On Tue, Oct 2, 2018 at 7:30 AM Reindl Harald h.reindl@thelounge.net wrote:
Am 02.10.18 um 13:25 schrieb Neal Gompa:
On Tue, Oct 2, 2018 at 7:01 AM Michal Schorm mschorm@redhat.com wrote:
I'm sending you this HTML email because Google dropped possibility to send plaintext emails. Sorry =(
Gmail -> compose -> options (bottom right) -> plain text mode
doesn't work for you?
(I'm sending this e-mail with this setting so who knows how can check it)
This is not respected by any of the mobile apps.
so get k9 mail and use imap/smtp - problem solved and a devloper should be capable to use mail proper
To be quite honest, I am not installing another mail app just for this. There's not a single legitimate reason for me to bother with the inconvenience of setting that up and effectively bypassing things like 2FA and whatnot.
On Tue, Oct 02, 2018 07:33:44 -0400, Neal Gompa wrote:
To be quite honest, I am not installing another mail app just for this. There's not a single legitimate reason for me to bother with the inconvenience of setting that up and effectively bypassing things like 2FA and whatnot.
A known, but unsolved issue from the looks of it (from a 2014 post): https://productforums.google.com/forum/#!topic/gmail/d7ySu-m1rgY
On Tue, Oct 2, 2018 at 4:46 AM Ankur Sinha sanjay.ankur@gmail.com wrote:
A known, but unsolved issue from the looks of it (from a 2014 post): https://productforums.google.com/forum/#!topic/gmail/d7ySu-m1rgY
Yep... and the issue is 4 years old and ends with the recommendation of "Send Feedback". I have no idea why Google doesn't add text mode to it's gmail phone apps... seems like it would be a simple thing for them to do - but then again, look at their other apps - especially their messaging strategy (or apparent lack thereof). It's the wild west over there.
As @Chris mentioned, if people collectively believe it's a big enough issue, start bouncing HTML emails.
El mar., 2 oct. 2018 a las 13:44, Neal Gompa (ngompa13@gmail.com) escribió:
On Tue, Oct 2, 2018 at 7:30 AM Reindl Harald h.reindl@thelounge.net wrote:
Am 02.10.18 um 13:25 schrieb Neal Gompa:
On Tue, Oct 2, 2018 at 7:01 AM Michal Schorm mschorm@redhat.com wrote:
I'm sending you this HTML email because Google dropped possibility to send plaintext emails. Sorry =(
Gmail -> compose -> options (bottom right) -> plain text mode
doesn't work for you?
(I'm sending this e-mail with this setting so who knows how can check it)
This is not respected by any of the mobile apps.
so get k9 mail and use imap/smtp - problem solved and a devloper should be capable to use mail proper
To be quite honest, I am not installing another mail app just for this. There's not a single legitimate reason for me to bother with the inconvenience of setting that up and effectively bypassing things like 2FA and whatnot.
IMO, it would be easier for everybody if the mailing list could automatically strip out the HTML part. Some projects do this (e.g., R).
-- 真実はいつも一つ!/ Always, there's only one truth!
El mar., 2 oct. 2018 a las 13:48, Iñaki Ucar (iucar@fedoraproject.org) escribió:
El mar., 2 oct. 2018 a las 13:44, Neal Gompa (ngompa13@gmail.com) escribió:
On Tue, Oct 2, 2018 at 7:30 AM Reindl Harald h.reindl@thelounge.net wrote:
Am 02.10.18 um 13:25 schrieb Neal Gompa:
On Tue, Oct 2, 2018 at 7:01 AM Michal Schorm mschorm@redhat.com wrote:
I'm sending you this HTML email because Google dropped possibility to send plaintext emails. Sorry =(
Gmail -> compose -> options (bottom right) -> plain text mode
doesn't work for you?
(I'm sending this e-mail with this setting so who knows how can check it)
This is not respected by any of the mobile apps.
so get k9 mail and use imap/smtp - problem solved and a devloper should be capable to use mail proper
To be quite honest, I am not installing another mail app just for this. There's not a single legitimate reason for me to bother with the inconvenience of setting that up and effectively bypassing things like 2FA and whatnot.
IMO, it would be easier for everybody if the mailing list could automatically strip out the HTML part. Some projects do this (e.g., R).
i.e., see multipart filtering and conversion to plaintext:
https://mailman.readthedocs.io/en/latest/src/mailman/handlers/docs/filtering...
It should be easy to set this up for all Fedora mailing lists.
-- 真実はいつも一つ!/ Always, there's only one truth!
-- Iñaki Ucar
On Tue, Oct 2, 2018 at 7:56 AM Iñaki Ucar iucar@fedoraproject.org wrote:
El mar., 2 oct. 2018 a las 13:48, Iñaki Ucar (iucar@fedoraproject.org) escribió:
El mar., 2 oct. 2018 a las 13:44, Neal Gompa (ngompa13@gmail.com) escribió:
On Tue, Oct 2, 2018 at 7:30 AM Reindl Harald h.reindl@thelounge.net wrote:
Am 02.10.18 um 13:25 schrieb Neal Gompa:
On Tue, Oct 2, 2018 at 7:01 AM Michal Schorm mschorm@redhat.com wrote:
> I'm sending you this HTML email because Google dropped possibility to send plaintext emails. Sorry =(
Gmail -> compose -> options (bottom right) -> plain text mode
doesn't work for you?
(I'm sending this e-mail with this setting so who knows how can check it)
This is not respected by any of the mobile apps.
so get k9 mail and use imap/smtp - problem solved and a devloper should be capable to use mail proper
To be quite honest, I am not installing another mail app just for this. There's not a single legitimate reason for me to bother with the inconvenience of setting that up and effectively bypassing things like 2FA and whatnot.
IMO, it would be easier for everybody if the mailing list could automatically strip out the HTML part. Some projects do this (e.g., R).
i.e., see multipart filtering and conversion to plaintext:
https://mailman.readthedocs.io/en/latest/src/mailman/handlers/docs/filtering...
It should be easy to set this up for all Fedora mailing lists.
I don't think it's a good idea to convert/transform people's emails in flight... But multipart mail renders as plain text by most plain text mail clients, so I'm not sure it's really a problem from Gmail (which always sends multipart mail when in rich text mode).
On Tuesday, 02 October 2018 at 13:55, Iñaki Ucar wrote:
El mar., 2 oct. 2018 a las 13:48, Iñaki Ucar (iucar@fedoraproject.org) escribió:
[...]
IMO, it would be easier for everybody if the mailing list could automatically strip out the HTML part. Some projects do this (e.g., R).
i.e., see multipart filtering and conversion to plaintext:
https://mailman.readthedocs.io/en/latest/src/mailman/handlers/docs/filtering...
It should be easy to set this up for all Fedora mailing lists.
While it might be easy, it would break GPG signatures on such e-mails. Arguably it could be skipped for such signed e-mails as a compromise.
Regards, Dominik
* Iñaki Ucar:
IMO, it would be easier for everybody if the mailing list could automatically strip out the HTML part. Some projects do this (e.g., R).
In many cases, the quoting in the text/html part is easier to read (it's clearer who wrote what) than in the text/plain part.
For example, Igor's message looks like this in HTML:
| Dear developers, | | Hello, Dominik ;) | | sorry for a slightly off-topic post, but I've noticed that a significant | number of posters is sending HTML e-mail to this list (not to mention
The first and last two lines of the snippet are quoted.
The corresponding text/plain version looks like this:
| > Dear developers, | > | | Hello, Dominik ;) | | sorry for a slightly off-topic post, but I've noticed that a significant | > number of posters is sending HTML e-mail to this list (not to mention
So the quoting has become corrupted. In this case, it is not much of a problem, but for inline replies it can be really confusing.
Thanks, Florian
On Wednesday, 03 October 2018 at 08:38, Florian Weimer wrote:
- Iñaki Ucar:
IMO, it would be easier for everybody if the mailing list could automatically strip out the HTML part. Some projects do this (e.g., R).
In many cases, the quoting in the text/html part is easier to read (it's clearer who wrote what) than in the text/plain part.
For me it's actually the other way around, because HTML part gets processed by lynx. Quoting is marked by identation, which is not automatically highlighted by VIM, while the > prefix in the text/plain part is.
But that's beside the point. My point was about excessive and unnecessary bandwith usage (the same information gets sent twice).
Regards, Dominik
On Thu, Oct 4, 2018 at 11:44 AM Dominik 'Rathann' Mierzejewski dominik@greysector.net wrote:
On Wednesday, 03 October 2018 at 08:38, Florian Weimer wrote:
- Iñaki Ucar:
IMO, it would be easier for everybody if the mailing list could automatically strip out the HTML part. Some projects do this (e.g., R).
In many cases, the quoting in the text/html part is easier to read (it's clearer who wrote what) than in the text/plain part.
For me it's actually the other way around, because HTML part gets processed by lynx. Quoting is marked by identation, which is not automatically highlighted by VIM, while the > prefix in the text/plain part is.
But that's beside the point. My point was about excessive and unnecessary bandwith usage (the same information gets sent twice).
I respect this issue, but I wonder if it this is a situation where we are trying to solve a problem for a small number of people by potential inconveniencing or excluding a lot of people.
While Internet plans vary, I wonder how many pay per bit? In September, if you assume every message sent to devel list was plain text and you further assume that the html version was 1.25 times longer then the total traffic for the month would have been about 10.6 MB. While this is 2.25 more bits than were actually sent (based on disk size), is it really a problem we should invest effort in fixing?
We know that more and more people are using mailers which can accidentally or purposefully send HTML mail. We know that we want to attract people, and not create unneeded barriers to entry. Saying that you have to change your email client to participate in Fedora is a huge disincentive. We are effectively putting a sign on the door saying "go away."
I suggest that if we think this is a serious issue and that we want to get the maximum possible help to those who are forced to pay by the bit that we stop sending them mail they don't need all together. Lets move this mailing list (and others) to discussion.fedoraproject.org. There the person will be given the ability to select exactly which subjects they see and allow to be transmitted down the wire.
regards,
bex
Regards, Dominik -- Fedora https://getfedora.org | RPMFusion http://rpmfusion.org There should be a science of discontent. People need hard times and oppression to develop psychic muscles. -- from "Collected Sayings of Muad'Dib" by the Princess Irulan _______________________________________________ devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-leave@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org
all this aside,HTML mail format is difficult to qoute and reply on mobile,plain text would be nice.
On Thu, Oct 4, 2018 at 3:34 PM, Brian (bex) Exelbierd bexelbie@redhat.com wrote:
On Thu, Oct 4, 2018 at 11:44 AM Dominik 'Rathann' Mierzejewski dominik@greysector.net wrote:
On Wednesday, 03 October 2018 at 08:38, Florian Weimer wrote:
- Iñaki Ucar:
IMO, it would be easier for everybody if the mailing list could automatically strip out the HTML part. Some projects do this (e.g., R).
In many cases, the quoting in the text/html part is easier to read
(it's
clearer who wrote what) than in the text/plain part.
For me it's actually the other way around, because HTML part gets processed by lynx. Quoting is marked by identation, which is not automatically highlighted by VIM, while the > prefix in the text/plain part is.
But that's beside the point. My point was about excessive and unnecessary bandwith usage (the same information gets sent twice).
I respect this issue, but I wonder if it this is a situation where we are trying to solve a problem for a small number of people by potential inconveniencing or excluding a lot of people.
While Internet plans vary, I wonder how many pay per bit? In September, if you assume every message sent to devel list was plain text and you further assume that the html version was 1.25 times longer then the total traffic for the month would have been about 10.6 MB. While this is 2.25 more bits than were actually sent (based on disk size), is it really a problem we should invest effort in fixing?
We know that more and more people are using mailers which can accidentally or purposefully send HTML mail. We know that we want to attract people, and not create unneeded barriers to entry. Saying that you have to change your email client to participate in Fedora is a huge disincentive. We are effectively putting a sign on the door saying "go away."
I suggest that if we think this is a serious issue and that we want to get the maximum possible help to those who are forced to pay by the bit that we stop sending them mail they don't need all together. Lets move this mailing list (and others) to discussion.fedoraproject.org. There the person will be given the ability to select exactly which subjects they see and allow to be transmitted down the wire.
regards,
bex
Regards, Dominik -- Fedora https://getfedora.org | RPMFusion http://rpmfusion.org There should be a science of discontent. People need hard times and oppression to develop psychic muscles. -- from "Collected Sayings of Muad'Dib" by the Princess Irulan _______________________________________________ devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-leave@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.
org/archives/list/devel@lists.fedoraproject.org
-- Brian (bex) Exelbierd | bexelbie@redhat.com | bex@pobox.com Fedora Community Action & Impact Coordinator @bexelbie | http://www.winglemeyer.org _______________________________________________ devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-leave@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@lists. fedoraproject.org
The fact is, the world has moved away from quoted mail with inline replies. Top posting rules basically everywhere except hold-out old-school mailing lists. Gmail, both on the web and _especially_ on mobile, makes it almost impossible to do quoting and inline replies properly. I can't imagine other big consumer email providers are much better.
Yes, this sucks for having nuanced point-by-point discussions. But I don't think that's a tide we're really going to hold back. We should look at other tools that fit with the way people do asynchronous online discussion today. As Brian noted, we're trying out Discourse at https://discussion.fedoraproject.org/ — if you're interested in trying it out, come join the conversation there.
On Thu, Oct 04, 2018 at 03:51:51PM +0530, Manas Mangaonkar wrote:
all this aside,HTML mail format is difficult to qoute and reply on mobile,plain text would be nice.
On Thu, Oct 4, 2018 at 3:34 PM, Brian (bex) Exelbierd bexelbie@redhat.com wrote:
On Thu, Oct 4, 2018 at 11:44 AM Dominik 'Rathann' Mierzejewski dominik@greysector.net wrote:
On Wednesday, 03 October 2018 at 08:38, Florian Weimer wrote:
- Iñaki Ucar:
IMO, it would be easier for everybody if the mailing list could automatically strip out the HTML part. Some projects do this (e.g., R).
In many cases, the quoting in the text/html part is easier to read
(it's
clearer who wrote what) than in the text/plain part.
For me it's actually the other way around, because HTML part gets processed by lynx. Quoting is marked by identation, which is not automatically highlighted by VIM, while the > prefix in the text/plain part is.
But that's beside the point. My point was about excessive and unnecessary bandwith usage (the same information gets sent twice).
I respect this issue, but I wonder if it this is a situation where we are trying to solve a problem for a small number of people by potential inconveniencing or excluding a lot of people.
While Internet plans vary, I wonder how many pay per bit? In September, if you assume every message sent to devel list was plain text and you further assume that the html version was 1.25 times longer then the total traffic for the month would have been about 10.6 MB. While this is 2.25 more bits than were actually sent (based on disk size), is it really a problem we should invest effort in fixing?
We know that more and more people are using mailers which can accidentally or purposefully send HTML mail. We know that we want to attract people, and not create unneeded barriers to entry. Saying that you have to change your email client to participate in Fedora is a huge disincentive. We are effectively putting a sign on the door saying "go away."
I suggest that if we think this is a serious issue and that we want to get the maximum possible help to those who are forced to pay by the bit that we stop sending them mail they don't need all together. Lets move this mailing list (and others) to discussion.fedoraproject.org. There the person will be given the ability to select exactly which subjects they see and allow to be transmitted down the wire.
regards,
bex
Regards, Dominik -- Fedora https://getfedora.org | RPMFusion http://rpmfusion.org There should be a science of discontent. People need hard times and oppression to develop psychic muscles. -- from "Collected Sayings of Muad'Dib" by the Princess Irulan _______________________________________________ devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-leave@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.
org/archives/list/devel@lists.fedoraproject.org
-- Brian (bex) Exelbierd | bexelbie@redhat.com | bex@pobox.com Fedora Community Action & Impact Coordinator @bexelbie | http://www.winglemeyer.org _______________________________________________ devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-leave@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@lists. fedoraproject.org
devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-leave@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org
On Thu, Oct 4, 2018, 23:33 Matthew Miller mattdm@fedoraproject.org wrote:
The fact is, the world has moved away from quoted mail with inline replies. Top posting rules basically everywhere except hold-out old-school mailing lists. Gmail, both on the web and _especially_ on mobile, makes it almost impossible to do quoting and inline replies properly. I can't imagine other big consumer email providers are much better.
Yes, this sucks for having nuanced point-by-point discussions. But I don't think that's a tide we're really going to hold back. We should look at other tools that fit with the way people do asynchronous online discussion today. As Brian noted, we're trying out Discourse at https://discussion.fedoraproject.org/ — if you're interested in trying it out, come join the conversation there.
How can we get our space there? Packaging and Rust topics.
On Thu, Oct 04, 2018 at 03:51:51PM +0530, Manas Mangaonkar wrote:
all this aside,HTML mail format is difficult to qoute and reply on mobile,plain text would be nice.
On Thu, Oct 4, 2018 at 3:34 PM, Brian (bex) Exelbierd <
bexelbie@redhat.com>
wrote:
On Thu, Oct 4, 2018 at 11:44 AM Dominik 'Rathann' Mierzejewski dominik@greysector.net wrote:
On Wednesday, 03 October 2018 at 08:38, Florian Weimer wrote:
- Iñaki Ucar:
IMO, it would be easier for everybody if the mailing list could automatically strip out the HTML part. Some projects do this
(e.g.,
R).
In many cases, the quoting in the text/html part is easier to read
(it's
clearer who wrote what) than in the text/plain part.
For me it's actually the other way around, because HTML part gets processed by lynx. Quoting is marked by identation, which is not automatically highlighted by VIM, while the > prefix in the
text/plain
part is.
But that's beside the point. My point was about excessive and unnecessary bandwith usage (the same information gets sent twice).
I respect this issue, but I wonder if it this is a situation where we are trying to solve a problem for a small number of people by potential inconveniencing or excluding a lot of people.
While Internet plans vary, I wonder how many pay per bit? In September, if you assume every message sent to devel list was plain text and you further assume that the html version was 1.25 times longer then the total traffic for the month would have been about 10.6 MB. While this is 2.25 more bits than were actually sent (based on disk size), is it really a problem we should invest effort in fixing?
We know that more and more people are using mailers which can accidentally or purposefully send HTML mail. We know that we want to attract people, and not create unneeded barriers to entry. Saying that you have to change your email client to participate in Fedora is a huge disincentive. We are effectively putting a sign on the door saying "go away."
I suggest that if we think this is a serious issue and that we want to get the maximum possible help to those who are forced to pay by the bit that we stop sending them mail they don't need all together. Lets move this mailing list (and others) to discussion.fedoraproject.org. There the person will be given the ability to select exactly which subjects they see and allow to be transmitted down the wire.
regards,
bex
Regards, Dominik -- Fedora https://getfedora.org | RPMFusion http://rpmfusion.org There should be a science of discontent. People need hard times and oppression to develop psychic muscles. -- from "Collected Sayings of Muad'Dib" by the Princess
Irulan
devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-leave@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines:
https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: https://lists.fedoraproject.
org/archives/list/devel@lists.fedoraproject.org
-- Brian (bex) Exelbierd | bexelbie@redhat.com | bex@pobox.com Fedora Community Action & Impact Coordinator @bexelbie | http://www.winglemeyer.org _______________________________________________ devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-leave@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines:
https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives:
https://lists.fedoraproject.org/archives/list/devel@lists.
fedoraproject.org
devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-leave@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives:
https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org
-- Matthew Miller mattdm@fedoraproject.org Fedora Project Leader _______________________________________________ devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-leave@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org
On Thu, 2018-10-04 at 17:32 -0400, Matthew Miller wrote:
The fact is, the world has moved away from quoted mail with inline replies. Top posting rules basically everywhere except hold-out old-school mailing lists.
...like this one. ;)
On Thu, 04 Oct 2018 16:26:21 -0700 Adam Williamson adamwill@fedoraproject.org wrote:
On Thu, 2018-10-04 at 17:32 -0400, Matthew Miller wrote:
The fact is, the world has moved away from quoted mail with inline replies. Top posting rules basically everywhere except hold-out old-school mailing lists.
...like this one. ;)
His email was from mutt as well! I guess fedora-devel just isn't old-school enough :(
On Fri, Oct 05, 2018 at 02:02:46AM +0100, Andrew Clayton wrote:
Adam Williamson adamwill@fedoraproject.org wrote:
On Thu, 2018-10-04 at 17:32 -0400, Matthew Miller wrote:
The fact is, the world has moved away from quoted mail with inline replies. Top posting rules basically everywhere except hold-out old-school mailing lists.
...like this one. ;)
His email was from mutt as well! I guess fedora-devel just isn't old-school enough :(
Okay, I admit to some level of trolling there. :)
My point, though, is that I don't think we can afford to be exclusively old-school to the point where it makes a significant barrier to entry. A few years ago, I think the landscape was different. As more young people come to the project with a new perspective on online communication, we should look at how _our_ practices should evolve.
Top-posting isn't just an annoying Gmail thing; it's a whole different approach to using email for conversations. And, yeah, it *is* a much worse one than our old-school threaded, minimal-quoting, usenet-inspired style. That's why the general trend is *away* from email.
The Foreman community recently switched away from mailing lists in this way, and https://theforeman.org/2018/07/discourse-6-months-on-impact-assesment.html is really interesting and helpful read on the topic for those who might have some ... trepidation.
I'm not sayin' we are ready to shut this list down, but it's honestly worth considering if a different approach will be more effective.
On Fri, Oct 5, 2018 at 10:20 AM, Matthew Miller mattdm@fedoraproject.org wrote:
On Fri, Oct 05, 2018 at 02:02:46AM +0100, Andrew Clayton wrote:
Adam Williamson adamwill@fedoraproject.org wrote:
On Thu, 2018-10-04 at 17:32 -0400, Matthew Miller wrote:
The fact is, the world has moved away from quoted mail with inline replies. Top posting rules basically everywhere except hold-out old-school mailing lists.
...like this one. ;)
His email was from mutt as well! I guess fedora-devel just isn't old-school enough :(
Okay, I admit to some level of trolling there. :)
My point, though, is that I don't think we can afford to be exclusively old-school to the point where it makes a significant barrier to entry.
I agree.
And I think this needs rephrasing since forever:
https://fedoraproject.org/wiki/Mailing_list_guidelines Place each part of your reply after the text it addresses (i.e., NO Top-Posting, please see "Wikipedia - Top Posting" and links therein for more on this).
It's one thing to state a preference for bottom posting, and why. But the present guidelines encourage people to flip out whenever there's a top post.
Top-posting isn't just an annoying Gmail thing; it's a whole different approach to using email for conversations. And, yeah, it *is* a much worse one than our old-school threaded, minimal-quoting, usenet-inspired style. That's why the general trend is *away* from email.
For all the faults of gmail, one thing I like is being able to search lists within gmail.
The Foreman community recently switched away from mailing lists in this way, and https://theforeman.org/2018/07/discourse-6-months-on-impact-assesment.html is really interesting and helpful read on the topic for those who might have some ... trepidation.
I'm not sayin' we are ready to shut this list down, but it's honestly worth considering if a different approach will be more effective.
Far worse would be fragmented list/forums resulting from mutually exclusive venues. It's probably a joke and not worth the effort to have automagic behind the scenes cross posting between discourse forums and hyperkitty (email) lists.
* Matthew Miller:
The Foreman community recently switched away from mailing lists in this way, and https://theforeman.org/2018/07/discourse-6-months-on-impact-assesment.html is really interesting and helpful read on the topic for those who might have some ... trepidation.
I'm more worried how these forums exclude commercial contributors. Rust has the same NC terms. How is this supposed to work for projects which are not exclusively run and used by hobbyists? How do you incorporate material that is posted to the forum into free software, when the terms clearly state that you cannot do this?
Thanks, Florian
On Fri, Oct 05, 2018 at 08:25:21PM +0200, Florian Weimer wrote:
I'm more worried how these forums exclude commercial contributors. Rust has the same NC terms. How is this supposed to work for projects which are not exclusively run and used by hobbyists? How do you incorporate material that is posted to the forum into free software, when the terms clearly state that you cannot do this?
It looks like Foreman hasn't updated the software's boilerplate terms of service, which puts user content under CC BY-NA-SA.
For the Fedora instance, we've changed that whole long page to a much simpler statement that
1. Participants are bound by the Fedora Code of Conduct, and
2. User contributions need to be under an acceptable-for-Fedora license, with the normal Fedora content license (which is CC BY-SA 3.0) by default.
(See https://discussion.fedoraproject.org/tos)
On Fri, Oct 5, 2018 at 9:21 AM Matthew Miller mattdm@fedoraproject.org wrote: ...
That's why the general trend is *away* from email.
The Foreman community recently switched away from mailing lists in this
way,
and
https://theforeman.org/2018/07/discourse-6-months-on-impact-assesment.html
is really interesting and helpful read on the topic for those who might
have
some ... trepidation.
I'm not sayin' we are ready to shut this list down, but it's honestly
worth
considering if a different approach will be more effective.
Before the daggers come out.... please take some time to checkout Fedora Discourse https://discussion.fedoraproject.org/ and read the Foreman community link above.
As I had previously mentioned, Discourse allows for RSS feeds and email notifications. It also has a mailing list mode.
It really is a much better solution.
On Tue, Oct 16, 2018, 8:13 AM Gerald B. Cox gbcox@bzb.us wrote:
On Fri, Oct 5, 2018 at 9:21 AM Matthew Miller mattdm@fedoraproject.org wrote: ...
That's why the general trend is *away* from email.
The Foreman community recently switched away from mailing lists in this
way,
and
https://theforeman.org/2018/07/discourse-6-months-on-impact-assesment.html
is really interesting and helpful read on the topic for those who might
have
some ... trepidation.
I'm not sayin' we are ready to shut this list down, but it's honestly
worth
considering if a different approach will be more effective.
Before the daggers come out.... please take some time to checkout Fedora Discourse https://discussion.fedoraproject.org/ and read the Foreman community link above.
As I had previously mentioned, Discourse allows for RSS feeds and email notifications. It also has a mailing list mode.
RSS is going away. And part of that, Mozilla is removing built-in RSS support in Firefox 64.
Chris Murphy
On Tue, Oct 16, 2018 at 7:25 AM Chris Murphy lists@colorremedies.com wrote:
On Tue, Oct 16, 2018, 8:13 AM Gerald B. Cox gbcox@bzb.us wrote:
On Fri, Oct 5, 2018 at 9:21 AM Matthew Miller mattdm@fedoraproject.org wrote: ...
That's why the general trend is *away* from email.
The Foreman community recently switched away from mailing lists in this
way,
and
https://theforeman.org/2018/07/discourse-6-months-on-impact-assesment.html
is really interesting and helpful read on the topic for those who might
have
some ... trepidation.
I'm not sayin' we are ready to shut this list down, but it's honestly
worth
considering if a different approach will be more effective.
Before the daggers come out.... please take some time to checkout Fedora Discourse https://discussion.fedoraproject.org/ and read the Foreman community link above.
As I had previously mentioned, Discourse allows for RSS feeds and email notifications. It also has a mailing list mode.
RSS is going away. And part of that, Mozilla is removing built-in RSS support in Firefox 64.
Well, for the moment, Discourse supports it (as well as Fedora updates) - and FWIW I've never used RSS Support in Fx. I use Tiny Tiny RSS and Inoreader and have yet to have a problem adding a feed from a site. There are also many many other clients available - too many to list here.
If it goes away however, such is life - and I'll use the next best thing.
But RSS really isn't the point of this thread. My point is that Fedora should move to Discourse. It's a much better solution for discussion. We should publish a timeline and just do it.
Il giorno mar 16 ott 2018 alle 16:39, "Gerald B. Cox" gbcox@bzb.us ha scritto:
On Tue, Oct 16, 2018 at 7:25 AM Chris Murphy lists@colorremedies.com wrote:
On Tue, Oct 16, 2018, 8:13 AM Gerald B. Cox gbcox@bzb.us wrote:
On Fri, Oct 5, 2018 at 9:21 AM Matthew Miller mattdm@fedoraproject.org wrote: ...
That's why the general trend is *away* from email.
The Foreman community recently switched away from mailing lists
in this way,
and
https://theforeman.org/2018/07/discourse-6-months-on-impact-assesment.html
is really interesting and helpful read on the topic for those who
might have
some ... trepidation.
I'm not sayin' we are ready to shut this list down, but it's
honestly worth
considering if a different approach will be more effective.
Before the daggers come out.... please take some time to checkout Fedora Discourse https://discussion.fedoraproject.org/ and read the Foreman community link above.
As I had previously mentioned, Discourse allows for RSS feeds and email notifications. It also has a mailing list mode.
RSS is going away. And part of that, Mozilla is removing built-in RSS support in Firefox 64.
Well, for the moment, Discourse supports it (as well as Fedora updates) - and FWIW I've never used RSS Support in Fx. I use Tiny Tiny RSS and Inoreader and have yet to have a problem adding a feed from a site. There are also many many other clients available - too many to list here.
If it goes away however, such is life - and I'll use the next best thing.
And anyway.. Firefox is dropping built-in support for RSS, but you'll be able to use RSS in Firefox through web extensions (or just use another browser or an external application).
https://www.zdnet.com/article/end-nears-for-rss-firefox-64-to-drop-built-in-...
Let me add a personal consideration on this regard. I'm deleting 90% of emails coming in this mailing list, but I have to be subscribed and filter tens of threads every day just to get the 10% I'm interested in. If you sum it day by day, it's a lot of waste.
In Discourse I'd be able to receive emails posted in the section I'm interested in. Or I may want to receive no email at all, just a notification on the website. My email client wouldn't be loaded of stuff I don't need. My mind would be happier too :-)
On Tue, Oct 16, 2018, 8:39 AM Gerald B. Cox gbcox@bzb.us wrote:
On Tue, Oct 16, 2018 at 7:25 AM Chris Murphy lists@colorremedies.com wrote:
On Tue, Oct 16, 2018, 8:13 AM Gerald B. Cox gbcox@bzb.us wrote:
On Fri, Oct 5, 2018 at 9:21 AM Matthew Miller mattdm@fedoraproject.org wrote: ...
That's why the general trend is *away* from email.
The Foreman community recently switched away from mailing lists in
this way,
and
https://theforeman.org/2018/07/discourse-6-months-on-impact-assesment.html
is really interesting and helpful read on the topic for those who
might have
some ... trepidation.
I'm not sayin' we are ready to shut this list down, but it's honestly
worth
considering if a different approach will be more effective.
Before the daggers come out.... please take some time to checkout Fedora Discourse https://discussion.fedoraproject.org/ and read the Foreman community link above.
As I had previously mentioned, Discourse allows for RSS feeds and email notifications. It also has a mailing list mode.
RSS is going away. And part of that, Mozilla is removing built-in RSS support in Firefox 64.
Well, for the moment, Discourse supports it (as well as Fedora updates) - and FWIW I've never used RSS Support in Fx. I use Tiny Tiny RSS and Inoreader and have yet to have a problem adding a feed from a site. There are also many many other clients available - too many to list here.
If it goes away however, such is life - and I'll use the next best thing.
But RSS really isn't the point of this thread. My point is that Fedora should move to Discourse. It's a much better solution for discussion. We should publish a timeline and just do it.
What does migration look like? Existing mail archive is imported into Discourse, or is it a hard break? If hard break, what's the possibility of a pseudo import, where Discourse searches include results from the mail archive? Can the existing subscriptions be transferred to Discourse? And could those transfers default to making list mode?
Infrastructure team would do this work? What's their position, available resources, and timeframe?
Chris Murphy
On Tue, Oct 16, 2018 at 10:39 AM Gerald B. Cox gbcox@bzb.us wrote:
But RSS really isn't the point of this thread. My point is that Fedora should move to Discourse. It's a much better solution for discussion. We should publish a timeline and just do it.
Fedora should not "just do it" because "it's a much better solution". If you want Fedora to consider moving to Discourse over email, it would be nice to see an analysis like:
1) What are the pros and cons? Is Discourse's mailing list mode an acceptable interface for people who prefer maillng lists and wouldn't want to interact with Discourse itself?
2) What's the migration strategy look like? Can users on the mailing lists be automatically added to the relevant discourse lists?
3) Will there be a transition period where the old list addresses continue to work? Can they be maintained in perpetuity, or at least aliased to the right thing?
I'm not saying *you* need to answer these questions, but I think *someone* should before proposing to move all of Fedora's mailing list infrastructure to Discourse. We would request answers to similar questions for a system-wide change proposal, and I think generally similar policies should be applied when considering major changes to the distribution's infrastructure. And the mailing lists are definitely an important part of the distribution's infrastructure.
Ben Rosser
On Tue, Oct 16, 2018 at 8:06 AM Ben Rosser rosser.bjr@gmail.com wrote:
On Tue, Oct 16, 2018 at 10:39 AM Gerald B. Cox gbcox@bzb.us wrote:
But RSS really isn't the point of this thread. My point is that Fedora
should move to Discourse. It's a much better solution for
discussion. We should publish a timeline and just do it.
Fedora should not "just do it" because "it's a much better solution". If you want Fedora to consider moving to Discourse over email, it would be nice to see an analysis like:
- What are the pros and cons? Is Discourse's mailing list mode an
acceptable interface for people who prefer maillng lists and wouldn't want to interact with Discourse itself?
- What's the migration strategy look like? Can users on the mailing
lists be automatically added to the relevant discourse lists?
- Will there be a transition period where the old list addresses
continue to work? Can they be maintained in perpetuity, or at least aliased to the right thing?
I'm not saying *you* need to answer these questions, but I think *someone* should before proposing to move all of Fedora's mailing list infrastructure to Discourse. We would request answers to similar questions for a system-wide change proposal, and I think generally similar policies should be applied when considering major changes to the distribution's infrastructure. And the mailing lists are definitely an important part of the distribution's infrastructure.
Did you checkout the Fedora Discourse site and read the Foreman analysis that Matt provided?
We're not talking about re-inventing the wheel here... nor are we the first organization that have done this.
I'm sure there are lots of lessons learned available to review - and I would expect that the Fedora team that is working on Discourse now is totally aware of this.
On Tue, Oct 16, 2018 at 11:12 AM Gerald B. Cox gbcox@bzb.us wrote:
Did you checkout the Fedora Discourse site and read the Foreman analysis that Matt provided?
Unless I'm missing something, I am not sure that either of these things answers my questions. I am not asking the question "will switching to Discourse increase user engagement" (which seems to be what the linked Foreman article covers, unless I missed something?)-- I'm asking "what does the transition actually look like for current users of the mailing lists".
We're not talking about re-inventing the wheel here... nor are we the first organization that have done this.
I'm sure there are lots of lessons learned available to review - and I would expect that the Fedora team that is working on Discourse now is totally aware of this.
Great, then it should be easy for someone to write up a change proposal-like document detailing the migration strategy.
I am not saying switching to Discourse is a *bad* idea. I am saying that I, at least, would like to see a more serious proposal than simply "just do it because it's better". That might require switching one list over and seeing how well it works (and it sounded like the Council was considering doing this with council-discuss, which is probably a good first step).
Ben Rosser
On Tue, Oct 16, 2018 at 11:30 AM Ben Rosser rosser.bjr@gmail.com wrote:
I am not saying switching to Discourse is a *bad* idea. I am saying that I, at least, would like to see a more serious proposal than simply "just do it because it's better". That might require switching one list over and seeing how well it works (and it sounded like the Council was considering doing this with council-discuss, which is probably a good first step).
Council is considering it, and CommOps just started a 2-month experiment to see how it goes. It makes sense to me that we wait and see how those go before we start considering larger moves. I'm worried in particular about larger lists such as this one where there's a large number of users and a high volume of traffic. I haven't used Discourse enough to know what the experience is like at that point.
I agree that moving to Discourse (or any other change in how we communicate) needs to be backed by a well-thought-out proposal. Whichever technology we choose, it will be better for some and worse for others. That's unavoidable, but we should be intentional about the changes we make and what impact they will have on the community.
On Tue, Oct 16, 2018 at 8:41 AM Ben Cotton bcotton@redhat.com wrote:
Council is considering it, and CommOps just started a 2-month experiment to see how it goes. It makes sense to me that we wait and see how those go before we start considering larger moves. I'm worried in particular about larger lists such as this one where there's a large number of users and a high volume of traffic. I haven't used Discourse enough to know what the experience is like at that point.
I agree that moving to Discourse (or any other change in how we communicate) needs to be backed by a well-thought-out proposal. Whichever technology we choose, it will be better for some and worse for others. That's unavoidable, but we should be intentional about the changes we make and what impact they will have on the community.
Hey Ben...
Yeah, based upon the trial and previous comments I knew it was being considered. The point of my thread was that many people probably haven't taken the time to checkout the Fedora Discourse website or read the Foreman analysis that Matt provided. They should.
The difference between using Discourse and mailing lists is night and day.
As far as larger lists... check out this link: https://meta.discourse.org/t/biggest-busiest-discourse-forums/30674
On Tue, 16 Oct 2018 08:51:33 -0700 "Gerald B. Cox" gbcox@bzb.us wrote:
Yeah, based upon the trial and previous comments I knew it was being considered. The point of my thread was that many people probably haven't taken the time to checkout the Fedora Discourse website or read the Foreman analysis that Matt provided. They should.
Did this. I didn't like all the clicking; I prefer shortcut keys. From what I can see on the web, Discourse is trying to mimic social platforms so users familiar with them find it easy to use. Also seems targeted to mobiles. That *could* be beneficial for getting new blood into Fedora, as most younger people who show interest are going to be in the intersection of those two factors.
The difference between using Discourse and mailing lists is night and day.
But which is night and which is day? :-) Seriously, this is a value judgement, which is true for *you*, but not necessarily for others. Are you similar to the people who are being targeted by this change? Yes? That's evidence it might work.
As far as larger lists... check out this link: https://meta.discourse.org/t/biggest-busiest-discourse-forums/30674
I don't think any of the Fedora lists I'm subscribed to are high traffic, even this one, so that isn't really relevant. I suppose it does show there is room for growth.
The fact that the software is open source gives insurance that if the site goes under, Fedora doesn't have to lose all its interaction archives. Maybe the database could be backed up periodically, so in an emergency Fedora could run the software and bring them back up.
On Tue, Oct 16, 2018 at 11:59 AM stan stanl-fedorauser@vfemail.net wrote:
On Tue, 16 Oct 2018 08:51:33 -0700 "Gerald B. Cox" gbcox@bzb.us wrote:
Yeah, based upon the trial and previous comments I knew it was being considered. The point of my thread was that many people probably haven't taken the time to checkout the Fedora Discourse website or read the Foreman analysis that Matt provided. They should.
Did this. I didn't like all the clicking; I prefer shortcut keys.
Press the ? key and a window will pop up with all the keyboard shutcuts
From what I can see on the web, Discourse is trying to mimic social
platforms so users familiar with them find it easy to use. Also seems targeted to mobiles. That *could* be beneficial for getting new blood into Fedora, as most younger people who show interest are going to be in the intersection of those two factors.
The difference between using Discourse and mailing lists is night and day.
But which is night and which is day? :-) Seriously, this is a value judgement, which is true for *you*, but not necessarily for others. Are you similar to the people who are being targeted by this change? Yes? That's evidence it might work.
I believe the product provides a more natural discussion environment, while
allowing a mailing list mode for those who prefer that. Considering the many current deployments, that appears to be a popular opinion.
As far as larger lists... check out this link: https://meta.discourse.org/t/biggest-busiest-discourse-forums/30674
I don't think any of the Fedora lists I'm subscribed to are high traffic, even this one, so that isn't really relevant. I suppose it does show there is room for growth.
That was mainly in response to a concern about the platform scalibility.
On Tue, 16 Oct 2018 13:27:12 -0700 "Gerald B. Cox" gbcox@bzb.us wrote:
Press the ? key and a window will pop up with all the keyboard shutcuts
I couldn't find a way to move to the next topic from within the current topic. I had to do a u, then a down arrow, then an enter. That is, I finished reading a response stream, and just want to open the next one. I tried J and K, but they did nothing except appear in a page search entry box.
The other thing, is there a way to put the responses in threaded mode, so they are in context like a tree, the way a newsreader would do it. The display seems to be a posting time based stream of responses. Maybe there is a cue I'm missing that tells the post the response is to? Or maybe it is because I'm not registered and logged in. Is there extra functionality then?
On Tue, 16 Oct 2018 14:11:21 -0700 stan stanl-fedorauser@vfemail.net wrote:
I couldn't find a way to move to the next topic from within the current topic. I had to do a u, then a down arrow, then an enter.
This is wrong. I did u and then had to click on a topic. It didn't put the focus on the previous topic.
On Tue, 16 Oct 2018 14:11:21 -0700 stan stanl-fedorauser@vfemail.net wrote:
The other thing, is there a way to put the responses in threaded mode, so they are in context like a tree, the way a newsreader would do it. The display seems to be a posting time based stream of responses. Maybe there is a cue I'm missing that tells the post the response is to? Or maybe it is because I'm not registered and logged in. Is there extra functionality then?
This is already done, it puts the number of replies and the replies under each post. The first topic I read must have been without threads. When I read a topic with threads it was obvious.
Il giorno mar 16 ott 2018 alle 20:58, stan stanl-fedorauser@vfemail.net ha scritto:
On Tue, 16 Oct 2018 08:51:33 -0700 "Gerald B. Cox" gbcox@bzb.us wrote:
Yeah, based upon the trial and previous comments I knew it was being considered. The point of my thread was that many people probably haven't taken the time to checkout the Fedora Discourse website or read the Foreman analysis that Matt provided. They should.
Did this. I didn't like all the clicking; I prefer shortcut keys. From what I can see on the web, Discourse is trying to mimic social platforms so users familiar with them find it easy to use. Also seems targeted to mobiles. That *could* be beneficial for getting new blood into Fedora, as most younger people who show interest are going to be in the intersection of those two factors.
I don't see it this way. I'm attracted by Discourse as I'm attracted by stackoverflow or similar platforms. There are obvious reasons why these platforms work better than the old mailing lists and people using these tools is growing. I'm not young (getting close to the fourties) and I don't use social, but I'm definitely in the target.
I've seen this same discussion in another (smaller) open source project I contribute to for some years. Basically the contrast is between "old school people" who are attached to their old habits (sometimes for a very good reason! I love and appreciate old school people) and people who are curious about new tools they think could improve their and others' experience.
The difference between using Discourse and mailing lists is night and day.
But which is night and which is day? :-) Seriously, this is a value judgement, which is true for *you*, but not necessarily for others. Are you similar to the people who are being targeted by this change? Yes? That's evidence it might work.
Of course it's subjective. But we don't know how many people in this community would be in favour of this or that. The success of Stackoverflow makes me think though...
As far as larger lists... check out this link: https://meta.discourse.org/t/biggest-busiest-discourse-forums/30674
I don't think any of the Fedora lists I'm subscribed to are high traffic, even this one, so that isn't really relevant. I suppose it does show there is room for growth.
Perhaps each single list is not high traffic, but if you are subscribed to 3 or 4 Fedora lists (as I am) then you get a lot of traffic. My Fedora folder is the busiest folder in my email client. Sometimes I unsubscribe to some lists, then subscribe again... Discourse would be my "day" (see above) :D
On Tue, Oct 16, 2018 at 11:29:33AM -0400, Ben Rosser wrote:
I am not saying switching to Discourse is a *bad* idea. I am saying that I, at least, would like to see a more serious proposal than simply "just do it because it's better". That might require switching one list over and seeing how well it works (and it sounded like the Council was considering doing this with council-discuss, which is probably a good first step).
Yeah, I certainly am not suggesting we switch devel list, at least not for the forseeable future. Even if this whole thread kind of started with observations about actual problems we are seeing with the world today and mailing lists.
Ben Rosser rosser.bjr@gmail.com wrote:
- What's the migration strategy look like? Can users on the mailing
lists be automatically added to the relevant discourse lists?
Can the mailing lists be transferred from Mailman to Discourse and continue to be the *same* mailing list, with the same address and the same list ID? Because otherwise you'll break things for subscribers and force them to redo their configuration.
Just force-subscribing people to a new mailing list won't cause their local programs to handle messages the same way as those from an old mailing list. Programs will correctly see a new mailing list that they don't know anything about, and everybody will have to tell their programs how to handle messages from this new mailing list.
Transitions like this are what the List-ID header field is for. Even if the list address changes, the list ID is supposed to remain unchanged, so that sorting and filtering rules can recognize the list. But I'm sure many people select on the list address instead, so in practice you have to preserve both the list ID and the list address to make the transition smooth.
Björn Persson
On Tue, Oct 16, 2018 at 3:22 PM, Björn Persson <Bjorn@rombobjörn.se> wrote:
Ben Rosser rosser.bjr@gmail.com wrote:
- What's the migration strategy look like? Can users on the mailing
lists be automatically added to the relevant discourse lists?
Can the mailing lists be transferred from Mailman to Discourse and continue to be the *same* mailing list, with the same address and the same list ID? Because otherwise you'll break things for subscribers and force them to redo their configuration.
Just force-subscribing people to a new mailing list won't cause their local programs to handle messages the same way as those from an old mailing list. Programs will correctly see a new mailing list that they don't know anything about, and everybody will have to tell their programs how to handle messages from this new mailing list.
Transitions like this are what the List-ID header field is for. Even if the list address changes, the list ID is supposed to remain unchanged, so that sorting and filtering rules can recognize the list. But I'm sure many people select on the list address instead, so in practice you have to preserve both the list ID and the list address to make the transition smooth.
Along those same lines, we have a bunch of permalinks floating about with the current mailing list. The permalink is available in hyperkitty UI, as well as a header in the mails. For example the email I'm replying to has header:
Archived-At: https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/message/EMXAAJU2KVICBYUQ5DF5IELVAGGMZWOX/
Ideally these get preserved somehow.
* Björn Persson:
Ben Rosser rosser.bjr@gmail.com wrote:
- What's the migration strategy look like? Can users on the mailing
lists be automatically added to the relevant discourse lists?
Can the mailing lists be transferred from Mailman to Discourse and continue to be the *same* mailing list, with the same address and the same list ID? Because otherwise you'll break things for subscribers and force them to redo their configuration.
Discourse uses labels instead of folders, and a discussion can have multiple labels, so filing email into separate folders per topic (devel vs legal vs docs) is tricky because you get only a single message. It can be done, but it's a bit of client-side work.
On the Rust Discourse instance, the email threading doesn't feel quite right, either.
Email responses end up with hard line breaks in the web interface, which may not look as nice as web-submitted contributions.
I assume that if you don't care about folders and threads, it will work well enough.
Although it would make me sad to see the move. Mailing lists are hardly the main source of Fedora overhead for me. That's dist-git and RPM patch management. 8-(
Thanks, Florian
On Wed, Oct 17, 2018 at 10:20:49 +0200, Florian Weimer fweimer@redhat.com wrote:
Email responses end up with hard line breaks in the web interface, which may not look as nice as web-submitted contributions.
Did you test this for format=flowed? I would expect hard line breaks if that wasn't used.
* Bruno Wolff, III:
On Wed, Oct 17, 2018 at 10:20:49 +0200, Florian Weimer fweimer@redhat.com wrote:
Email responses end up with hard line breaks in the web interface, which may not look as nice as web-submitted contributions.
Did you test this for format=flowed? I would expect hard line breaks if that wasn't used.
I did not test with format=flowed. It may look better there, but I stopped using it because it looks really bad for mailing list usage. (Really long lines in most web-based mailing list archives and stuff like that.)
Thanks, Florian
On Tue, Oct 16, 2018 at 11:06 AM Ben Rosser rosser.bjr@gmail.com wrote:
- What's the migration strategy look like? Can users on the mailing
lists be automatically added to the relevant discourse lists?
- Will there be a transition period where the old list addresses
continue to work? Can they be maintained in perpetuity, or at least aliased to the right thing?
I agree these are absolutely the most important questions to answer. Silently breaking the experience for anyone not paying enough attention is not acceptable.
-Dan
Il giorno mar 16 ott 2018 alle 16:12, "Gerald B. Cox" gbcox@bzb.us ha scritto:
On Fri, Oct 5, 2018 at 9:21 AM Matthew Miller mattdm@fedoraproject.org wrote: ...
That's why the general trend is *away* from email.
The Foreman community recently switched away from mailing lists in
this way,
and
https://theforeman.org/2018/07/discourse-6-months-on-impact-assesment.html
is really interesting and helpful read on the topic for those who
might have
some ... trepidation.
I'm not sayin' we are ready to shut this list down, but it's
honestly worth
considering if a different approach will be more effective.
Before the daggers come out.... please take some time to checkout Fedora Discourse https://discussion.fedoraproject.org/ and read the Foreman community link above.
As I had previously mentioned, Discourse allows for RSS feeds and email notifications. It also has a mailing list mode.
It really is a much better solution.
I agree with you and I'm happy about the discussion.fedoraproject.org "experiment".
On Tue, 2018-10-16 at 07:12 -0700, Gerald B. Cox wrote:
On Fri, Oct 5, 2018 at 9:21 AM Matthew Miller mattdm@fedoraproject.org wrote: ...
That's why the general trend is *away* from email.
The Foreman community recently switched away from mailing lists in this
way,
and
https://theforeman.org/2018/07/discourse-6-months-on-impact-assesment.html
is really interesting and helpful read on the topic for those who might
have
some ... trepidation.
I'm not sayin' we are ready to shut this list down, but it's honestly
worth
considering if a different approach will be more effective.
Before the daggers come out.... please take some time to checkout Fedora Discourse https://discussion.fedoraproject.org/ and read the Foreman community link above.
As I had previously mentioned, Discourse allows for RSS feeds and email notifications. It also has a mailing list mode.
From the foreman post, in the comments: Email interface does not really work, there should be rather called “notifications”. You must visit the site eventually.
I personally find this a fatal flaw, I do not have time to waste visiting slow websites for each project I follow. If I have to switch away from my mail client I just end up unfollowing the project unless I am focusing specifically on it at that time.
It really is a much better solution.
Better for whom and for what purpose ?
I am sure it won't be better for me, all "web forums" I ever used made for my disappearence from the platform as it was too expensive (time- wise) for me to keep up.
Simo.
On Tue, 2018-10-16 at 13:31 -0400, Simo Sorce wrote:
On Tue, 2018-10-16 at 07:12 -0700, Gerald B. Cox wrote:
On Fri, Oct 5, 2018 at 9:21 AM Matthew Miller mattdm@fedoraproject.org wrote: ...
That's why the general trend is *away* from email.
The Foreman community recently switched away from mailing lists in this
way,
and
https://theforeman.org/2018/07/discourse-6-months-on-impact-assesment.html
is really interesting and helpful read on the topic for those who might
have
some ... trepidation.
I'm not sayin' we are ready to shut this list down, but it's honestly
worth
considering if a different approach will be more effective.
Before the daggers come out.... please take some time to checkout Fedora Discourse https://discussion.fedoraproject.org/ and read the Foreman community link above.
As I had previously mentioned, Discourse allows for RSS feeds and email notifications. It also has a mailing list mode.
From the foreman post, in the comments: Email interface does not really work, there should be rather called “notifications”. You must visit the site eventually.
I personally find this a fatal flaw, I do not have time to waste visiting slow websites for each project I follow. If I have to switch away from my mail client I just end up unfollowing the project unless I am focusing specifically on it at that time.
It really is a much better solution.
Better for whom and for what purpose ?
I am sure it won't be better for me, all "web forums" I ever used made for my disappearence from the platform as it was too expensive (time- wise) for me to keep up.
Ah nice side-thing I just found out, discussion.fedoraproject.org does not show *anything* unless I let my browser access this website http://discourse-cdn-sjc1.com
That doesn't feel right.
Simo.
On Tue, Oct 16, 2018 at 1:35 PM Simo Sorce simo@redhat.com wrote:
Ah nice side-thing I just found out, discussion.fedoraproject.org does not show *anything* unless I let my browser access this website http://discourse-cdn-sjc1.com
That doesn't feel right.
Fedora is not hosting its Discourse instance. And I think it appears to be a general trend of no longer self-hosting the infrastructure we leverage. Personally, I think that's a mistake that will probably bite us again (remember Transifex?). But if that's the way we want to go, oh well.
On Tue, Oct 16, 2018 at 10:39 AM Neal Gompa ngompa13@gmail.com wrote:
On Tue, Oct 16, 2018 at 1:35 PM Simo Sorce simo@redhat.com wrote:
Ah nice side-thing I just found out, discussion.fedoraproject.org does not show *anything* unless I let my browser access this website http://discourse-cdn-sjc1.com
That doesn't feel right.
Fedora is not hosting its Discourse instance. And I think it appears to be a general trend of no longer self-hosting the infrastructure we leverage. Personally, I think that's a mistake that will probably bite us again (remember Transifex?). But if that's the way we want to go, oh well.
It's most likely about budgets and manpower. Many companies now have outsourced many parts of their infrastructure... especially email - it's usually cheaper, and many companies no longer want to deal with supporting it.
On Tuesday, 16 October 2018 at 19:35, Simo Sorce wrote:
On Tue, 2018-10-16 at 13:31 -0400, Simo Sorce wrote:
On Tue, 2018-10-16 at 07:12 -0700, Gerald B. Cox wrote:
On Fri, Oct 5, 2018 at 9:21 AM Matthew Miller mattdm@fedoraproject.org wrote: ...
That's why the general trend is *away* from email.
The Foreman community recently switched away from mailing lists in this
way,
and
https://theforeman.org/2018/07/discourse-6-months-on-impact-assesment.html
is really interesting and helpful read on the topic for those who might
have
some ... trepidation.
I'm not sayin' we are ready to shut this list down, but it's honestly
worth
considering if a different approach will be more effective.
Before the daggers come out.... please take some time to checkout Fedora Discourse https://discussion.fedoraproject.org/ and read the Foreman community link above.
As I had previously mentioned, Discourse allows for RSS feeds and email notifications. It also has a mailing list mode.
From the foreman post, in the comments: Email interface does not really work, there should be rather called “notifications”. You must visit the site eventually.
I personally find this a fatal flaw, I do not have time to waste visiting slow websites for each project I follow. If I have to switch away from my mail client I just end up unfollowing the project unless I am focusing specifically on it at that time.
It really is a much better solution.
Better for whom and for what purpose ?
I am sure it won't be better for me, all "web forums" I ever used made for my disappearence from the platform as it was too expensive (time- wise) for me to keep up.
Ah nice side-thing I just found out, discussion.fedoraproject.org does not show *anything* unless I let my browser access this website http://discourse-cdn-sjc1.com
That doesn't feel right.
It kind of works with JavaScript not allowed from both discussion.fedoraproject.org and discourse-cdn-sjc1.com, but if you allow JS from discussion.fedoraproject.org only, it just shows you a blank page.
And I agree, too much clicking. Reading e-mail with mutt on a terminal is so much faster.
Regards, Dominik
On Tuesday, 16 October 2018 at 22:38, Dominik 'Rathann' Mierzejewski wrote: [...]
And I agree, too much clicking. Reading e-mail with mutt on a terminal is so much faster.
So, I tried the keyboard navigation mentioned elsewhere in the thread. Not easy to use at all and still have to use the browser.
I tried the mailing list mode to have it send me e-mail messages, but the messages are multipart/alternative (html+text) with the text part mangled and containing pieces of HTML code. Failure here as well.
For someone who doesn't like to use a browser to access e-mail or any kind of discussion-type messaging, Discourse is not a good tool in my opinion.
If it had a NNTP interface, I'd give it another try.
Regards, Dominik
On Wed, Oct 17, 2018 at 5:07 AM Dominik 'Rathann' Mierzejewski < dominik@greysector.net> wrote:
On Tuesday, 16 October 2018 at 22:38, Dominik 'Rathann' Mierzejewski wrote: [...]
And I agree, too much clicking. Reading e-mail with mutt on a terminal is so much faster.
So, I tried the keyboard navigation mentioned elsewhere in the thread. Not easy to use at all and still have to use the browser.
I believe you meant to say "still have to use a mouse." The keyboard navigation is for the browser mode - so you are using the browser to access it. I've tried the shortcuts and use a few... I mix and match.
I tried the mailing list mode to have it send me e-mail messages, but the messages are multipart/alternative (html+text) with the text part mangled and containing pieces of HTML code. Failure here as well.
Could this be because of the particular email client you're using? I use gmail and they look fine. Probably not fair to judge the system on the capabilities of your email client. Wouldn't the same thing happen if people sent you email from gmail, etc. with embedded html? What the mailing list mode does do is to allow you to read and respond without a browser.
For someone who doesn't like to use a browser to access e-mail or any kind of discussion-type messaging, Discourse is not a good tool in my opinion.
That's fair... if your don't like to use a browser to access email. You're entitled to your preferences and opinion - but many people do like using browser based email.
On Wed, Oct 17, 2018 at 07:23:28AM -0700, Gerald B. Cox wrote:
On Wed, Oct 17, 2018 at 5:07 AM Dominik 'Rathann' Mierzejewski < dominik@greysector.net> wrote:
I tried the mailing list mode to have it send me e-mail messages, but the messages are multipart/alternative (html+text) with the text part mangled and containing pieces of HTML code. Failure here as well.
Could this be because of the particular email client you're using? I use gmail and they look fine. Probably not fair to judge the system on the capabilities of your email client. Wouldn't the same thing happen if people sent you email from gmail, etc. with embedded html? What the mailing list mode does do is to allow you to read and respond without a browser.
No, if an email sender (Discourse) uses multipart/alternative with two attachments, text/plain and text/html, then it is required that the sender format the text/plain part as text with no HTML in it. It is also required to send semantically similar contents in both the text/plain and text/html parts, so that the text/plain part can act as a real human-readable alternative in the multipart/alternative message (i.e. it shouldn't set the text/plain part to be something like "Your email client doesn't support HTML. Please open this message in an HTML-capable client.")
In my experience so far with the 3 messages I've received from Discourse in mailing-list mode, it doesn't send raw HTML in the text/plain attachment. It sends something that looks like BBcode:
[quote="znmeb, post:12, topic:129, full:true"] GitHub is best for discovery - nobody searches GitLab or Bitbucket for projects. I use them for private repos, but if I want to engage a wider community than my workstation or laptop, I’m going to put my repo on GitHub. [/quote]
On Wed, Oct 17, 2018 at 7:38 AM Anderson, Charles R cra@wpi.edu wrote:
On Wed, Oct 17, 2018 at 07:23:28AM -0700, Gerald B. Cox wrote:
On Wed, Oct 17, 2018 at 5:07 AM Dominik 'Rathann' Mierzejewski < dominik@greysector.net> wrote:
I tried the mailing list mode to have it send me e-mail messages, but the messages are multipart/alternative (html+text) with the text part mangled and containing pieces of HTML code. Failure here as well.
Could this be because of the particular email client you're using? I use gmail and they look fine. Probably not fair to judge the system on the capabilities of your email client. Wouldn't the same thing happen if people sent you email from gmail, etc. with embedded html? What the mailing list mode does do is
to
allow you to read and respond without a browser.
No, if an email sender (Discourse) uses multipart/alternative with two attachments, text/plain and text/html, then it is required that the sender format the text/plain part as text with no HTML in it.
Where is that defined? Probably would be a good idea to open a bug with them if they are violating some official standard.
It is also required to send semantically similar contents in both the text/plain and text/html parts, so that the text/plain part can act as a real human-readable alternative in the multipart/alternative message (i.e. it shouldn't set the text/plain part to be something like "Your email client doesn't support HTML. Please open this message in an HTML-capable client.")
In my experience so far with the 3 messages I've received from Discourse in mailing-list mode, it doesn't send raw HTML in the text/plain attachment. It sends something that looks like BBcode:
Well, which is it? Rathann says html, you say bbcode? Regardless, as I mentioned above, if they have a bug, then it should be reported for them to address. Mailing list mode has been out for several years now - seems to me if this were a pervasive issue with a published standardize solution, they should take care of it.
Again, I use gmail and things look perfectly fine for me.
On Wed, Oct 17, 2018 at 10:55 AM Samuel Sieb samuel@sieb.net wrote:
On 10/17/18 8:52 AM, Gerald B. Cox wrote:
Again, I use gmail and things look perfectly fine for me.
That's because gmail only shows the HTML part.
Ah... so it's a client issue. Good to know.
On Wed, Oct 17, 2018 at 11:02:58AM -0700, Gerald B. Cox wrote:
On Wed, Oct 17, 2018 at 10:55 AM Samuel Sieb samuel@sieb.net wrote:
On 10/17/18 8:52 AM, Gerald B. Cox wrote:
Again, I use gmail and things look perfectly fine for me.
That's because gmail only shows the HTML part.
Ah... so it's a client issue. Good to know.
No. You are either completely misunderstanding the issue, or are arrogantly choosing to ignore it. Your tone comes across as the latter, but I apologize if I misconstrue your intent.
Let me try to explain again. Different people prefer different clients. Some people prefer text-only clients that have no capability to render HTML. That's okay--MIME (Multipurpose Internet Mail Extenstions RFC 2045 and RFC 2046) provide a way to support both text-only and HTML clients, called multipart/alternative in MIME. It is up to the sender of the email to support MIME multipart/alternative correctly by supplying meaningful content in two separate mail attachments--text/plain for text-only clients, and text/html for Gmail and other HTML-capable clients. If you are using Gmail you will only see the HTML part. You need to see the text/plain part to see what issues it has.
To claim that your preference of email client, Gmail, "looks fine" and then dismissing anyone else's issues as "a client issue" implies that you don't care about people who use different clients. That is not a way to garner support and reaching consensus for changing how the Fedora Project communicates with its members.
On Wed, Oct 17, 2018, 15:31 Anderson, Charles R cra@wpi.edu wrote:
On Wed, Oct 17, 2018 at 11:02:58AM -0700, Gerald B. Cox wrote:
On Wed, Oct 17, 2018 at 10:55 AM Samuel Sieb samuel@sieb.net wrote:
On 10/17/18 8:52 AM, Gerald B. Cox wrote:
Again, I use gmail and things look perfectly fine for me.
That's because gmail only shows the HTML part.
Ah... so it's a client issue. Good to know.
No. You are either completely misunderstanding the issue, or are arrogantly choosing to ignore it. Your tone comes across as the latter, but I apologize if I misconstrue your intent.
Let me try to explain again. Different people prefer different clients. Some people prefer text-only clients that have no capability to render HTML. That's okay--MIME (Multipurpose Internet Mail Extenstions RFC 2045 and RFC 2046) provide a way to support both text-only and HTML clients, called multipart/alternative in MIME. It is up to the sender of the email to support MIME multipart/alternative correctly by supplying meaningful content in two separate mail attachments--text/plain for text-only clients, and text/html for Gmail and other HTML-capable clients. If you are using Gmail you will only see the HTML part. You need to see the text/plain part to see what issues it has.
To claim that your preference of email client, Gmail, "looks fine" and then dismissing anyone else's issues as "a client issue" implies that you don't care about people who use different clients. That is not a way to garner support and reaching consensus for changing how the Fedora Project communicates with its members.
I'm not an email expert by any means. What I said was that it works perfectly fine for me. If people have an issue with it they should file a bug or enhancement request with the discourse project. That way the issue could be addressed and everyone could benefit.
On Wed, Oct 17, 2018 at 03:52:17PM -0300, Gerald B. Cox wrote:
I'm not an email expert by any means. What I said was that it works perfectly fine for me. If people have an issue with it they should file a bug or enhancement request with the discourse project. That way the issue could be addressed and everyone could benefit.
Okay, so you care only for your own needs? Everyone else is on their own?
You are advocating for a change to how Fedora communicates, while dismissing people's concerns about that change, and then putting the onus on those same people to help improve the third-party cloud hosted product so the change can work for them. How about we simply leave things the way they are instead? It has worked fine for 15+ years.
On Wed, Oct 17, 2018, 16:35 Anderson, Charles R cra@wpi.edu wrote:
On Wed, Oct 17, 2018 at 03:52:17PM -0300, Gerald B. Cox wrote:
I'm not an email expert by any means. What I said was that it works perfectly fine for me. If people have an issue with it they should file a bug or enhancement request with the discourse project. That way the issue could be addressed and everyone could benefit.
Okay, so you care only for your own needs? Everyone else is on their own?
You are advocating for a change to how Fedora communicates, while dismissing people's concerns about that change, and then putting the onus on those same people to help improve the third-party cloud hosted product so the change can work for them. How about we simply leave things the way they are instead? It has worked fine for 15+ years.
Again... I am not an email expert nor am I experiencing the issue. If you are having the issue you are best capable to report it.
As far as not being concerned... I wouldn't be engaging in a conversation if I weren't.
Things change... Technology advances...
On Wed, 2018-10-17 at 11:02 -0700, Gerald B. Cox wrote:
On Wed, Oct 17, 2018 at 10:55 AM Samuel Sieb samuel@sieb.net wrote:
On 10/17/18 8:52 AM, Gerald B. Cox wrote:
Again, I use gmail and things look perfectly fine for me.
That's because gmail only shows the HTML part.
Ah... so it's a client issue. Good to know.
No it is a service issue that just happen not to affect *your* client of choice. There is a big difference.
Simo.
On Thu, Oct 18, 2018 at 11:40 AM Simo Sorce simo@redhat.com wrote:
On Wed, 2018-10-17 at 11:02 -0700, Gerald B. Cox wrote:
On Wed, Oct 17, 2018 at 10:55 AM Samuel Sieb samuel@sieb.net wrote:
On 10/17/18 8:52 AM, Gerald B. Cox wrote:
Again, I use gmail and things look perfectly fine for me.
That's because gmail only shows the HTML part.
Ah... so it's a client issue. Good to know.
No it is a service issue that just happen not to affect *your* client of choice. There is a big difference.
Simo.
You need to read the entire thread in context, including subsequent responses. I realize that can be difficult on a mailing list, with all the top-posting, conversation snippets, etc.
On Thu, 2018-10-18 at 11:51 -0700, Gerald B. Cox wrote:
On Thu, Oct 18, 2018 at 11:40 AM Simo Sorce simo@redhat.com wrote:
On Wed, 2018-10-17 at 11:02 -0700, Gerald B. Cox wrote:
On Wed, Oct 17, 2018 at 10:55 AM Samuel Sieb samuel@sieb.net wrote:
On 10/17/18 8:52 AM, Gerald B. Cox wrote:
Again, I use gmail and things look perfectly fine for me.
That's because gmail only shows the HTML part.
Ah... so it's a client issue. Good to know.
No it is a service issue that just happen not to affect *your* client of choice. There is a big difference.
Simo.
You need to read the entire thread in context, including subsequent responses. I realize that can be difficult on a mailing list, with all the top-posting, conversation snippets, etc.
I did read the thread, and there is no difficulty at all to read a thread, given my client can do threading just fine, the comment stands.
You like gmail, and it just so happen discourse will generate mails that rendere well in gmail ... I would be ready to bet that is because the developers of that feature only test with gmail ...
The point was, you are looking only at your experience and not making an analysis of what the impact of a change is for *all* the users. You also state an opinion "discourse is just better" as fact, with a few other *opinions* only as supporting evidence.
I think discourse may be a decent tool for fedora-users@ or some other forum, where discussion is shorter and generally more of a quick answer/reply shot.
But devel@ is a completely different story, if you think discourse is much better is it possible that it's because your client doesn't handle mailing list very well and you should look for a better client ?
Simo.
On Thu, Oct 18, 2018 at 12:03 PM Simo Sorce simo@redhat.com wrote:
On Thu, 2018-10-18 at 11:51 -0700, Gerald B. Cox wrote:
On Thu, Oct 18, 2018 at 11:40 AM Simo Sorce simo@redhat.com wrote:
On Wed, 2018-10-17 at 11:02 -0700, Gerald B. Cox wrote:
On Wed, Oct 17, 2018 at 10:55 AM Samuel Sieb samuel@sieb.net
wrote:
On 10/17/18 8:52 AM, Gerald B. Cox wrote:
Again, I use gmail and things look perfectly fine for me.
That's because gmail only shows the HTML part.
Ah... so it's a client issue. Good to know.
No it is a service issue that just happen not to affect *your* client of choice. There is a big difference.
Simo.
You need to read the entire thread in context, including subsequent responses. I realize that can be difficult on a mailing list, with all the top-posting, conversation snippets, etc.
I did read the thread, and there is no difficulty at all to read a thread, given my client can do threading just fine, the comment stands.
Well, maybe you need to read it again - because you're missing the point.
On Thu, 2018-10-18 at 12:12 -0700, Gerald B. Cox wrote:
On Thu, Oct 18, 2018 at 12:03 PM Simo Sorce simo@redhat.com wrote:
On Thu, 2018-10-18 at 11:51 -0700, Gerald B. Cox wrote:
On Thu, Oct 18, 2018 at 11:40 AM Simo Sorce simo@redhat.com wrote:
On Wed, 2018-10-17 at 11:02 -0700, Gerald B. Cox wrote:
On Wed, Oct 17, 2018 at 10:55 AM Samuel Sieb samuel@sieb.net
wrote:
On 10/17/18 8:52 AM, Gerald B. Cox wrote: > Again, I use gmail and things look perfectly fine for me.
That's because gmail only shows the HTML part.
Ah... so it's a client issue. Good to know.
No it is a service issue that just happen not to affect *your* client of choice. There is a big difference.
Simo.
You need to read the entire thread in context, including subsequent responses. I realize that can be difficult on a mailing list, with all the top-posting, conversation snippets, etc.
I did read the thread, and there is no difficulty at all to read a thread, given my client can do threading just fine, the comment stands.
Well, maybe you need to read it again - because you're missing the point.
Your quoting tells me half the issue is in the setup you chose or are forced to use.
I think I understand the point, I just do not reach the same conclusions you do, each of us reach their conclusions based on their experiences. So perhaps we are both reaching wrong conclusions, I am not saying I am *right*, what I am saying is that you are not offering compelling reasons *to me*.
Simo.
On Thu, Oct 18, 2018 at 03:02:52PM -0400, Simo Sorce wrote:
On Thu, 2018-10-18 at 11:51 -0700, Gerald B. Cox wrote:
On Thu, Oct 18, 2018 at 11:40 AM Simo Sorce simo@redhat.com wrote:
On Wed, 2018-10-17 at 11:02 -0700, Gerald B. Cox wrote:
On Wed, Oct 17, 2018 at 10:55 AM Samuel Sieb samuel@sieb.net wrote:
On 10/17/18 8:52 AM, Gerald B. Cox wrote:
Again, I use gmail and things look perfectly fine for me.
That's because gmail only shows the HTML part.
Ah... so it's a client issue. Good to know.
No it is a service issue that just happen not to affect *your* client of choice. There is a big difference.
You need to read the entire thread in context, including subsequent responses. I realize that can be difficult on a mailing list, with all the top-posting, conversation snippets, etc.
I did read the thread, and there is no difficulty at all to read a thread, given my client can do threading just fine, the comment stands.
You like gmail […]
Using GMail (both legacy and Inbox) as a representation of email workflow and ergonomy is not fair. Gmail as a client is abysmal. No threading, no coloring of different level of citation, no integrated GPG support, no comfortable editor. Honest comparison would be between Discourse and Mutt+procmail, or Gnus. If one insist on using GMail as a server (instead of running one), I believe there's a IMAP access method which should work with mutt.
On Thu, Oct 18, 2018 at 12:24 PM Tomasz Torcz tomek@pipebreaker.pl wrote:
You like gmail […]
Using GMail (both legacy and Inbox) as a representation of email workflow and ergonomy is not fair. Gmail as a client is abysmal. No threading, no coloring of different level of citation, no integrated GPG support, no comfortable editor. Honest comparison would be between Discourse and Mutt+procmail, or Gnus. If one insist on using GMail as a server (instead of running one), I believe there's a IMAP access method which should work with mutt.
I use gmail... I don't like it... it's a software tool... there is no emotional attachment. That said, the point of this thread isn't gmail.
On Thu, Oct 18, 2018 at 09:24:03PM +0200, Tomasz Torcz wrote:
Using GMail (both legacy and Inbox) as a representation of email workflow and ergonomy is not fair. Gmail as a client is abysmal. No threading, no coloring of different level of citation, no integrated GPG support, no comfortable editor. Honest comparison would be between Discourse and Mutt+procmail, or Gnus. If one insist on using GMail as a server (instead of running one), I believe there's a IMAP access method which should work with mutt.
I agree that Gmail is awful. And, moreso, specifically hostile to mailing lists.
However — and I say this as a die-hard mutt user with my own postfix server — we absolutely, no way, should make it seem like running such a setup is a prerequisite to involvement in Fedora development. If that's "fair", I don't want "fair".
On Thu, Oct 18, 2018 at 11:40 AM Simo Sorce simo@redhat.com wrote:
You need to read the entire thread in context, including subsequent responses. I realize that can be difficult on a mailing list, with all the top-posting, conversation snippets, etc.
We have been doing that. The features touted for Discourse are automatic and natural for email lists, and said to be poorly mimicked in Discourse.
On Thu, Oct 18, 2018 at 12:28 PM tonynelson@georgeanelson.com wrote:
On Thu, Oct 18, 2018 at 11:40 AM Simo Sorce simo@redhat.com wrote:
You need to read the entire thread in context, including subsequent responses. I realize that can be difficult on a mailing list, with all the top-posting, conversation snippets, etc.
We have been doing that. The features touted for Discourse are automatic and natural for email lists, and said to be poorly mimicked in Discourse.
You're missing the point of this thread - it's about the capabilities of a
tool to foster discussion and communication between mulitiple people - it's not about cloning email software.
On Thu, Oct 18, 2018 at 12:34:45PM -0700, Gerald B. Cox wrote:
You're missing the point of this thread - it's about the capabilities of a tool to foster discussion and communication between mulitiple people - it's not about cloning email software.
What is "email software", if not a tool to foster discussion and communication between multiple people?
- Solomon
On Wednesday, 17 October 2018 at 17:52, Gerald B. Cox wrote:
On Wed, Oct 17, 2018 at 7:38 AM Anderson, Charles R cra@wpi.edu wrote:
[...]
It is also required to send semantically similar contents in both the text/plain and text/html parts, so that the text/plain part can act as a real human-readable alternative in the multipart/alternative message (i.e. it shouldn't set the text/plain part to be something like "Your email client doesn't support HTML. Please open this message in an HTML-capable client.")
In my experience so far with the 3 messages I've received from Discourse in mailing-list mode, it doesn't send raw HTML in the text/plain attachment. It sends something that looks like BBcode:
Well, which is it? Rathann says html, you say bbcode?
I didn't say it was HTML, I only said it was mangled, but Charles is right, it looks like BBcode, which doesn't really work in text mode. The standard for quoting is prefixing each line of quoted text with "> ". Highlighting that is implemented by default in mutt.
Regardless, as I mentioned above, if they have a bug, then it should be reported for them to address.
I wouldn't call it a bug, just bad UX for a minority(?) target audience.
Mailing list mode has been out for several years now - seems to me if this were a pervasive issue with a published standardize solution, they should take care of it.
I suspect the kind of people that would be bothered by Discourse's mailing list mode deficiencies simply stop using it as soon as they find out how bad it is for them. I don't blame them.
Again, I use gmail and things look perfectly fine for me.
That's why. Try mutt or another text-based client and you'll see the difference.
And personally, I recommend against Gmail because it's known to accept and silently drop some e-mail, which is against RFC5321 section 6.1.-6.2.
Regards,
Required disclosure: Mailman dev, and my sympathies are with the list advocates for this channel both for that reason, and for more objective ones. I don't really argue against a move to Discourse here, but I do know a bit about the problem space, and I'd like to discuss *some* aspects here. I expect it's clear which I prefer, but there are a lot of arguments "for" that I don't deal with.
Dominik 'Rathann' Mierzejewski writes:
Gerald B. Cox writes:
Regardless, as I mentioned above, if they have a bug, then it should be reported for them to address.
I wouldn't call it a bug, just bad UX for a minority(?) target audience.
I'm pretty sure the Discourse developers would concede that bad UX for email users is undesirable. I'm *not* sure they would concede that it's bad UX. We know how to render HTML to plain text, to RTF, and so on; Discourse deliberately chose not to do so, but rather chose a format of their own or perhaps some existing format (this is the first I've heard of BBcode, so I can't judge).
Thus, I doubt that reporting an issue against Discourse's email functionality would get action any time soon from the developers, and might get a lot of pushback. Given that support from Red Hat/Fedora community for Mailman development has decreased dramatically, I don't see why they would want to pick up responsibility for patching up Discourse instead. So I see years of annoyance for those who have really effective email workflows, being constrained to Discourse's.
I suspect that the number of people who really want email is at least as big as those who really want a forum, and they probably contribute more code and admin (which are typically communication-intensive, multicast activities), though forum-oriented users may contribute greatly in other ways (user support has been mentioned, and I wouldn't be surprised if that's easier both for the OP and responders in forum format -- synchronous communication does a lot of work to avoid redundancy, and editing out mistaken information is a real feature for user support). And I suppose there are a lot of people in the middle who don't care much either way (ignoring costs of moving to a new channel, which are significant in the short run but not really in the long run IMO), probably a majority.
Gerald's personal anecdotes suggest that people who are weakly attached to the channel (jumping into threads in the middle, signing up for their one issue, etc) are going to find the forum format more convenient. I find his discussion persuasive on that, as I have no problem with the forum format when I have a drive-by interest in some channel. This is *not* a put-down, it's simply a statement that a certain group of users would probably be happier with a forum.
(I'm not sure user support is a goal of this channel, but I've seen many threads devoted to issues of system configuration that have zero likelihood of inducing development of Fedora.)
Mailing list mode has been out for several years now - seems to me if this were a pervasive issue with a published standardize solution, they should take care of it.
Email doesn't have a standardized solution or standardized format for content. That's why people like HTML email -- it's a quasi-standard mess that browsers and browser-derived messaging clients have huge code bases for handling, and 25 years on have gotten pretty good at it. Most clients' outgoing messages now are pretty close to HTML 5 conformant, but there's still a huge mismatch with "traditional" email clients because email standards use a different mechanism from HTML for embedding objects in the data stream. (That may not be a big problem with Discourse depending on how it handles such media, and how the lists are configured.)
The standards that have been referred to are SMTP (RFC 5321, which basically requires that if you accept a connection at the TCP level, you should say you're refusing to accept mail rather than dropping it silently), the Internet message format (RFC 5322, which basically defines email as an ASCII text stream divided into a header containing various metadata and a body), and the Multimedia Mail Extensions (MIME, with a plethora of RFCs starting with the 2045-2049 series) which define ways to encode non-ASCII data (both text and non-text), ways to include non-text data, and ways to mix various media, in a message body. Among the latter is the multipart/alternative body part (defined in RFC 2046, IIRC), which allows catering to various user agents, and does require that the alternatives have as close to the same semantics as the media make possible.
This provision of RFC 2046 is violated by a lot of HTML-oriented clients, which embed attachments in the HTML where they can't easily be accessed by non-HTML clients, or even provide a plain text alternative that says "use an HTML client", rather than a plain text version of the HTML content. I don't think that's a problem with Discourse, but I don't understand why they wouldn't use the traditional email conventions for quoted material in the body of a text/plain part. The fact that they deliberately did something different suggests that they think they know better than we do. :-(
To the extent that HTML email is the source of posts to Discourse, there's not much that the software can do much about these differences (without incorporating the huge body of code mentioned above). MIME on the other hand can be handled with a reasonable amount of code.
I suspect the kind of people that would be bothered by Discourse's mailing list mode deficiencies simply stop using it as soon as they find out how bad it is for them. I don't blame them.
I've heard reports of that in multiple communities now (this one, Python, and Emacs, and gratuitously on Mailman lists[1]). That said, while I think the great majority of list users would be at least temporarily inconvenienced by a switch to forum software, those who really care either way are likely to be minorities. My suspicion is that those who really dislike the forum format contribute a lot more to the community than those who really like it. It's anecdotal, but in all cases where I've seen suggestions to convert MLs to forums, the strongest proponents are people whose main current contribution to discussion is to advocate a forum, while people who do a lot of day-to-day work generally resist, some very strongly. Of course it's possible that for proponents forums are so efficient that it's worth dropping everything else to advocate them.
OT, but I can't resist.
Again, I use gmail and things look perfectly fine for me.
Speak of bad UX for a minority audience, and GMail will be presented as an example of perfectly fine. :-(
And personally, I recommend against Gmail because it's known to accept and silently drop some e-mail, which is against RFC5321 section 6.1.-6.2.
I've talked to some people who should be in the know (via DMARC dev channels) about that. They claim that they do it only for for identified bad actors, where they're 99% sure it's spam and 99% sure that an SMTP "administrative denial" will result in backscatter, as the return path goes through, uh, less-than-diligent intermediate MXes.
As a mailing list owner, I think that's reasonable, though a real problem for the sender and receiver in the 0.01% of cases where they get it wrong. They probably get it wrong thousands of times a day, they process an awful lot of email.... OTOH, if some list's MTA gets banned by Yahoo! or Microsoft for backscatter, that's often a couple thousand emails per day that don't go through.
YMMV, of course. I'm just presenting the strongest argument I know in favor of the practice. It's also evidence that it won't go away (and I suspect other large freemail providers do the same).
Footnotes: [1] It's off-topic, and we try to keep badmouthing alternative paradigms to a minimum. ;-)
On Sat, Oct 20, 2018 at 6:09 AM Stephen J. Turnbull stephen@xemacs.org wrote:
Dominik 'Rathann' Mierzejewski writes:
Gerald B. Cox writes:
Regardless, as I mentioned above, if they have a bug, then it should be reported for them to address.
I wouldn't call it a bug, just bad UX for a minority(?) target audience.
I'm pretty sure the Discourse developers would concede that bad UX for email users is undesirable. I'm *not* sure they would concede that it's bad UX. We know how to render HTML to plain text, to RTF, and so on; Discourse deliberately chose not to do so, but rather chose a format of their own or perhaps some existing format (this is the first I've heard of BBcode, so I can't judge).
It is not a format of their own, but it's not appropriate for plaintext, so it sounds like a bug to me.
https://en.wikipedia.org/wiki/BBCode
-Dan
Dan Book writes:
It is not a format of their own, but it's not appropriate for plaintext, so it sounds like a bug to me.
But that's exactly my point. *We* think it's a bug, but *they* chose it deliberately. Undoubtedly people have tools expecting it, etc.
I've been on both sides of that proposition, and the maintainers bat about .900.
On Sat, Oct 20, 2018 at 3:05 AM Stephen J. Turnbull stephen@xemacs.org wrote:
Required disclosure: Mailman dev, and my sympathies are with the list advocates for this channel both for that reason, and for more objective ones. I don't really argue against a move to Discourse here, but I do know a bit about the problem space, and I'd like to discuss *some* aspects here. I expect it's clear which I prefer, but there are a lot of arguments "for" that I don't deal with.
Dominik 'Rathann' Mierzejewski writes:
Gerald B. Cox writes:
Regardless, as I mentioned above, if they have a bug, then it should be reported for them to address.
I wouldn't call it a bug, just bad UX for a minority(?) target audience.
I'm pretty sure the Discourse developers would concede that bad UX for email users is undesirable. I'm *not* sure they would concede that it's bad UX. We know how to render HTML to plain text, to RTF, and so on; Discourse deliberately chose not to do so, but rather chose a format of their own or perhaps some existing format (this is the first I've heard of BBcode, so I can't judge).
Thus, I doubt that reporting an issue against Discourse's email functionality would get action any time soon from the developers, and might get a lot of pushback.
Thanks very much for your detailed reply. You summed it up pretty well. As far as the mailing list issue for some email clients I did find this over on the Discourse support forum:
https://meta.discourse.org/t/plaintext-and-or-raw-emails-for-mailing-list-mo...
Again, I'm far from being an email expert... I'm just a user of it - but people who do have the knowledge might want to take a look and see if this discussion reflects some of their concerns and possibly participate. I don't know if the situation could or could not be improved... but it doesn't hurt to ask.
"GBC" == Gerald B Cox gbcox@bzb.us writes:
GBC> Probably not fair to judge the system on the capabilities of your GBC> email client.
It is entirely fair. Moving to this system will remove capabilities that I have now and will not provide adequate replacements for them. Since moving to this system will most likely result in my disconnection from this portion of the project, I don't intend to ignore its faults.
GBC> That's fair... if your don't like to use a browser to access email. GBC> You're entitled to your preferences and opinion - but many people GBC> do like using browser based email.
I completely agree. And the nice thing about our existing mailing list setup is that they can do that. And people who don't like using browser based email don't have to. Because it's all just email, and we can all access it and deal with it according to our own workflows.
Discourse simply enforces a workflow upon its users. Yes, there's a "mailing list mode". It's not sufficient. Yes, there is some keyboard navigation. It's not sufficient. I can fire up some browser addons to try to make the page look like something I can use (and hope they don't ever change their layout), but I can't add capabilities to it that it doesn't have. And if I insist on sticking with email, I get a second class experience.
So contrary to what you write above, I wouldn't be entitled to my preferences, because discourse doesn't give me the opportunity to have them. If I don't like my email client, I can change it. I can't change the discourse system.
On project I used to follow quite a bit was the RT request tracker. They moved to discourse some time ago. The end result was that email became a mostly useless stream of context free messages and you basically have to open a browser to follow a conversation. Offline? Too bad. High latency connection? Too bad. I don't bother even attempting to follow that project any longer.
I don't expect to sway anyone's opinion or that Fedora will keep mailing lists around just for me. But in this push for some nebulous "engagement" I believe there will also be disengagement, and that by doing this Fedora itself is making a trade.
- J<
On Wed, Oct 17, 2018 at 12:28 PM Jason L Tibbitts III tibbs@math.uh.edu wrote:
On project I used to follow quite a bit was the RT request tracker. They moved to discourse some time ago. The end result was that email became a mostly useless stream of context free messages and you basically have to open a browser to follow a conversation. Offline? Too bad. High latency connection? Too bad. I don't bother even attempting to follow that project any longer.
I don't expect to sway anyone's opinion or that Fedora will keep mailing lists around just for me. But in this push for some nebulous "engagement" I believe there will also be disengagement, and that by doing this Fedora itself is making a trade.
While I'm not a contributor to these lists, I do follow them and have so for probably 20 years now. Like Jason, I have had other projects move away from mailing lists, and that now means I have to make a concious effort to 'poll' their system, rather than having a push notification. Needless to say, I don't follow them any more, and I don't know what's happening there.
Push means I find out whenever something happens. If nothing happens... then that's OK with me, I'm not bothered. But to be constantly polling for 'is there anything new... is there anything new' (whether there is or not), is just plain inefficient.
I hope that doesn't happen to Fedora too.
[Just my 2 cents worth.]
On Wed, Oct 17, 2018 at 9:40 AM Fulko Hew fulko.hew@gmail.com wrote:
On Wed, Oct 17, 2018 at 12:28 PM Jason L Tibbitts III tibbs@math.uh.edu wrote:
While I'm not a contributor to these lists, I do follow them and have so for probably 20 years now. Like Jason, I have had other projects move away from mailing lists, and that now means I have to make a concious effort to 'poll' their system, rather than having a push notification. Needless to say, I don't follow them any more, and I don't know what's happening there.
Push means I find out whenever something happens. If nothing happens... then that's OK with me, I'm not bothered. But to be constantly polling for 'is there anything new... is there anything new' (whether there is or not), is just plain inefficient.
I hope that doesn't happen to Fedora too.
[Just my 2 cents worth.]
Again, especially for a user of gmail, there is mailing list mode. People
keep saying it isn't sufficient or it doesn't work. I've been using it for 3 days and looks and acts like a normal mailing list.
On Wed, Oct 17, 2018 at 09:49:49AM -0700, Gerald B. Cox wrote:
On Wed, Oct 17, 2018 at 9:40 AM Fulko Hew fulko.hew@gmail.com wrote:
On Wed, Oct 17, 2018 at 12:28 PM Jason L Tibbitts III tibbs@math.uh.edu wrote:
While I'm not a contributor to these lists, I do follow them and have so for probably 20 years now. Like Jason, I have had other projects move away from mailing lists, and that now means I have to make a concious effort to 'poll' their system, rather than having a push notification. Needless to say, I don't follow them any more, and I don't know what's happening there.
Push means I find out whenever something happens. If nothing happens... then that's OK with me, I'm not bothered. But to be constantly polling for 'is there anything new... is there anything new' (whether there is or not), is just plain inefficient.
I hope that doesn't happen to Fedora too.
[Just my 2 cents worth.]
Again, especially for a user of gmail, there is mailing list mode. People
keep saying it isn't sufficient or it doesn't work. I've been using it for 3 days and looks and acts like a normal mailing list.
How do I start a thread on Discourse from email? We should start this discussion over there so we can experience it ourselves.
On Wed, Oct 17, 2018 at 05:00:32PM +0000, Anderson, Charles R wrote:
How do I start a thread on Discourse from email? We should start this discussion over there so we can experience it ourselves.
So, yeah, that's a thing: we currently have that feature turned off, because I'm worried about spam. So you can reply to threads, but not start them.
Discourse is *definitely* not a smooth, drop-in mailing list replacement like Hyperkitty is.
On Wed, Oct 17, 2018 at 01:14:32PM -0400, Matthew Miller wrote:
On Wed, Oct 17, 2018 at 05:00:32PM +0000, Anderson, Charles R wrote:
How do I start a thread on Discourse from email? We should start this discussion over there so we can experience it ourselves.
So, yeah, that's a thing: we currently have that feature turned off, because I'm worried about spam. So you can reply to threads, but not start them.
Discourse is *definitely* not a smooth, drop-in mailing list replacement like Hyperkitty is.
Well then, that's a showstopper for me. I was honestly ready to give it a try.
On Wed, Oct 17, 2018 at 10:15 AM Matthew Miller mattdm@fedoraproject.org wrote:
On Wed, Oct 17, 2018 at 05:00:32PM +0000, Anderson, Charles R wrote:
How do I start a thread on Discourse from email? We should start this discussion over there so we can experience it ourselves.
So, yeah, that's a thing: we currently have that feature turned off, because I'm worried about spam. So you can reply to threads, but not start them.
Matt, would probably be a good idea to turn that on so people could try it out. People can also try it out at the following sites:
discourse.mozilla.org meta.discourse.org
There are countless others, just do a search.
Discourse is *definitely* not a smooth, drop-in mailing list replacement like Hyperkitty is.
Of course it's not identical to a what we have now... that is kind of the point. However, from what I have seen the mailing list mode is a good compromise for those who want to continue that mode of communication.
I would submit bug reports if I had a problem with the features or capabilities of mailing list mode - however, for me it works perfectly. I would encourage those who aren't happy with it to take a few moments to follow up with the discourse people to either file a bug or request an enhancement. It is to everyone's best interest to help to improve the product.
On Wed, Oct 17, 2018 at 1:28 PM Gerald B. Cox gbcox@bzb.us wrote:
On Wed, Oct 17, 2018 at 10:15 AM Matthew Miller mattdm@fedoraproject.org wrote:
On Wed, Oct 17, 2018 at 05:00:32PM +0000, Anderson, Charles R wrote:
How do I start a thread on Discourse from email? We should start this discussion over there so we can experience it ourselves.
So, yeah, that's a thing: we currently have that feature turned off, because I'm worried about spam. So you can reply to threads, but not start them.
Matt, would probably be a good idea to turn that on so people could try it out. People can also try it out at the following sites:
discourse.mozilla.org meta.discourse.org
There are countless others, just do a search.
Discourse is *definitely* not a smooth, drop-in mailing list replacement like Hyperkitty is.
Of course it's not identical to a what we have now... that is kind of the point. However, from what I have seen the mailing list mode is a good compromise for those who want to continue that mode of communication.
I would submit bug reports if I had a problem with the features or capabilities of mailing list mode - however, for me it works perfectly. I would encourage those who aren't happy with it to take a few moments to follow up with the discourse people to either file a bug or request an enhancement. It is to everyone's best interest to help to improve the product.
The thing is, it doesn't matter. Discourse is *not* designed to support the types of discussions that do happen on these lists, nor is it designed to handle the load or the number of disparate conversations. I've experienced the switchover from mail lists to Discourse before with OpenMandriva, and asynchronous development discussions basically died. The OpenMandriva developers (myself included) rely on IRC and IRC meetings even more so than we did before, because Discourse is just awful for this. And OpenMandriva is a hundredth of the scale of Fedora development list.
I cannot *ever* recommend, in good conscious, moving to Discourse for Fedora development discussions.
However, I think it's fantastic for user support, as those are much more context free, incidental, and so on. I've wished for a long time that we'd integrate fedoraforum.org into the main Fedora Project, but we never did that...
The thing is, it doesn't matter. Discourse is *not* designed to
support the types of discussions that do happen on these lists, nor is it designed to handle the load or the number of disparate conversations.
I've experienced the switchover from mail lists to
Discourse before with OpenMandriva, and asynchronous development discussions basically died. The OpenMandriva developers (myself included) rely on IRC and IRC meetings even more so than we did before, because Discourse is just awful for this. And OpenMandriva is a hundredth of the scale of Fedora development list.
Can you please articulate that difference? As far as traffic is concerned, that was addressed earlier in the thread and I don't see how that is an issue.
On Wed, Oct 17, 2018 at 10:48 AM Gerald B. Cox gbcox@bzb.us wrote:
The thing is, it doesn't matter. Discourse is *not* designed to
support the types of discussions that do happen on these lists, nor is it designed to handle the load or the number of disparate conversations.
I've experienced the switchover from mail lists to
Discourse before with OpenMandriva, and asynchronous development discussions basically died. The OpenMandriva developers (myself included) rely on IRC and IRC meetings even more so than we did before, because Discourse is just awful for this. And OpenMandriva is a hundredth of the scale of Fedora development list.
Can you please articulate that difference? As far as traffic is concerned, that was addressed earlier in the thread and I don't see how that is an issue.
Here is also a good link: https://meta.discourse.org/t/discourse-vs-email-mailing-lists/54298
On Wed, Oct 17, 2018 at 01:39:30PM -0400, Neal Gompa wrote:
I cannot *ever* recommend, in good conscious, moving to Discourse for Fedora development discussions.
However, I think it's fantastic for user support, as those are much more context free, incidental, and so on. I've wished for a long time
Yeah, I'm not suggesting moving all development discussion there, although I think there are people who *do* want to do development discussion that way, and I wouldn't want to _block_ that.
On Thu, Oct 18, 2018 at 5:33 AM Matthew Miller mattdm@fedoraproject.org wrote:
On Wed, Oct 17, 2018 at 01:39:30PM -0400, Neal Gompa wrote:
I cannot *ever* recommend, in good conscious, moving to Discourse for Fedora development discussions.
However, I think it's fantastic for user support, as those are much more context free, incidental, and so on. I've wished for a long time
Yeah, I'm not suggesting moving all development discussion there, although I think there are people who *do* want to do development discussion that way, and I wouldn't want to _block_ that.
Actually, I think that just creating a new discourse instance for discussions keeping the mailing lists around would be a good solution. You wouldn't have to worry about registration. People who wanted to use it or try it out could do that themselves - and you wouldn't have to be concerned about moving archives, etc. You could then just let normal attrition handle it.
On 10/18/18 6:31 AM, Gerald B. Cox wrote:
On Thu, Oct 18, 2018 at 5:33 AM Matthew Miller mattdm@fedoraproject.org wrote:
On Wed, Oct 17, 2018 at 01:39:30PM -0400, Neal Gompa wrote:
I cannot *ever* recommend, in good conscious, moving to Discourse for Fedora development discussions.
However, I think it's fantastic for user support, as those are much more context free, incidental, and so on. I've wished for a long time
Yeah, I'm not suggesting moving all development discussion there, although I think there are people who *do* want to do development discussion that way, and I wouldn't want to _block_ that.
Actually, I think that just creating a new discourse instance for discussions keeping the mailing lists around would be a good solution. You wouldn't have to worry about registration. People who wanted to use it or try it out could do that themselves - and you wouldn't have to be concerned about moving archives, etc. You could then just let normal attrition handle it.
Sure, but it might be not a great experience if not enough people watch both or move to discourse. ie, people asking questions there or starting discussions and no one answering them.
So, if people want to discuss in discourse, great, but I don't think we should create a vast sea of empty topics until/unless people are needing them.
kevin
On Thu, Oct 18, 2018 at 12:40 PM Kevin Fenzi kevin@scrye.com wrote:
On 10/18/18 6:31 AM, Gerald B. Cox wrote:
Actually, I think that just creating a new discourse instance for discussions keeping the mailing lists around would be a good solution. You wouldn't have to worry about registration. People who wanted to use it or try it out could do that themselves - and you wouldn't have
to
be concerned about moving archives, etc. You could then just let normal attrition handle it.
Sure, but it might be not a great experience if not enough people watch both or move to discourse. ie, people asking questions there or starting discussions and no one answering them.
So, if people want to discuss in discourse, great, but I don't think we should create a vast sea of empty topics until/unless people are needing them.
Yeah, it's kind of a chicken / egg situation - you're not going to know
until you throw something out there and let people give it a whirl and determine how it works for them in a real world situation - not some lab mockup. The Foreman results that Matt posted earlier appeared to indicate it would be worthwhile to try.
* Gerald B. Cox [18/10/2018 06:31] :
Actually, I think that just creating a new discourse instance for discussions keeping the mailing lists around would be a good solution.
The way I read this, it implies we would fragment developement disccusion in two, the people using mailing lists on one side and the people using discourse on the other.
Whatever is decided, I hope this does not happen.
Emmanuel
On Wed, 2018-10-17 at 13:14 -0400, Matthew Miller wrote:
On Wed, Oct 17, 2018 at 05:00:32PM +0000, Anderson, Charles R wrote:
How do I start a thread on Discourse from email? We should start this discussion over there so we can experience it ourselves.
So, yeah, that's a thing: we currently have that feature turned off, because I'm worried about spam. So you can reply to threads, but not start them.
Discourse is *definitely* not a smooth, drop-in mailing list replacement like Hyperkitty is.
I am trying to understand what is the goal of moving development discussions to Discourse in this case.
<a little snarky>Is it to deter busy people from participating in the hopes of some fleeting engagement by superficially interested people will fill the void ?</>
Simo.
On Wed, 2018-10-17 at 13:14 -0400, Matthew Miller wrote:
Discourse is *definitely* not a smooth, drop-in mailing list replacement like Hyperkitty is.
I'm curious what is insufficient about Hyperkitty that Discourse does well at. Wasn't Hyperkitty supposed to give people the forum experience? I admit I haven't used it that much for reading or posting (though I do use it for archive links sometimes and it seems fine for that), since I really like the e-mail interface, so I am not familiar with how nice or un-nice it is to use.
On Thu, Oct 18, 2018 at 1:15 PM Randy Barlow bowlofeggs@fedoraproject.org wrote:
On Wed, 2018-10-17 at 13:14 -0400, Matthew Miller wrote:
Discourse is *definitely* not a smooth, drop-in mailing list replacement like Hyperkitty is.
I'm curious what is insufficient about Hyperkitty that Discourse does well at. Wasn't Hyperkitty supposed to give people the forum experience? I admit I haven't used it that much for reading or posting (though I do use it for archive links sometimes and it seems fine for that), since I really like the e-mail interface, so I am not familiar with how nice or un-nice it is to use.
Here is a link to discourse features: https://www.discourse.org/features
I view hyperkitty as just a web interface for mailing lists - not much more than that. Discourse provides a more complete conversation / collaboration environment. If you take a look at the features page, you'll see it offers some nice capabilities.
I wouldn't expect it to be a "drop-in" mailing list replacement. Yes, it allows some backward compatibility by providing "mailing-list mode" - but you're going to get a richer experience if you use native interfaces.
On Thu, 2018-10-18 at 14:04 -0700, Gerald B. Cox wrote:
Here is a link to discourse features: https://www.discourse.org/features
I view hyperkitty as just a web interface for mailing lists - not much more than that. Discourse provides a more complete conversation / collaboration environment. If you take a look at the features page, you'll see it offers some nice capabilities.
Reading that page, I don't see a lot that seems substantially different from hyperkitty to me. Can you highlight what in particular you view as a valuable difference? What is lacking about Hyperkitty?
On Thu, Oct 18, 2018 at 9:11 PM Randy Barlow bowlofeggs@fedoraproject.org wrote:
On Thu, 2018-10-18 at 14:04 -0700, Gerald B. Cox wrote:
Here is a link to discourse features: https://www.discourse.org/features
I view hyperkitty as just a web interface for mailing lists - not much more than that. Discourse provides a more complete conversation / collaboration environment. If you take a look at the features page, you'll see it offers some nice capabilities.
Reading that page, I don't see a lot that seems substantially different from hyperkitty to me. Can you highlight what in particular you view as a valuable difference? What is lacking about Hyperkitty?
(CC'd Mairin Duffy, as she's probably the right person to give the feedback to, and could respond appropriately to my comments)
In my opinion, there's only a few things needed to make HyperKitty excel:
* Notifications hub for unread threads and such. For people who'd rather use HyperKitty to interface with the list, there needs to be a way to track threads and topics that you've participated in or are watching.
* HyperKitty completely lacks a means of handling rich content. Some people may not agree with me on this, but having *zero* formatting sucks. Supporting at least Markdown (probably with CommonMark specification?) would be a major improvement.
* There *needs* to be a flat view option. A lot of people don't realize this, but the reason why threaded views aren't normal in web forums is because they're absolutely awful for deep threads. The only reason it works in email is because email clients are usually too dumb to represent them accurately.
* Topic splits aren't rendered as separate conversations to follow, which makes it difficult to understand what's happening because per the thing about threaded views, it gets really weird and awkward to follow multiple conversations spawned from the same mail.
A nice to have would be a better visual style that's more responsive, exciting, and cleaner looking, but that's a matter of taste. I'm not much of a fan of the current era of Fedora web styles, as it feels like the life is sucked out of the site and it's just _boring_... Heck, I think the current Koji web style is more exciting than the current Bodhi and HyperKitty ones...
I'm open to all of those suggestions as well as committing to design and CSS work for them. I would need a web dev to help me though; I'm not great with Django.
Please note, the reason Hyperkitty didn't cause this sort of thread or honestly any sort of drama or controversy when it was deployed is because it required no current users to change anything about their workflow, with one small exception - mm3 didn't come out w topic support which was used on the packagers list. (I don't believe that's an issue anymore.)
The whole point of the design was to enable a new group of would be contributors be able to participate alongside the folks already there, so that everyone could participate together. Existing ML users never need to use Hyperkitty if they don't want to, and yet, users new to the project can start reading and participating in threads right away w no mail client config and never receiving a single email if they so desire.
I believe quite strongly (and have from the start when I first heard of the project) that Discourse's basic UX model is fundamentally flawed. If we deploy discourse and roll it out, we *may* get new users, but as noted in this thread, we will lose existing ones. Participating in upstream effort on Discourse, improving it, etc is foolish bc the fundamental model is broken.
When some people think of email, they think of mutt or thunderbird, annoying client config, setting up procmail or fetchmail or whatever other complex elaborate tools many of the ppl reading this use. Email is just a basic standard. Discourse does not follow that standard. There is no reason a social media timeline like experience for the teenagers is not possible using email as the underlying system. Jabber never really took off, except Google Hangouts and FB messenger both used it (no idea if they still do.) The reason our open standards like email and xmpp are dying off is bc the primary biz model of the companies that used them relies on getting eyes on ads, and scanning content in ways that mean giving users a choice of client that works best for their needs is off the table.
Basically dont confuse the front end youre used to with the underlying tech.
I think it's a better idea to use a tool based on open standards, that allows users to use the client experience that works best for them. If you try to force everyone down one road it won't work.
~m
Hi, Máirín.
On Friday, 19 October 2018 at 05:43, Máirín Duffy wrote: [...]
I believe quite strongly (and have from the start when I first heard of the project) that Discourse's basic UX model is fundamentally flawed. If we deploy discourse and roll it out, we *may* get new users, but as noted in this thread, we will lose existing ones. Participating in upstream effort on Discourse, improving it, etc is foolish bc the fundamental model is broken.
Agreed. Well said.
[...]
There is no reason a social media timeline like experience for the teenagers is not possible using email as the underlying system. Jabber never really took off, except Google Hangouts and FB messenger both used it (no idea if they still do.)
They used to, but no longer. Both have (d)evolved into their own proprietary protocols. For FB, there's a plugin for pidgin/libpurple that actually works quite well (purple-facebook package).
[...]
I think it's a better idea to use a tool based on open standards, that allows users to use the client experience that works best for them. If you try to force everyone down one road it won't work.
I fully agree.
If the company behind Discourse goes away for some reason and the project is not picked up by community, we'll end up having to support it ourselves and I don't think we have resources for this.
I don't think e-mail is going away anytime soon.
Regards, Dominik
On Fri, 2018-10-19 at 03:43 +0000, Máirín Duffy wrote:
I'm open to all of those suggestions as well as committing to design and CSS work for them. I would need a web dev to help me though; I'm not great with Django.
Please note, the reason Hyperkitty didn't cause this sort of thread or honestly any sort of drama or controversy when it was deployed is because it required no current users to change anything about their workflow, with one small exception - mm3 didn't come out w topic support which was used on the packagers list. (I don't believe that's an issue anymore.)
The whole point of the design was to enable a new group of would be contributors be able to participate alongside the folks already there, so that everyone could participate together. Existing ML users never need to use Hyperkitty if they don't want to, and yet, users new to the project can start reading and participating in threads right away w no mail client config and never receiving a single email if they so desire.
I believe quite strongly (and have from the start when I first heard of the project) that Discourse's basic UX model is fundamentally flawed. If we deploy discourse and roll it out, we *may* get new users, but as noted in this thread, we will lose existing ones. Participating in upstream effort on Discourse, improving it, etc is foolish bc the fundamental model is broken.
When some people think of email, they think of mutt or thunderbird, annoying client config, setting up procmail or fetchmail or whatever other complex elaborate tools many of the ppl reading this use. Email is just a basic standard. Discourse does not follow that standard. There is no reason a social media timeline like experience for the teenagers is not possible using email as the underlying system. Jabber never really took off, except Google Hangouts and FB messenger both used it (no idea if they still do.) The reason our open standards like email and xmpp are dying off is bc the primary biz model of the companies that used them relies on getting eyes on ads, and scanning content in ways that mean giving users a choice of client that works best for their needs is off the table.
Can we be careful with association of timelines with teenagers here? It's a tiny bit belittling, and makes us all sound old and out of touch, since my parents (who are obviously older than me) know what they are. I've been using Linux exclusively on the desktop for a bit over 15 years, so you can guess I'm not a teen.
Basically dont confuse the front end youre used to with the underlying tech.
I think it's a better idea to use a tool based on open standards, that allows users to use the client experience that works best for them. If you try to force everyone down one road it won't work.
~m _______________________________________________ devel mailing list -- devel@lists.fedoraproject.org
To unsubscribe send an email to devel-leave@lists.fedoraproject.org
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org
Re: teenagers and timelines, I'm just addressing the specific concerns that were raised to me.
~m
Máirín,
On 2018-10-19 14:43, Máirín Duffy wrote:
I'm open to all of those suggestions as well as committing to design and CSS work for them. I would need a web dev to help me though; I'm not great with Django.
Please note, the reason Hyperkitty didn't cause this sort of thread or honestly any sort of drama or controversy when it was deployed is because it required no current users to change anything about their workflow, with one small exception - mm3 didn't come out w topic support which was used on the packagers list. (I don't believe that's an issue anymore.)
The whole point of the design was to enable a new group of would be contributors be able to participate alongside the folks already there, so that everyone could participate together. Existing ML users never need to use Hyperkitty if they don't want to, and yet, users new to the project can start reading and participating in threads right away w no mail client config and never receiving a single email if they so desire.
I believe quite strongly (and have from the start when I first heard of the project) that Discourse's basic UX model is fundamentally flawed. If we deploy discourse and roll it out, we *may* get new users, but as noted in this thread, we will lose existing ones. Participating in upstream effort on Discourse, improving it, etc is foolish bc the fundamental model is broken.
When some people think of email, they think of mutt or thunderbird, annoying client config, setting up procmail or fetchmail or whatever other complex elaborate tools many of the ppl reading this use. Email is just a basic standard. Discourse does not follow that standard. There is no reason a social media timeline like experience for the teenagers is not possible using email as the underlying system. Jabber never really took off, except Google Hangouts and FB messenger both used it (no idea if they still do.) The reason our open standards like email and xmpp are dying off is bc the primary biz model of the companies that used them relies on getting eyes on ads, and scanning content in ways that mean giving users a choice of client that works best for their needs is off the table.
Basically dont confuse the front end youre used to with the underlying tech.
I think it's a better idea to use a tool based on open standards, that allows users to use the client experience that works best for them. If you try to force everyone down one road it won't work.
A nice, informed response!
Thanks,
Phil.
Again, I believe some are trying to do an apples to apples comparison with Discourse and mailing list technologies. Discourse was build from the ground up with the goal of fostering communication and collaboration. Hyperkitty is a bolt on HTML to mailing list archives. It's good for what it is, but it isn't Discourse - and usage numbers tend to bear that out.
The fact is that email usage is declining. People are moving away from it and prefer to use other platforms for collaboration. As with many things... when something new comes out, there are a group of people who push back and want things to stay as they are - history has proven time and time again, that change is inevitable. If something new is the better solution, as people become aware of it and use it - it will become the go to solution. The examples are endless and span multiple disciplines.
Fedora has a long history of supporting new and innovative solutions and toolsets. That is what helps differentiate us as a distribution. We need a tool that will encourage more participation. I believe Discourse will help with this - people will discover new and more efficient ways to do their work and the sun will rise the next day.
On Fri, Oct 19, 2018 at 9:16 AM Gerald B. Cox gbcox@bzb.us wrote:
Again, I believe some are trying to do an apples to apples comparison with Discourse and mailing list technologies. Discourse was build from the ground up with the goal of fostering communication and collaboration. Hyperkitty is a bolt on HTML to mailing list archives. It's good for what it is, but it isn't Discourse - and usage numbers tend to bear that out.
That's an unfair characterization. HyperKitty was designed from the ground up with that goal in mind too. The _sole_ difference is the backend approach. Discourse uses a database system while HyperKitty uses a mail list engine.
You know why the usage numbers bear that out? Because the upgrade to HyperKitty was mishandled and delayed over and over. We were screwed over by the fact that our infrastructure doesn't run on Fedora, so that made it harder to get it working. The initial deployment was very slow and unoptimized. Bugs in the UI remained unfixed in Fedora's installation even though upstream fixed them. I would not be surprised if upstream ignores us because we don't seem to be upgrading.
The development process for HyperKitty basically stalled out because migrations were impossible from Mailman 2 to Mailman 3 for a *very* long time. Fedora somehow did it, and that seemed to have not gone back upstream, so until *very recently*, upstream did not recommend doing mm2 to mm3 upgrades.
That *completely* handicapped adoption of HyperKitty, because HyperKitty requires Mailman 3. What's worse, because it's almost impossible to run on RHEL due to the lack of Python 3 (which continues to anger and frustrate me), Red Hat never migrated their mailing lists. Red Hat's lists are one of the larger installations, and it was a real blow to not have that migrate.
The irony that I can probably get SUSE to deploy Mailman 3 and HyperKitty before Red Hat will is not lost on me.
The fact is that email usage is declining. People are moving away from it and prefer to use other platforms for collaboration. As with many things... when something new comes out, there are a group of people who push back and want things to stay as they are - history has proven time and time again, that change is inevitable. If something new is the better solution, as people become aware of it and use it - it will become the go to solution. The examples are endless and span multiple disciplines.
You know what? I bounce back and forth between HyperKitty and my email client. If all the lists I subscribed to used HyperKitty, I wouldn't be using my email client at all. While I don't generally reply from HK, it's mainly because of bugs that I know are fixed in newer versions.
We have not taken good care of our mail list infrastructure. I don't blame our infra team. I blame the fact our infra runs on RHEL, and RHEL has handicapped us in so many ways because of their own choices. Fedora can't control its own (infrastructure) destiny because we have no power to influence RHEL at all. And that's broken.
Fedora has a long history of supporting new and innovative solutions and toolsets. That is what helps differentiate us as a distribution. We need a tool that will encourage more participation. I believe Discourse will help with this - people will discover new and more efficient ways to do their work and the sun will rise the next day.
And yet, for 15 years, Fedora didn't have a web forum for user support. People have asked for it over the years, and Fedora refused. That's why things like FedoraForum.org exist instead of being part of Fedora itself.
I'm a guy who started with web forums who later used email lists, not the other way around. I vastly prefer forum-style environments. I still don't like Discourse for this kind of stuff, because it's just not designed for handling contextual conversations.
And there are problems with Discourse too: doing partial quoting with attribution is annoying and requires editing the quote to restore that information. In addition, searching for posts and topics becomes exponentially slower as the system handles more content, which is a huge problem if you're trying to find information to cross reference.
It's also bad for archiving, since threads are inherently unstable. Conversation splitting and merging is very awkward (as I've observed in the Snapcraft Discourse). I can keep going, but it doesn't matter, because you're dead set on this anyway.
On Fri, Oct 19, 2018 at 09:43:48AM -0400, Neal Gompa wrote:
That *completely* handicapped adoption of HyperKitty, because HyperKitty requires Mailman 3. What's worse, because it's almost impossible to run on RHEL due to the lack of Python 3 (which continues to anger and frustrate me), Red Hat never migrated their mailing lists. Red Hat's lists are one of the larger installations, and it was a real blow to not have that migrate.
I don't know why Red Hat's mailman impl isn't upgraded, but it is not blocked by lack of Python 3 on RHEL.
Red Hat Software Collections have been providing Python 3.x versions that run on RHEL since ~2013. They exist as add-on yum repos, rather than part of the base since they have a different support lifecycle and to avoid interfering with the system python binary which is depended on by many other apps/tools
https://access.redhat.com/solutions/472793 https://access.redhat.com/support/policy/updates/rhscl
Regards, Daniel
Le vendredi 19 octobre 2018 à 14:55 +0100, Daniel P. Berrangé a écrit :
I don't know why Red Hat's mailman impl isn't upgraded, but it is not blocked by lack of Python 3 on RHEL.
Red Hat Software Collections have been providing Python 3.x versions that run on RHEL since ~2013. They exist as add-on yum repos,
Well that's another thing 200% broken in the current Fedora RHEL universe (and I speak both with my Fedora contributor hat, and as someone who was tasked with procuring RHEL systems in a fortune xxx company not so long ago).
The whole "but it's in an optional RHEL repo" just kills *any* form RHEL/Fedora symbiosis.
No one but RH knows what ends up in what optional repo for what reason, they're ignored by Centos, it's completely impossible to push anything a tad complex from Fedora to EPEL because it will depend on things RH stashed away in an optional repo centos does not rebuild, and you're forbidden to put a copy in EPEL because it may collide with the optional repos, and so on.
That's how we end up with the hilarious situation where the Go EPEL6 stack is both newer and more complete than the Go EPEL7 stack because RH vacuumed some Go packages in an optional EL7 repo and now it's impossible to do anything Go-related in EPEL7.
I'm sure the RH marketoïds *love* the optional repos, it's segmentation market 101, but concretely? They're the kiss of death for anything in Fedora that has enterprise applications, because almost no one is going to bother contributing things in Fedora, that he needs enterprise-side, if the result has zero chance of ending up in EPEL.
The end result is that no one but RH contributes to EL optional repos, and no one who is working on RH EL7 repos has the slightest interest in integrating with Fedora since they do not see any stream of EPEL Fedora contributions.
Or, you end up deploying enterprise systems with your own private rebuild of Fedora packages for EL, and you know what? At this point the bean counters just ask “why are we paying $$$ to RH again, I see you spend your time rebuilding Fedora packages, can't you use Debian if the EL part of RHEL is useless?”
Regards,
And if this conversation were in Discourse, we could simply move it to a new topic.... ;-)
On Fri, Oct 19, 2018 at 7:21 AM Nicolas Mailhot nicolas.mailhot@laposte.net wrote:
Le vendredi 19 octobre 2018 à 14:55 +0100, Daniel P. Berrangé a écrit :
I don't know why Red Hat's mailman impl isn't upgraded, but it is not blocked by lack of Python 3 on RHEL.
Red Hat Software Collections have been providing Python 3.x versions that run on RHEL since ~2013. They exist as add-on yum repos,
Well that's another thing 200% broken in the current Fedora RHEL universe (and I speak both with my Fedora contributor hat, and as someone who was tasked with procuring RHEL systems in a fortune xxx company not so long ago).
The whole "but it's in an optional RHEL repo" just kills *any* form RHEL/Fedora symbiosis.
No one but RH knows what ends up in what optional repo for what reason, they're ignored by Centos, it's completely impossible to push anything a tad complex from Fedora to EPEL because it will depend on things RH stashed away in an optional repo centos does not rebuild, and you're forbidden to put a copy in EPEL because it may collide with the optional repos, and so on.
That's how we end up with the hilarious situation where the Go EPEL6 stack is both newer and more complete than the Go EPEL7 stack because RH vacuumed some Go packages in an optional EL7 repo and now it's impossible to do anything Go-related in EPEL7.
I'm sure the RH marketoïds *love* the optional repos, it's segmentation market 101, but concretely? They're the kiss of death for anything in Fedora that has enterprise applications, because almost no one is going to bother contributing things in Fedora, that he needs enterprise-side, if the result has zero chance of ending up in EPEL.
The end result is that no one but RH contributes to EL optional repos, and no one who is working on RH EL7 repos has the slightest interest in integrating with Fedora since they do not see any stream of EPEL Fedora contributions.
Or, you end up deploying enterprise systems with your own private rebuild of Fedora packages for EL, and you know what? At this point the bean counters just ask “why are we paying $$$ to RH again, I see you spend your time rebuilding Fedora packages, can't you use Debian if the EL part of RHEL is useless?”
Regards,
-- Nicolas Mailhot _______________________________________________ devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-leave@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org
Le vendredi 19 octobre 2018 à 07:28 -0700, Gerald B. Cox a écrit :
And if this conversation were in Discourse, we could simply move it to a new topic.... ;-)
And if it where is discourse I would’t participate in it.
Basically, as others said, not interested in shiny tech that has no notion of interop, has a single implementation, and a shelf life of a couple years.
There's a reason every single web site out there quietly drops the shiny things bit to fall back on email and sms as soon as it manipulates anything $$$ related. email and sms are an ugly but reliable and standard way to reach anyone. The native web forums are just ephemeral eye-cather mutually incompatible things designed make you read ads (as Mairin explained better than me).
Now, it it were packaged in Fedora, and deployed on Fedora infra, without any magic call to a third party website, that would be something else. Wouldn't change the probability upstream would give up on it as soon as it was not shiny and cool anymore, but at least I and Fedora would not totally depend on them for our data.
On Fri, Oct 19, 2018 at 7:43 AM Nicolas Mailhot nicolas.mailhot@laposte.net wrote:
Le vendredi 19 octobre 2018 à 07:28 -0700, Gerald B. Cox a écrit :
And if this conversation were in Discourse, we could simply move it to a new topic.... ;-)
And if it where is discourse I would’t participate in it.
Basically, as others said, not interested in shiny tech that has no notion of interop, has a single implementation, and a shelf life of a couple years.
There's a reason every single web site out there quietly drops the shiny things bit to fall back on email and sms as soon as it manipulates anything $$$ related. email and sms are an ugly but reliable and standard way to reach anyone. The native web forums are just ephemeral eye-cather mutually incompatible things designed make you read ads (as Mairin explained better than me).
Now, it it were packaged in Fedora, and deployed on Fedora infra, without any magic call to a third party website, that would be something else. Wouldn't change the probability upstream would give up on it as soon as it was not shiny and cool anymore, but at least I and Fedora would not totally depend on them for our data.
You really should try it, you might like it. BTW, there are no ads in the Fedora Discourse instance, so not sure what you are talking about there. As far as email is concerned the trends are clear... just do a web search.
Le vendredi 19 octobre 2018 à 07:54 -0700, Gerald B. Cox a écrit :
On Fri, Oct 19, 2018 at 7:43 AM Nicolas Mailhot <
You really should try it, you might like it. BTW, there are no ads in the Fedora Discourse instance, so not sure what you are talking about there. As far as email is concerned the trends are clear... just do a web search.
Ah yes, the advertising industry, that created the web forum bubble and its social media derivative, publishes report after report they're the thing and no one is using mail any more, so you can safely invest even more money advertising on web forums and social media.
In the meanwhile? Websites still use mail. People who do work still use mail. Anyone who needs to publish anything he wants to be able to find some years later still uses mail, or simple websites. Even so-called millenials move to mail, as soon as they need to fill taxes, or anything else that requires tracking things for years, and they notice the hard way the transient nature of web forums and social media (transient *except* for the extracted personal data).
On Fri, 2018-10-19 at 07:28 -0700, Gerald B. Cox wrote:
And if this conversation were in Discourse, we could simply move it to a new topic.... ;-)
Some people see it as history rewriting ... one of the reasons I like email is that *you* can't change stuff after the fact, because *I* have it archived.
Simo.
On Fri, Oct 19, 2018 at 6:45 AM Neal Gompa ngompa13@gmail.com wrote:
On Fri, Oct 19, 2018 at 9:16 AM Gerald B. Cox gbcox@bzb.us wrote:
Again, I believe some are trying to do an apples to apples comparison
with Discourse and mailing list technologies. Discourse was build from the ground up with the goal of fostering communication and collaboration. Hyperkitty is a bolt on HTML to mailing list archives. It's good for what it is, but it isn't Discourse - and usage numbers tend to bear that out.
That's an unfair characterization. HyperKitty was designed from the ground up with that goal in mind too. The _sole_ difference is the backend approach. Discourse uses a database system while HyperKitty uses a mail list engine.
if you think the "sole" difference between HyperKitty and Discourse is the
backend approach you're not looking very hard. It's quite apparent just by looking at it. If yperKitty's design goals are the exact same as Discourse they hid it pretty well in their online documentation.
You know why the usage numbers bear that out? Because the upgrade to HyperKitty was mishandled and delayed over and over. We were screwed over by the fact that our infrastructure doesn't run on Fedora, so that made it harder to get it working. The initial deployment was very slow and unoptimized. Bugs in the UI remained unfixed in Fedora's installation even though upstream fixed them. I would not be surprised if upstream ignores us because we don't seem to be upgrading.
I don't agree with that being the reason - I believe it is the design approach and goals - but even if that were true - that ship has sailed.
The development process for HyperKitty basically stalled out because migrations were impossible from Mailman 2 to Mailman 3 for a *very* long time. Fedora somehow did it, and that seemed to have not gone back upstream, so until *very recently*, upstream did not recommend doing mm2 to mm3 upgrades.
This thread isn't about making excuses for Hyperkitty - it's about Discourse.
On 19/10/18 07:03 -0700, Gerald B. Cox wrote:
On Fri, Oct 19, 2018 at 6:45 AM Neal Gompa ngompa13@gmail.com wrote:
On Fri, Oct 19, 2018 at 9:16 AM Gerald B. Cox gbcox@bzb.us wrote:
Again, I believe some are trying to do an apples to apples comparison
with Discourse and mailing list technologies. Discourse was build from the ground up with the goal of fostering communication and collaboration. Hyperkitty is a bolt on HTML to mailing list archives. It's good for what it is, but it isn't Discourse - and usage numbers tend to bear that out.
That's an unfair characterization. HyperKitty was designed from the ground up with that goal in mind too. The _sole_ difference is the backend approach. Discourse uses a database system while HyperKitty uses a mail list engine.
if you think the "sole" difference between HyperKitty and Discourse is the
backend approach you're not looking very hard. It's quite apparent just by looking at it. If yperKitty's design goals are the exact same as Discourse they hid it pretty well in their online documentation.
The one benefit that Discourse might offer to me is that Gerald's replies might use quoting properly :-P
(Yes, I know this is a problem with the text/plain part generated by Gmail's web UI, which can be worked around by just adding a blank line between the quoted text and the reply).
On Fri, Oct 19, 2018 at 09:43:48AM -0400, Neal Gompa wrote:
It's also bad for archiving, since threads are inherently unstable. Conversation splitting and merging is very awkward (as I've observed in the Snapcraft Discourse). I can keep going, but it doesn't matter, because you're dead set on this anyway.
I ... don't know how to engage constructively with this accusation, because it it seems to come from absolutely nowhere. Yes, we're *definitely* trying out Discourse. That's not a conspiracy — it's live! We're also trying out HyperKitty. It's live too! We try stuff!
The rest of everything — development resources, infrastructure, whether there is widespread adoption — that's not something arguing over will change. When we can *see* something is not working, we can't just complain that those are unfair. we should try something else.
But the comments from several people implying some sort of fait accompli here, that some backroom decision has been made, that we're somehow trying to destroy communication.... seriously?
On Fri, Oct 19, 2018 at 10:10:22AM -0400, Matthew Miller wrote:
I ... don't know how to engage constructively with this accusation, because it it seems to come from absolutely nowhere. Yes, we're *definitely* trying out Discourse. That's not a conspiracy — it's live! We're also trying out HyperKitty. It's live too! We try stuff!
Can you please turn on the ability to create new threads via email/mailing list mode in Discourse?
Thank you.
On Fri, Oct 19, 2018 at 10:11 AM Matthew Miller mattdm@fedoraproject.org wrote:
On Fri, Oct 19, 2018 at 09:43:48AM -0400, Neal Gompa wrote:
It's also bad for archiving, since threads are inherently unstable. Conversation splitting and merging is very awkward (as I've observed in the Snapcraft Discourse). I can keep going, but it doesn't matter, because you're dead set on this anyway.
I ... don't know how to engage constructively with this accusation, because it it seems to come from absolutely nowhere. Yes, we're *definitely* trying out Discourse. That's not a conspiracy — it's live! We're also trying out HyperKitty. It's live too! We try stuff!
Do not misunderstand me, I don't mind trying out Discourse for certain types of conversations if we want to, and clearly some communities in Fedora do.
But I feel like we've done a bad job with giving HyperKitty a fair shake. The same thing happened with Pagure too, but someone had the wherewithal to push everything forward enough to make it a success, and even now I'll say that Pagure is a nice environment. Yeah, Pagure is still missing things I wish it had, but it's evolving and growing. It took a few years, but it happened.
I know that the big promotion push for HyperKitty was supposed to come with Hubs, but somehow that project died before it could go anywhere, so...
I'm happy to help with infra related stuff, but I'm just not equipped to do so. Only a few people are. :(
The rest of everything — development resources, infrastructure, whether there is widespread adoption — that's not something arguing over will change. When we can *see* something is not working, we can't just complain that those are unfair. we should try something else.
The reason I'm arguing about it now is because it doesn't seem to be fixable from the background. We just keep hacking around it over and over.
I don't like saying this, but what it comes down to is that our relationship with RHEL has evolved into a one-sided affair. I wish someone who is empowered to do something about it would, but the rest of us can't.
Frankly, I suspect you're the only person who could maybe do anything about it, and I'm not certain you could do anything about it.
But the comments from several people implying some sort of fait accompli here, that some backroom decision has been made, that we're somehow trying to destroy communication.... seriously?
I'm not talking about you in the Fedora sense. I'm talking about Gerald and his saying "we must move everything to Discourse".
On Fri, Oct 19, 2018 at 7:41 AM Neal Gompa ngompa13@gmail.com wrote:
I'm not talking about you in the Fedora sense. I'm talking about Gerald and his saying "we must move everything to Discourse".
Oh really... I said that... perhaps you should take 5 seconds and read the subject of the thread.
As far as Hyperkitty is concerned you need to take a few minutes and re-read Matt's response.
But again, Hyperkitty is not the point of this thread.
On Fri, Oct 19, 2018 at 07:49:41AM -0700, Gerald B. Cox wrote:
Oh really... I said that... perhaps you should take 5 seconds and read the subject of the thread.
Hey, let's please keep this friendly.
On Fri, Oct 19, 2018 at 10:40:18AM -0400, Neal Gompa wrote:
I don't like saying this, but what it comes down to is that our relationship with RHEL has evolved into a one-sided affair. I wish someone who is empowered to do something about it would, but the rest of us can't.
Frankly, I suspect you're the only person who could maybe do anything about it, and I'm not certain you could do anything about it.
Time to pull back the curtain a little bit, I guess. :)
I think the issue isn't really RHEL-Fedora, but Red Hat-Fedora. Red Hat isn't a large company in the sense of Oracle or SAP or whatever, let alone Microsoft or Apple, but it's much bigger than it used to be, and much bigger than RHEL.
Most of Red Hat's _financial_ engagement in Fedora comes from two places: Platform (i.e., RHEL) and associated groups like QE, and from the CTO's office via OSAS. In the first bucket, that's hardware and infrastructure, the people paid to be on the community infra team (and work in Fedora Infrastructure and Rel-Eng), and me. In the second, the CTO's office, that's Bex — and now Sanja on CoreOS and Silverblue — and also our community budget.
On the Platform side, it's easy (and, increasingly so, really!) to get Red Hat interested Fedora technology that has a direct connection to improving, well, the platform. This is the answer to why funding for Pagure and not HyperKitty or Hubs. The src.fedoraproject.org thing and what we're working on with shared dist-git with CentOS... easy dots to connect. This part of Red Hat *cares* about Fedora as a succcessful community, but also has to justify spending, and as the company overall invests in OpenShift in the Enterprise, there's not a lot of extra.
Meanwhile, the other side of the coin, over in the CTO's office — Fedora's community budget and staffing is a significant chunk already. (The Discourse experiment funding is coming through Sanja and not the Fedora community budget, FWIW.)
As much as I think the CTO's office *should*, they don't have a group of programmers available to work on community open source tooling. I definitely am pushing as much as I can for more of that kind of investment, and ... maybe some things will bear fruit. I have to say, though (since it's super-relevant to the discussion here) one of the very first questions I get every time is: "Why does Fedora have so much of its own stuff when there are open source alternatives? What's with the huge NIH complex?" I do a lot of shoving rocks uphill on that one!
But if we want this to be a two-way balanced relationship, it can't be all "Red Hat isn't spending enough money on Fedora's non-engineering needs!". What else do you (not just Neal — take this as an open question!) think we should do differently from a Red Hat side?
On Fri, 19 Oct 2018, Matthew Miller wrote:
Neal Gompa: because you're dead set on this anyway.
Matt Miller: I ... don't know how to engage constructively with this accusation, because it it seems to come from absolutely nowhere. Yes, we're *definitely* trying
This seems very tone deaf and lacking in introspection, Matt
perhaps by reading the subject line you chose to start this thread with
No mention of trying out, no mention of a trial. Just the usual 'this is a done deal' announcement on an extremely high volume mailing list
I had trialled Discourse six months ago, leaving it on for days at a time. The web client had a leak down in it somewhere, My baseline load average went up and I ended up with a frozen web browser once all free ram and swap was exhausted. Migrate or not as you wish, it will simply be another place where Fedoraproject went off the tracks, chasing a mythical low involvement 'tire kicker' instead of supporting developers using mature tools and technologies (mailing lists, bugzilla, procmail sorting, mutt or alpine)
The non-migratation capabilities of early MM3 efforts have been detailed earlier in this thread. Hyperkitty pretends a 33 character hash conveys more information than the MM2 model for pipermail, or say: YYYYMMDD-NNNN, but no-one was willing to critique a bad choice
Pagure is great and all, but I recall filing a bug ... somewhere ... pagure, git, how knows ... about a cross-site inter-panel info leak, but there is no SPOT -- single point of truth -- in Fedora's hall of mirrors to find it later ad hoc. That does not happen with Bugzilla as I have a portfolio of several dozen custom searches that would reveal it to me with a couple clicks
But charging off to a new shiny hill seems more important to Fedoraproject
** shrug **
-- Russ herrold
Once upon a time, R P Herrold herrold@owlriver.com said:
This seems very tone deaf and lacking in introspection, Matt
perhaps by reading the subject line you chose to start this thread with
Matt didn't choose that - that subject was set by Gerald B. Cox.
On Fri, Oct 19, 2018 at 8:31 AM Chris Adams linux@cmadams.net wrote:
Once upon a time, R P Herrold herrold@owlriver.com said:
This seems very tone deaf and lacking in introspection, Matt
perhaps by reading the subject line you chose to start this thread with
Matt didn't choose that - that subject was set by Gerald B. Cox.
As I previously mentioned with all the top-posting, excerpts and hyperbole
interjected by others people get lost and run with mis-quotes - perhaps Discourse could help with that. ;-)
Software is a tool for me. I don't get emotionally attached to it - as some people apparently are. It's a bit telling that many people seem to be afraid that Discourse will be a success.
On Fri, Oct 19, 2018 at 12:00 PM Gerald B. Cox gbcox@bzb.us wrote:
On Fri, Oct 19, 2018 at 8:31 AM Chris Adams linux@cmadams.net wrote:
Once upon a time, R P Herrold herrold@owlriver.com said:
This seems very tone deaf and lacking in introspection, Matt
perhaps by reading the subject line you chose to start this thread with
Matt didn't choose that - that subject was set by Gerald B. Cox.
As I previously mentioned with all the top-posting, excerpts and hyperbole interjected by others people get lost and run with mis-quotes - perhaps Discourse could help with that. ;-)
Software is a tool for me. I don't get emotionally attached to it - as some people apparently are. It's a bit telling that many people seem to be afraid that Discourse will be a success.
I'm more afraid that it'll be a success with casualties. In other words, it'll be a failure but not look like one at a glance. Driving people away and making it harder to keep track of topics of import is going to necessarily constrain how much people are able and willing to do. It doesn't get simpler than that. And I have *not* seen a Discourse instance be successful in that with large teams, much less large groups like the development groups within Fedora.
On Fri, Oct 19, 2018 at 12:19:43PM -0400, Neal Gompa wrote:
I'm more afraid that it'll be a success with casualties. In other words, it'll be a failure but not look like one at a glance. Driving people away and making it harder to keep track of topics of import is going to necessarily constrain how much people are able and willing to do. It doesn't get simpler than that. And I have *not* seen a Discourse instance be successful in that with large teams, much less large groups like the development groups within Fedora.
Yeah -- this is completely reasonable!
On Fri, Oct 19, 2018, 18:30 Neal Gompa ngompa13@gmail.com wrote:
On Fri, Oct 19, 2018 at 12:00 PM Gerald B. Cox gbcox@bzb.us wrote:
On Fri, Oct 19, 2018 at 8:31 AM Chris Adams linux@cmadams.net wrote:
Once upon a time, R P Herrold herrold@owlriver.com said:
This seems very tone deaf and lacking in introspection, Matt
perhaps by reading the subject line you chose to start this thread with
Matt didn't choose that - that subject was set by Gerald B. Cox.
As I previously mentioned with all the top-posting, excerpts and
hyperbole interjected by others people
get lost and run with mis-quotes - perhaps Discourse could help with
that. ;-)
Software is a tool for me. I don't get emotionally attached to it - as
some people apparently are. It's a bit telling that
many people seem to be afraid that Discourse will be a success.
I'm more afraid that it'll be a success with casualties. In other words, it'll be a failure but not look like one at a glance. Driving people away and making it harder to keep track of topics of import is going to necessarily constrain how much people are able and willing to do. It doesn't get simpler than that. And I have *not* seen a Discourse instance be successful in that with large teams, much less large groups like the development groups within Fedora.
Actually, Rust ecosystem is using discourse quite well. Both for development discussions and for user discussions.
On Fri, 19 Oct 2018 08:58:57 -0700 "Gerald B. Cox" gbcox@bzb.us wrote:
On Fri, Oct 19, 2018 at 8:31 AM Chris Adams linux@cmadams.net wrote:
Once upon a time, R P Herrold herrold@owlriver.com said:
This seems very tone deaf and lacking in introspection, Matt
perhaps by reading the subject line you chose to start this thread with
Matt didn't choose that - that subject was set by Gerald B. Cox.
Yeah, it is kind of confrontational; there's a big difference between 'Should Fedora replace mailing lists with Discourse?' and 'Fedora should replace mailing lists with Discourse.'
One implies a community discourse, the other implies a decision already made.
As I previously mentioned with all the top-posting, excerpts and hyperbole interjected by others people get lost and run with mis-quotes - perhaps Discourse could help with that. ;-)
Software is a tool for me. I don't get emotionally attached to it - as some people apparently are. It's a bit telling that many people seem to be afraid that Discourse will be a success.
It isn't free for people who have already a good working system to adopt Discourse. They have to invest time and effort in learning, adapting, and helping fix the new system. What's in it for them? I've had this happen to me in the past, but I was paid by my employer for the time and effort.
You aren't really selling Discourse, so much as trying to bludgeon people into going along with your opinion. Anyone who's been around the block a few times knows that there are lots of fads with fabulous press that don't pan out.
So, you are really gung-ho for Discourse. What is your personal experience of using it on a project, before and after? Did you actually see more engagement? Did the process of development run smoother than it had before? Did new contributors join the project you were working on? Did experienced contributors keep contributing? Did it help build community spirit? Did it make *your* life easier?
On Fri, Oct 19, 2018 at 3:58 PM stan stanl-fedorauser@vfemail.net wrote:
On Fri, 19 Oct 2018 08:58:57 -0700 "Gerald B. Cox" gbcox@bzb.us wrote:
Software is a tool for me. I don't get emotionally attached to it - as some people apparently are. It's a bit telling that many people seem to be afraid that Discourse will be a success.
It isn't free for people who have already a good working system to adopt Discourse. They have to invest time and effort in learning, adapting, and helping fix the new system. What's in it for them? I've had this happen to me in the past, but I was paid by my employer for the time and effort.
You aren't really selling Discourse, so much as trying to bludgeon people into going along with your opinion. Anyone who's been around the block a few times knows that there are lots of fads with fabulous press that don't pan out.
Indeed-- this thread feels like it has deteriorated into pro-Discourse and pro-email people sniping at each other. I am not sure how useful continuing to say things like "And if this conversation were in Discourse..." or "perhaps Discourse could help with that" really is when trying to convince people who don't agree with you.
I don't know that I feel strongly one way or the other about Discourse vs. email. What I do feel strongly about, as I wrote much earlier in the thread, is that any plan to move something as important to the project as email to a new system really needs to be carefully considered. And part of that careful consideration needs to be an honest assessment of what the negative consequences are likely to be. I suspect the negative consequences (as many have already said) are that it won't be possible to use Discourse as a drop-in replacement for these mailing lists, which will break the workflows of numerous current contributors, causing them to become less involved in discussions and perhaps the project altogether.
It is a bit disheartening that some of those advocating the change seem unwilling to acknowledge this, or have dismissed it as people who aren't willing to move with the times, or as just "hyperbole".
Ben Rosser
stan writes:
So, you are really gung-ho for Discourse.
IMO, that's not nasty, but it wasn't necessary and could be taken badly in context. Just, "as a proponent, I'd like to ask you" is good when things are getting heated.
Your questions are important[1], and I'd like to gloss them:
What is your personal experience of using it on a project, before and after?
It would be helpful to specify the project. It's also really important to be specific about what was merely comfortable, and what directly contributed to solving problems of developing the project's software.
Did you actually see more engagement?
And by whom? Developers? Users with requirements? Randos who engaged in controversy but in the end contributed neither code nor to refining requirements, specifications, or design?
Did you see *less* engagement (especially of the rando type)?
Did the process of development run smoother than it had before?
And if so, what aspects of the software contributed to smoothness, and how?
Which parts of the process (requirements, specification, design, coding, testing, documentation) benefited specifically?
Were there any aspects that were less smooth?
Did new contributors join the project you were working on?
Did experienced contributors keep contributing?
Did it help build community spirit?
How did "channel culture" change? Were people more or less polite? Were experienced participants more or less likely to help on-boarding newcomers? Mentor new contributors?
Did it make *your* life easier?
And for others? How might these things generalize to this list?
Steve GNU Mailman Project
Footnotes: [1] To Fedora in deciding to move or not, but also to Mailman to learn requirements to improve our product. "required disclosure"
On 10/19/18 8:58 AM, Gerald B. Cox wrote:
Software is a tool for me. I don't get emotionally attached to it - as some people apparently are. It's a bit telling that many people seem to be afraid that Discourse will be a success.
I'm not emotionally attached to it, but I am somewhat afraid that Discourse will be a success. Forum-type communication websites do not work for me personally, so if this mailing list moved to Discourse, my involvement could drop drastically depending on how the email integration works. For example, github and bugzilla work because they have full email messages. I don't have to go to the website to get the rest of the message. I do have to go there in order to reply, but that's mostly ok. However, most forums just send you an email saying there's something new on the topic with maybe a truncated bit of the message and I have to go to the website to figure out what's happened. That does not work and I have very little participation on those.
I find that email works really well. It's easy to search, I control the labelling, organizing and filtering, I can back it up, I can use it offline, minimal data usage on my phone, and so on. Features that just aren't going to work on a website-based system.
On Fri, 2018-10-19 at 13:05 -0700, Samuel Sieb wrote:
For example, github and bugzilla work because they have full email messages. I don't have to go to the website to get the rest of the message.
You actually can reply to the e-mails from GitHub (not sure about Bugzilla). They do thread weirdly when you do that to a review comment vs. replying to the review comment on the HTML, so it's a bit janky, but it does work.
Gerald B. Cox wrote:
On Fri, Oct 19, 2018 at 8:31 AM Chris Adams linux@cmadams.net wrote:
Once upon a time, R P Herrold herrold@owlriver.com said:
perhaps by reading the subject line you chose to start this thread with
Matt didn't choose that - that subject was set by Gerald B. Cox.
As I previously mentioned with all the top-posting, excerpts and hyperbole
interjected by others people get lost and run with mis-quotes - perhaps Discourse could help with that.
Moments before I read this I checked for myself who had changed the subject line. It took me about five seconds to scroll up in the tree view, find the point where the subject changed, and read the name on the same line. Having just done that so easily, I can really appreciate how utterly misdirected your remark is.
If it's difficult for you to check who wrote a subject line, then your problem is that you're trying to read a mailing list through a user agent that isn't up to the task. And, speaking of misquotes, the misattributed line that can be seen above ("As I previously ...") is a visible indication that you're not using one of the better Mail User Agents.
Humans will always misremember things, and not bother to check because they think they remember. No technology is going to help with that until we get direct neural interfaces.
Björn Persson
On Fri, 19 Oct 2018, Chris Adams wrote:
Once upon a time, R P Herrold herrold@owlriver.com said:
This seems very tone deaf and lacking in introspection, Matt
perhaps by reading the subject line you chose to start this thread with
Matt didn't choose that - that subject was set by Gerald B. Cox.
If you say so (and I have no reason to doubt, but cannot confirm, so: sorry for the mis-attribution, Matt) ... at the URL in your email message headers is a link to: https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/...
Perma-link: https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/...
and at that page, the link titled ' Back to the thread ' points into a wholly different discussion
https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/...
One more reason to dislike KyperKitty hashes over pipermail
-- Russ herrold
On 10/19/18 6:43 AM, Neal Gompa wrote:
On Fri, Oct 19, 2018 at 9:16 AM Gerald B. Cox gbcox@bzb.us wrote:
Again, I believe some are trying to do an apples to apples comparison with Discourse and mailing list technologies. Discourse was build from the ground up with the goal of fostering communication and collaboration. Hyperkitty is a bolt on HTML to mailing list archives. It's good for what it is, but it isn't Discourse - and usage numbers tend to bear that out.
That's an unfair characterization. HyperKitty was designed from the ground up with that goal in mind too. The _sole_ difference is the backend approach. Discourse uses a database system while HyperKitty uses a mail list engine.
You know why the usage numbers bear that out? Because the upgrade to HyperKitty was mishandled and delayed over and over. We were screwed over by the fact that our infrastructure doesn't run on Fedora, so that made it harder to get it working. The initial deployment was very slow and unoptimized. Bugs in the UI remained unfixed in Fedora's installation even though upstream fixed them. I would not be surprised if upstream ignores us because we don't seem to be upgrading.
Huh, you do realize that things take as long as they take, and there's no magic wand for 'it's magically done'. mailman3 was a massive undertaking with a very small group of developers, many of whom were wanting things to be really done before releasing them.
Much of our infrastructure does run on Fedora, and mailman3 could too if we needed it to. It just wasn't the decision at the time. Bugs remain unfixed because there's not too much bugfixing going on. I see two commits in this month, 2 last month, 1 the month before...
You can always ask why we aren't upgrading. In this case it's because we are moving stuff to python36 from 34. If these fixes are urgent let us know and we can re-evaluate and try and get things faster. I was under the impression that the fixes were pretty minor.
...snip...
We have not taken good care of our mail list infrastructure. I don't blame our infra team. I blame the fact our infra runs on RHEL, and RHEL has handicapped us in so many ways because of their own choices. Fedora can't control its own (infrastructure) destiny because we have no power to influence RHEL at all. And that's broken.
Thats simply not the case. We run many things on Fedora. We do so where it makes sense and RHEL where that makes sense, and increasingly OpenShift when that makes sense. ;)
I attempted to use the discourse rss feeds this last week, and they were... not great. I still need to try mailing list mode before speaking much more on it.
kevin
Kevin Fenzi writes:
On 10/19/18 6:43 AM, Neal Gompa wrote:
You know why the usage numbers bear that out? Because the upgrade to HyperKitty was mishandled and delayed over and over. We were screwed over by the fact that our infrastructure doesn't run on Fedora, so that made it harder to get it working. The initial deployment was very slow and unoptimized. Bugs in the UI remained unfixed in Fedora's installation even though upstream fixed them. I would not be surprised if upstream ignores us because we don't seem to be upgrading.
Huh, you do realize that things take as long as they take, and there's no magic wand for 'it's magically done'. mailman3 was a massive undertaking with a very small group of developers, many of whom were wanting things to be really done before releasing them.
Yeah, as a core Mailman developer I was really disappointed when the whole crew of Fedora/RH-supported developers just disappeared without leaving behind successors. I understand why that happens, but I wish I'd known they were able to participate only as long as they were assigned to it. Unfortunately, it's clear from the current installation supporting this list that no, they didn't get things "really done", or at least they were restricted to their direct relationship with HyperKitty -- in Fedora Postorius, even the explanatory blurbs in the user config screens are frequently incomplete (eg, lack information about current, default, and inherited values) and at least one is outright confusing (the semantics of the associated variable are the reverse of the option name). Again, I understand why such things happen, but *somebody* is going to have to commit to better care and feeding of the channel, whatever software is supporting it.
We at Mailman are very happy to help. We're also a small crew of part-timers, so that's going to be limited, but at least we're aware that Fedora's is one of the most heavily-used Mailman 3 installations, so we have a strong interest in it working well! Máirín's mail to our dev list got immediate and enthusiastic reaction. But we can't help if you have no support for upgrading to upstream current release; that's not our job (unless paid, and I'm not even sure that would break our developers loose from other responsibilities and core work).
I'm not sure you can count on such support from Discourse, but I have said more about that elsewhere in the thread, so I won't belabor it here.
You can always ask why we aren't upgrading. In this case it's because we are moving stuff to python36 from 34. If these fixes are urgent let us know and we can re-evaluate and try and get things faster. I was under the impression that the fixes were pretty minor.
HyperKitty is a fairly complex piece of software. I never did make head or tail of it (most of my time is devoted to core Mailman, especially email security such as DMARC and crypto), and there's nobody associated with the Mailman project to teach me about it anymore. To me it's not surprising that the only things people are willing to touch are minor. And even the original developers are unlikely to be familiar with the current state of upstream, as upstream has changes, some significant I think.
Steve
On 10/20/18 9:09 AM, Stephen J. Turnbull wrote:
Kevin Fenzi writes:
On 10/19/18 6:43 AM, Neal Gompa wrote:
You know why the usage numbers bear that out? Because the upgrade to HyperKitty was mishandled and delayed over and over. We were screwed over by the fact that our infrastructure doesn't run on Fedora, so that made it harder to get it working. The initial deployment was very slow and unoptimized. Bugs in the UI remained unfixed in Fedora's installation even though upstream fixed them. I would not be surprised if upstream ignores us because we don't seem to be upgrading.
Huh, you do realize that things take as long as they take, and there's no magic wand for 'it's magically done'. mailman3 was a massive undertaking with a very small group of developers, many of whom were wanting things to be really done before releasing them.
Yeah, as a core Mailman developer I was really disappointed when the whole crew of Fedora/RH-supported developers just disappeared without leaving behind successors. I understand why that happens, but I wish I'd known they were able to participate only as long as they were assigned to it.
Huh. The only person I know of from Fedora at least that was working on it was abompard. While he's working on other things now, as far as I know he's still working on mailman3/hyperkitty as time permits.
Unfortunately, it's clear from the current installation supporting this list that no, they didn't get things "really done", or at least they were restricted to their direct relationship with HyperKitty -- in Fedora Postorius, even the explanatory blurbs in the user config screens are frequently incomplete (eg, lack information about current, default, and inherited values) and at least one is outright confusing (the semantics of the associated variable are the reverse of the option name). Again, I understand why such things happen, but *somebody* is going to have to commit to better care and feeding of the channel, whatever software is supporting it.
Sure, but it could be anyone who wants to fix those things. By "really done" I didn't mean that the software was bug free and implemented 100% of it's intended uses. I just meant that they want to ship a usable product, of course there are still bugs or things that in hindshight would have been good to fix before release.
We at Mailman are very happy to help. We're also a small crew of part-timers, so that's going to be limited, but at least we're aware that Fedora's is one of the most heavily-used Mailman 3 installations, so we have a strong interest in it working well! Máirín's mail to our dev list got immediate and enthusiastic reaction. But we can't help if you have no support for upgrading to upstream current release; that's not our job (unless paid, and I'm not even sure that would break our developers loose from other responsibilities and core work).
Sure, and I don't know of anyone who said it was your responsibility. We have a ton of things going on, so we just had this as low priority currently. If it needs to be moved up higher the reasons for that would be great to know/discuss.
I'm not sure you can count on such support from Discourse, but I have said more about that elsewhere in the thread, so I won't belabor it here.
You can always ask why we aren't upgrading. In this case it's because we are moving stuff to python36 from 34. If these fixes are urgent let us know and we can re-evaluate and try and get things faster. I was under the impression that the fixes were pretty minor.
HyperKitty is a fairly complex piece of software. I never did make head or tail of it (most of my time is devoted to core Mailman, especially email security such as DMARC and crypto), and there's nobody associated with the Mailman project to teach me about it anymore. To me it's not surprising that the only things people are willing to touch are minor. And even the original developers are unlikely to be familiar with the current state of upstream, as upstream has changes, some significant I think.
While I know he's got many other things on his plate, I thought Aurelien was still doing upstream work and helping folks. CCing him on this, I could be mistaken.
kevin
Kevin Fenzi writes:
Huh. The only person I know of from Fedora at least that was working on it was abompard. While he's working on other things now, as far as I know he's still working on mailman3/hyperkitty as time permits.
pingu and abadger also contributed. Don't know their exact affiliations or what they're doing for Fedora's installation, but all three have stopped substantial upstream contribution for a couple years now. abompard has a few commits, he and pingu still answer mail but that's about it. abadger we still hang with at PyCons, but he wasn't involved in the later development, and would have to put a lot of effort in to get up to speed. As you say "it could be anyone who wants to", so we're not going to ask any of them to do more than they want to at any given time. They're not core any more.
but *somebody* is going to have to commit to better care and feeding of the channel, whatever software is supporting it.
Sure, but it could be anyone who wants to fix those things.
No, it can't, can it. Fedora is not keeping up with upstream, which means that "anyone who wants to" isn't upgrading on the Fedora system. And that's because they'd need to get permission or effort out of the operators, and apparently that's scarce. The people who manage the machines have to make this commitment, or it goes nowhere.
By "really done" I didn't mean that the software was bug free and implemented 100% of it's intended uses. I just meant that they want to ship a usable product, of course there are still bugs or things that in hindshight would have been good to fix before release.
I accept your definition, but I've never seen "really done" used to mean just "usable" before. ;-)
Sure, and I don't know of anyone who said it was your responsibility.
I didn't say anybody said that; I was pointing out that we are ready to help by accepting and maintaining bugfixes and improvements done by Fedora workers (as before), and to a limited extent to use our own resources to work on bug reports and RFEs. We have a proven record of doing exactly that. Discourse, OTOH, is an unknown quantity in that respect to me, and some of the things that I consider bugs in Discourse (eg, sending BBcode in text/plain parts) appear to be deliberate choices, so I would not be optimistic about getting some of the changes Fedora will want through quickly.
We have a ton of things going on, so we just had this as low priority currently. If it needs to be moved up higher the reasons for that would be great to know/discuss.
I don't know the reasons, but somebody at Fedora does: they're the same reasons that fedoraforum.org exists.
While I know he's got many other things on his plate, I thought Aurelien was still doing upstream work and helping folks. CCing him on this, I could be mistaken.
He's got a spate of commits every summer. But I think that abompard, as well as pingu and abadger, are a moot point here: they focused on core functions and scalability. Máirín is the author of the UI layer and has offered to work on these UX issues. I'm confident that she'll produce results for us! The question is will they get used by Fedora?
Much as I detest forums personally, I consider this an open question for Fedora. The "try it, you'll like it" advocacy from forum proponents worries me, but Fedora is not my thing, Mailman is. You all have to decide.
I will say I've seen this happen in several communities now (Python, Emacs). The communities seem to scale past the capacity of the communication channels. But to me (and the economics of information is what I do for a living), I don't think it's a problem with the channels, it's a PEBKAC. The *people* at the terminals don't have the bandwidth, and they're desperately searching for a way to reduce the time investment. But in my experience, they don't reallocate from fighting with email to more fruitful conversations on the channel (although they may be more polite with more moderation). Instead, they reduce engagement all around. And that's what I don't like about forums: in my experience, they undermine the productivity of the channel at the same time as they increase collegiality. YMMV, of course, but I evidently have a lot of company in this perception. :-(
Steve
On 10/21/18 6:12 PM, Stephen J. Turnbull wrote:
Kevin Fenzi writes:
Huh. The only person I know of from Fedora at least that was working on it was abompard. While he's working on other things now, as far as I know he's still working on mailman3/hyperkitty as time permits.
pingu and abadger also contributed. Don't know their exact affiliations or what they're doing for Fedora's installation, but all three have stopped substantial upstream contribution for a couple
Toshio (abadger) works on ansible now. Pierre (pingou) is still in the Fedora engineering group but is focused on pagure and ci work. I only see 2 commits ever from Toshio and all of pingou's commits were at the very start of the project in 2012.
years now. abompard has a few commits, he and pingu still answer mail but that's about it.
Well, he has 980 commits, much more than anyone else.
abadger we still hang with at PyCons, but he wasn't involved in the later development, and would have to put a lot of effort in to get up to speed. As you say "it could be anyone who wants to", so we're not going to ask any of them to do more than they want to at any given time. They're not core any more.
but *somebody* is going to have to commit to better care and feeding of the channel, whatever software is supporting it.
Sure, but it could be anyone who wants to fix those things.
No, it can't, can it. Fedora is not keeping up with upstream, which means that "anyone who wants to" isn't upgrading on the Fedora system. And that's because they'd need to get permission or effort out of the operators, and apparently that's scarce. The people who manage the machines have to make this commitment, or it goes nowhere.
We intend to keep up with upstream, but we haven't dropped other items to do so urgently. Unless there is some decision to move to something else, we definitely plan to catch back up to upstream when we can.
...snip...
We have a ton of things going on, so we just had this as low priority currently. If it needs to be moved up higher the reasons for that would be great to know/discuss.
I don't know the reasons, but somebody at Fedora does: they're the same reasons that fedoraforum.org exists.
I'm not able to parse this. This very thread is more information that may get us to move it up in priority. Can you rephrase?
While I know he's got many other things on his plate, I thought Aurelien was still doing upstream work and helping folks. CCing him on this, I could be mistaken.
He's got a spate of commits every summer. But I think that abompard, as well as pingu and abadger, are a moot point here: they focused on core functions and scalability. Máirín is the author of the UI layer and has offered to work on these UX issues. I'm confident that she'll produce results for us! The question is will they get used by Fedora?
Sure, if they go in upstream they will be used by us when we update.
Much as I detest forums personally, I consider this an open question for Fedora. The "try it, you'll like it" advocacy from forum proponents worries me, but Fedora is not my thing, Mailman is. You all have to decide.
Sure. I appreciate your input though!
kevin
Kevin Fenzi writes:
On 10/21/18 6:12 PM, Stephen J. Turnbull wrote:
years now. abompard has a few commits, he and pingu still answer mail but that's about it.
Well, he has 980 commits, much more than anyone else.
Sure, but in the last couple of years there are only a couple of dozen commits from all committers.
We intend to keep up with upstream, but we haven't dropped other items to do so urgently.
OK, you're right: it's not urgent; HyperKitty is at 1.2.0a1 in our repo, 1.1.5 is dated October 17 last year, but little has changed and no new features in the UI. Fedora's Postorius is 1.1.2 (upstream 1.2.3), but I doubt that many of the issues I have with Fedora's Postorius are fixed upstream, they're more local installation stuff. There are many more upstream changes in Postorius vs. HyperKitty, but I didn't notice any missing Postorius functionality (except web form signups, and that's deliberate).
Máirín will probably have results in a couple months but we are unlikely to release again until PyCon (in May, I think).
I don't know the reasons, but somebody at Fedora does: they're the same reasons that fedoraforum.org exists.
I'm not able to parse this. This very thread is more information that may get us to move it up in priority. Can you rephrase?
I don't know what "Fedora management" values, or what is needed by the people who are doing the work that devel is designed for (coordinating packages, troubleshooting dependencies, etc). However, there are forums related to Fedora that people have put up (and IIUC fedoraforum is an experiment set up by the Fedora project itself). That means there is some dissatisfaction. But turning that into requirements, or justifying using admin resources, etc to upgrade requires more knowledge about the community than I have.
I haven't found this thread very informative, to be honest. The email advocates say "it works, could be better" (which given the traffic levels I'd have to say is true) which isn't very specific about *what* works and how it works to enhance productivity, while the forum advocates are not at all precise about how a forum is supposed to improve *productivity* of this list. They're mostly focused on certain issues, such as ease of entry that I agree are important for a user support forum, but I don't think they really help the devel workers (that's my personal opinion, not anything I can claim expertise on).
Sure. I appreciate your input though!
Try and stop me! :-)
Seriously, as I said before, Fedora is currently one of our biggest publicly accessible Mailman 3 installations. We want the best for you first, but if "the best for Fedora" can be Mailman 3, we want it to shine!
Steve
On Mon, 22 Oct 2018 at 15:05, Stephen J. Turnbull stephen@xemacs.org wrote:
I'm not able to parse this. This very thread is more information that may get us to move it up in priority. Can you rephrase?
I don't know what "Fedora management" values, or what is needed by the people who are doing the work that devel is designed for (coordinating packages, troubleshooting dependencies, etc). However, there are forums related to Fedora that people have put up (and IIUC fedoraforum is an experiment set up by the Fedora project itself). That means
Fedora Forum is from a long time ago when Fedora was just starting up. At that time things were being set up in a more confederated method where teams would run the products they wanted. The discussions of mailing lists vs web-something have gone on since 2004 at least. And they branch out into 'you should use slack!', 'you should use irc', 'telegram!', 'USEnet!' 'you all are wrong. the Plato system had a superior method', etc etc etc
there is some dissatisfaction. But turning that into requirements, or justifying using admin resources, etc to upgrade requires more knowledge about the community than I have.
I haven't found this thread very informative, to be honest. The email advocates say "it works, could be better" (which given the traffic levels I'd have to say is true) which isn't very specific about *what*
Which pretty sums up every one of these conversations over the last 15 years. I don't think either side really can formulate things any more than a left handed person can say why they are left handed versus right handed.. one method works better for their brain and the other doesn't. Does it work perfectly? No but the other method is like using a tool which always feels wrong. You might be able to learn to use it but it always itches. Trying to explain it beyond that comes down to things like 'well it must be superior because I work better on it!' versus 'well my brain likes to categorize things in this visual method and your brain uses that method.'
And then a bunch of people try to come up with a tool which works for everyone but find out that the larger the net, the bigger the holes. Roll around 2-3 years and we will see another version.
On 10/22/18 11:14 AM, Stephen J. Turnbull wrote:
Kevin Fenzi writes:
On 10/21/18 6:12 PM, Stephen J. Turnbull wrote:
years now. abompard has a few commits, he and pingu still answer mail but that's about it.
Well, he has 980 commits, much more than anyone else.
Sure, but in the last couple of years there are only a couple of dozen commits from all committers.
True. ;(
We intend to keep up with upstream, but we haven't dropped other items to do so urgently.
OK, you're right: it's not urgent; HyperKitty is at 1.2.0a1 in our repo, 1.1.5 is dated October 17 last year, but little has changed and no new features in the UI. Fedora's Postorius is 1.1.2 (upstream 1.2.3), but I doubt that many of the issues I have with Fedora's Postorius are fixed upstream, they're more local installation stuff. There are many more upstream changes in Postorius vs. HyperKitty, but I didn't notice any missing Postorius functionality (except web form signups, and that's deliberate).
Máirín will probably have results in a couple months but we are unlikely to release again until PyCon (in May, I think).
Yeah, but staying closer to upstream is still valuable, so people who commit fixes don't have to wait ages to see them in production.
Another thing I'd like to do at some point, and perhaps we should just do this soon is move our instance to openshift. That would allow for a lot faster deployment schedule and a lot more flexability about what container(s) we use for things as well as scaling, etc. But again thats some work we need to prioritize and get people working on.
I don't know the reasons, but somebody at Fedora does: they're the same reasons that fedoraforum.org exists.
I'm not able to parse this. This very thread is more information that may get us to move it up in priority. Can you rephrase?
I don't know what "Fedora management" values, or what is needed by the people who are doing the work that devel is designed for (coordinating packages, troubleshooting dependencies, etc). However, there are forums related to Fedora that people have put up (and IIUC fedoraforum is an experiment set up by the Fedora project itself). That means there is some dissatisfaction. But turning that into requirements, or justifying using admin resources, etc to upgrade requires more knowledge about the community than I have.
Fedoraforum is not now, nor has it ever been run by the Fedora project. It's always been a 3rd party site that runs how they like. I used to be a moderator there, but I've not had time to even login there in ages. (perhaps proving that polling sites always ends up slipping down your list until you no longer remember to do it).
I haven't found this thread very informative, to be honest. The email advocates say "it works, could be better" (which given the traffic levels I'd have to say is true) which isn't very specific about *what* works and how it works to enhance productivity, while the forum advocates are not at all precise about how a forum is supposed to improve *productivity* of this list. They're mostly focused on certain issues, such as ease of entry that I agree are important for a user support forum, but I don't think they really help the devel workers (that's my personal opinion, not anything I can claim expertise on).
Yep. Perhaps we could (re)direct people to file issues with upstreams so their problems are at least tracked.
Sure. I appreciate your input though!
Try and stop me! :-)
Seriously, as I said before, Fedora is currently one of our biggest publicly accessible Mailman 3 installations. We want the best for you first, but if "the best for Fedora" can be Mailman 3, we want it to shine!
Thanks!
kevin
Kevin Fenzi writes:
Thanks for the conversation and a lot of useful information. I think it's mostly moved to "users", which is a channel which I think is more suited to forum style. So I'll go back to lurking, after one comment.
Yep. Perhaps we could (re)direct people to file issues with upstreams so their problems are at least tracked.
These should be triaged by somebody who understands Mailman, as many of the issues that have been mentioned are specific to the Fedora instance. Most prominent is the closure of the web form, with no explicit redirection to social auth or email signup. No need to be massively careful about it, we understand that users find it difficult to identify the right venue to report mail issues in general and list issues in particular. But it's very frustrating for them to get bounced back and forth.
Steve
On Sun, Oct 21, 2018 at 06:33:56PM -0700, Kevin Fenzi wrote:
On 10/21/18 6:12 PM, Stephen J. Turnbull wrote:
Kevin Fenzi writes:
Huh. The only person I know of from Fedora at least that was working on it was abompard. While he's working on other things now, as far as I know he's still working on mailman3/hyperkitty as time permits.
pingu and abadger also contributed. Don't know their exact affiliations or what they're doing for Fedora's installation, but all three have stopped substantial upstream contribution for a couple
Toshio (abadger) works on ansible now. Pierre (pingou) is still in the Fedora engineering group but is focused on pagure and ci work. I only see 2 commits ever from Toshio and all of pingou's commits were at the very start of the project in 2012.
Toshio and I implemented the early version of HyperKitty based on Máirín's designs at a time I was still a part-time contributor. Aurélien has been hired by RH a few months after HK took off and basically he took over from Toshio and I. Since then I did join RH and the Fedora engineering group but since we already had Aurélien working on HK, I kept hanging around a little but focused on other projects. This explains the profile of contributions you're seeing from the three of us :)
Pierre
abompard has a few commits, he and pingu still answer mail but that's about it.
I agree that my activity on HyperKitty has slowed down a lot these last years, because I got involved with other projects too.
No, it can't, can it. Fedora is not keeping up with upstream, which means that "anyone who wants to" isn't upgrading on the Fedora system.
I'm planning to upgrade but I'm currently stuck with the fact that Mailman Core upped it's dependency on Python 3.5+, and only Python 3.4 is easily accessible in EPEL. So we need to come up with a plan to rebuild all the dependency packages for both Python 3.4 and 3.6, which was part of the initial python lib spec file template in EPEL, but never got really implemented by python lib packagers. (see https://lists.fedoraproject.org/archives/list/python-devel@lists.fedoraproje...) Or run the lists on Fedora instead of CentOS. Or install in a venv instead of RPMs. All these solutions come with advandages and tradeoffs.
But the plan is definitely to upgrade, once we have a solution to this issue.
Aurélien
On 10/22/18 7:35 AM, Aurelien Bompard wrote:
abompard has a few commits, he and pingu still answer mail but that's about it.
I agree that my activity on HyperKitty has slowed down a lot these last years, because I got involved with other projects too.
No, it can't, can it. Fedora is not keeping up with upstream, which means that "anyone who wants to" isn't upgrading on the Fedora system.
I'm planning to upgrade but I'm currently stuck with the fact that Mailman Core upped it's dependency on Python 3.5+, and only Python 3.4 is easily accessible in EPEL. So we need to come up with a plan to rebuild all the dependency packages for both Python 3.4 and 3.6, which was part of the initial python lib spec file template in EPEL, but never got really implemented by python lib packagers. (see https://lists.fedoraproject.org/archives/list/python-devel@lists.fedoraproje...)
EPEL folks are trying to plan a python3 work day with whoever we can get from the python sig soon, where we try and move as much of the stack as we can to python36. That should very much help this.
Or run the lists on Fedora instead of CentOS. Or install in a venv instead of RPMs. All these solutions come with advandages and tradeoffs.
But the plan is definitely to upgrade, once we have a solution to this issue.
Yep.
kevin
On Mon, Oct 22, 2018 at 12:40 PM Kevin Fenzi kevin@scrye.com wrote:
On 10/22/18 7:35 AM, Aurelien Bompard wrote:
abompard has a few commits, he and pingu still answer mail but that's about it.
I agree that my activity on HyperKitty has slowed down a lot these last years, because I got involved with other projects too.
No, it can't, can it. Fedora is not keeping up with upstream, which means that "anyone who wants to" isn't upgrading on the Fedora system.
I'm planning to upgrade but I'm currently stuck with the fact that Mailman Core upped it's dependency on Python 3.5+, and only Python 3.4 is easily accessible in EPEL. So we need to come up with a plan to rebuild all the dependency packages for both Python 3.4 and 3.6, which was part of the initial python lib spec file template in EPEL, but never got really implemented by python lib packagers. (see https://lists.fedoraproject.org/archives/list/python-devel@lists.fedoraproje...)
EPEL folks are trying to plan a python3 work day with whoever we can get from the python sig soon, where we try and move as much of the stack as we can to python36. That should very much help this.
Would it be too difficult to just move to Fedora for the Mailman + HyperKitty + Postorius system?
On 10/22/18 9:49 AM, Neal Gompa wrote:
On Mon, Oct 22, 2018 at 12:40 PM Kevin Fenzi kevin@scrye.com wrote:
On 10/22/18 7:35 AM, Aurelien Bompard wrote:
abompard has a few commits, he and pingu still answer mail but that's about it.
I agree that my activity on HyperKitty has slowed down a lot these last years, because I got involved with other projects too.
No, it can't, can it. Fedora is not keeping up with upstream, which means that "anyone who wants to" isn't upgrading on the Fedora system.
I'm planning to upgrade but I'm currently stuck with the fact that Mailman Core upped it's dependency on Python 3.5+, and only Python 3.4 is easily accessible in EPEL. So we need to come up with a plan to rebuild all the dependency packages for both Python 3.4 and 3.6, which was part of the initial python lib spec file template in EPEL, but never got really implemented by python lib packagers. (see https://lists.fedoraproject.org/archives/list/python-devel@lists.fedoraproje...)
EPEL folks are trying to plan a python3 work day with whoever we can get from the python sig soon, where we try and move as much of the stack as we can to python36. That should very much help this.
Would it be too difficult to just move to Fedora for the Mailman + HyperKitty + Postorius system?
Well, it would still be some work, just different. I know mailman3 is under review, we would want that to finish up, then any other packages we need, then fixing up any config for fedora vs rhel, etc.
Might be easier, not sure.
kevin
On Mon, Oct 22, 2018 at 2:22 PM Kevin Fenzi kevin@scrye.com wrote:
On 10/22/18 9:49 AM, Neal Gompa wrote:
On Mon, Oct 22, 2018 at 12:40 PM Kevin Fenzi kevin@scrye.com wrote:
On 10/22/18 7:35 AM, Aurelien Bompard wrote:
abompard has a few commits, he and pingu still answer mail but that's about it.
I agree that my activity on HyperKitty has slowed down a lot these last years, because I got involved with other projects too.
No, it can't, can it. Fedora is not keeping up with upstream, which means that "anyone who wants to" isn't upgrading on the Fedora system.
I'm planning to upgrade but I'm currently stuck with the fact that Mailman Core upped it's dependency on Python 3.5+, and only Python 3.4 is easily accessible in EPEL. So we need to come up with a plan to rebuild all the dependency packages for both Python 3.4 and 3.6, which was part of the initial python lib spec file template in EPEL, but never got really implemented by python lib packagers. (see https://lists.fedoraproject.org/archives/list/python-devel@lists.fedoraproje...)
EPEL folks are trying to plan a python3 work day with whoever we can get from the python sig soon, where we try and move as much of the stack as we can to python36. That should very much help this.
Would it be too difficult to just move to Fedora for the Mailman + HyperKitty + Postorius system?
Well, it would still be some work, just different. I know mailman3 is under review, we would want that to finish up, then any other packages we need, then fixing up any config for fedora vs rhel, etc.
Might be easier, not sure.
I'm doing the review for mailman3, I'm just waiting on Aurelien to respond in it.
Gerald, I'm the person who designed Hyperkitty's concept on a napkin on a shuttlebus with Luke Macken some years ago. Your characterization of it here is incorrect.
I say this with respect, please try to listen more than you post. Hyperkitty stats show you're dominating this conveesation.
~m
On Thursday, 18 October 2018 at 23:04, Gerald B. Cox wrote: [...]
I wouldn't expect it to be a "drop-in" mailing list replacement. Yes, it allows some backward compatibility by providing "mailing-list mode" - but you're going to get a richer experience if you use native interfaces.
So, I tried to answer in one of the threads and it's quite difficult, in my opinion. The message compose pop-up doesn't quote the message I'm replying to, so I can't write my comments in response to specific passages by the previous person. Not being able to reply in context is a deal-breaker to me. What do you call "a richer experience", I wonder.
Regards, Dominik
On Thursday, 18 October 2018 at 23:04, Dominik wrote: [...]
So, I tried to answer in one of the threads and it's quite difficult, in my opinion. The message compose pop-up doesn't quote the message I'm replying to, so I can't write my comments in response to specific passages by the previous person. Not being able to reply in context is a deal-breaker to me. What do you call "a richer experience", I wonder.
In Discourse you have two options:
1. Quote the whole post: first hit reply, then click on the bubble icon to quote the whole post you are replying to.
2. Quote part of the message: select the text you want to quote with the mouse and a Quote button will pop up; click on it and you are ready to reply.
See this cheat sheet: https://www.sitepoint.com/community/t/discourse-cheat-sheet/733
BTW, Hyperkitty doesn't have the partial quoting feature AFAICT.
On Wed, 2018-10-17 at 13:14 -0400, Matthew Miller wrote:
I'm curious what is insufficient about Hyperkitty that Discourse does well at. Wasn't Hyperkitty supposed to give people the forum experience? I admit I haven't used it that much for reading or posting (though I do use it for archive links sometimes and it seems fine for that), since I really like the e-mail interface, so I am not familiar with how nice or un-nice it is to use.
I'm replying from Hyperkitty interface.
I clicked on Quote to quote your email, Randy, but your words appear as written by Matthew. Not good... Also it seems there's no way to mark a thread as read/unread or special. I have the feeling that following a list in Hyperkitty is almost impossible (but I may be missing something as I'm new to it).
So our Hyperkitty version is old here. I can't reporduce the issue on mailman3.org's HK, which is newer. I suspect this is a bug that's been fixed.
~m
Máirín Duffy wrote:
So our Hyperkitty version is old here. I can't reporduce the issue on mailman3.org's HK, which is newer. I suspect this is a bug that's been fixed.
Indeed it was. An infrastructure ticket was filed ~7 months back¹ and the issue was addressed upstream in 7558682 ("Fix quoting of nested replies", 2018-04-24)².
The issue was that hyperkitty simply didn't include any attribution of the quoted text, while many users assumed it did and was quoting the wrong person.
Updating to hyperkitty >= 1.2.0 should resolve the issue. Or the relatively small patch could be applied to our instance -- if updating is a problem.
¹ https://pagure.io/fedora-infrastructure/issue/6802 ² https://gitlab.com/mailman/hyperkitty/commit/7558682
I'm replying from Hyperkitty interface.
I clicked on Quote to quote your email, Randy, but your words appear as written by Matthew. Not good... Also it seems there's no way to mark a thread as read/unread or special. I have the feeling that following a list in Hyperkitty is almost impossible (but I may be missing something as I'm new to it).
Now quoting myself and no name appears.
I was wrong about the read/unread thing. After you visited a link, the thread won't have the envelope icon anymore, *but only after you refreshed the page*. It would be nice to have an immediate update. Visually the read/unread difference should be highlighted more: for example, unread may have some color background.
I have to understand yet how to flag a conversation. Flag = star? https://gitlab.com/mailman/hyperkitty/issues/80
There's a _user_ documentation or a help page for Hyperkitty?
On Wed, 2018-10-17 at 13:14 -0400, Randy Barlow (manually corrected) wrote:
I'm curious what is insufficient about Hyperkitty that Discourse does well at.
A lot of stuff. If you try any Discourse website out there you'll see it immediately.
Just an example. In Hyperkitty, when a new unread message appears in a thread and you click on it, you don't jump to the unread but start from the beginning and you have to search around to guess what's the new message. This is a big showstopper. In Discourse the usability is just miles ahead.
Randy Barlow bowlofeggs@fedoraproject.org writes:
On Wed, 2018-10-17 at 13:14 -0400, Matthew Miller wrote:
Discourse is *definitely* not a smooth, drop-in mailing list replacement like Hyperkitty is.
I'm curious what is insufficient about Hyperkitty that Discourse does well at.
Badges, those are super important. I already have the "First Emoji" and "First Quote" badge on Discourse and I can't wait to get the "First Bold Markup" one.
;-)
"GBC" == Gerald B Cox gbcox@bzb.us writes:
People
GBC> keep saying it isn't sufficient or it doesn't work. I've been GBC> using it for 3 days and looks and acts like a normal mailing list. GBC> _______________________________________________ devel mailing list GBC> -- devel@lists.fedoraproject.org To unsubscribe send an email to GBC> devel-leave@lists.fedoraproject.org Fedora Code of Conduct: GBC> https://getfedora.org/code-of-conduct.html List Guidelines: GBC> https://fedoraproject.org/wiki/Mailing_list_guidelines List GBC> Archives: GBC> https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org
"GBC" == Gerald B Cox gbcox@bzb.us writes:
GBC> People keep saying it isn't sufficient or it doesn't work. I've GBC> been using it for 3 days and looks and acts like a normal mailing GBC> list.
And I've been using it since the RT project switched a couple of years ago. And... it doesn't really look and act like a normal mailing list. Putting aside the issues with formatting of messages which is discussed elsewhere in this thread, certainly it generates messages and accepts replies and so technically it's a mailing list. But as a medium for discussion which involves email, it simply doesn't work out. the system appears to either foster or encourage changes which result in the delivered messages being less useful than messages from a mailing list.
The fundamental problem is that web forums tend to make less use of quoting. The end result is similar to top-posting, but in the other direction. Instead of complete "context" that's primarily useless, the result is no context at all. So you have to browse the rest of the thread in your email program (which thankfully is still possible) or click over to the web site. Neither is ideal.
So, really, you've used it for a couple of days, have declared it fine, and then boldly declared "Fedora should replace mailing lists with Discourse". But I've used it for a bit longer, and my experience has simply been negative. The discourse system simply does not serve to foster email-based discussion.
- J<
On Wed, Oct 17, 2018 at 10:31 AM Jason L Tibbitts III tibbs@math.uh.edu wrote:
"GBC" == Gerald B Cox gbcox@bzb.us writes:
GBC> People keep saying it isn't sufficient or it doesn't work. I've GBC> been using it for 3 days and looks and acts like a normal mailing GBC> list.
And I've been using it since the RT project switched a couple of years ago. And... it doesn't really look and act like a normal mailing list. Putting aside the issues with formatting of messages which is discussed elsewhere in this thread, certainly it generates messages and accepts replies and so technically it's a mailing list. But as a medium for discussion which involves email, it simply doesn't work out. the system appears to either foster or encourage changes which result in the delivered messages being less useful than messages from a mailing list.
The fundamental problem is that web forums tend to make less use of quoting. The end result is similar to top-posting, but in the other direction. Instead of complete "context" that's primarily useless, the result is no context at all. So you have to browse the rest of the thread in your email program (which thankfully is still possible) or click over to the web site. Neither is ideal.
So, really, you've used it for a couple of days, have declared it fine, and then boldly declared "Fedora should replace mailing lists with Discourse". But I've used it for a bit longer, and my experience has simply been negative. The discourse system simply does not serve to foster email-based discussion.
Yes, I've only used it for a few days - but I disagree with your wholesale assumption regarding context. There have been countless times when I've started reading a topic mid discussion for various reasons and I have no idea what the original point was. I then either have to trudge through my mail archives or go to the website and search through the entire topic to understand what is going on.
IMO, it's easier to do that with discourse than with the mailing list.
On 2018-10-17 13:41, Gerald B. Cox wrote:
On Wed, Oct 17, 2018 at 10:31 AM Jason L Tibbitts III <tibbs@math.uh.edu mailto:tibbs@math.uh.edu> wrote:
>>>>> "GBC" == Gerald B Cox <gbcox@bzb.us <mailto:gbcox@bzb.us>> writes: GBC> People keep saying it isn't sufficient or it doesn't work. I've GBC> been using it for 3 days and looks and acts like a normal mailing GBC> list. And I've been using it since the RT project switched a couple of years ago. And... it doesn't really look and act like a normal mailing list. Putting aside the issues with formatting of messages which is discussed elsewhere in this thread, certainly it generates messages and accepts replies and so technically it's a mailing list. But as a medium for discussion which involves email, it simply doesn't work out. the system appears to either foster or encourage changes which result in the delivered messages being less useful than messages from a mailing list. The fundamental problem is that web forums tend to make less use of quoting. The end result is similar to top-posting, but in the other direction. Instead of complete "context" that's primarily useless, the result is no context at all. So you have to browse the rest of the thread in your email program (which thankfully is still possible) or click over to the web site. Neither is ideal. So, really, you've used it for a couple of days, have declared it fine, and then boldly declared "Fedora should replace mailing lists with Discourse". But I've used it for a bit longer, and my experience has simply been negative. The discourse system simply does not serve to foster email-based discussion.
Yes, I've only used it for a few days - but I disagree with your wholesale assumption regarding context. There have been countless times when I've started reading a topic mid discussion for various reasons and I have no idea what the original point was. I then either have to trudge through my mail archives or go to the website and search through the entire topic to understand what is going on.
IMO, it's easier to do that with discourse than with the mailing list.
I've never used Discourse other than to do a very quick perusal these last few days when this came up. I am quite familiar with the forum model however. How does Discourse handle posts you've already read in a thread that's still active. With things like reddit or LWN, you get to read it over and over and over again if you really want to see whats new now. With my mail client, it's extremely easy and effective to put it into threaded mode and highlight the unread (or filter out the already read). With that, I can usually follow along very easily. Yes, sometimes I need to go back in the thread to get context. But its rare that I'll go back to a thread on a forum to find new comments later on once I've read many already. That's like watching a movie, getting a third of the way through it and realizing you've seen it before.
Now for historical review where a thread is dead, something like Discourse would be great due to the cleanliness of not having repeated context. But that's an entirely different use-case and not the primary one.
On Wed, Oct 17, 2018 at 1:55 PM John Florian jflorian@doubledog.org wrote:
How does Discourse handle posts you've already read in a thread that's still active. With things like reddit or LWN, you get to read it over and over and over again if you really want to see whats new now.
Discourse handles this quite well actually, as long as you're logged in. It will keep track of the last post you've seen as well as show you counts of new messages in a particular thread. I wouldn't claim that it's as good as an email client but it's much better than other forums and comment sites.
On Wednesday, 17 October 2018 at 21:15, Jeffrey Ollie wrote:
On Wed, Oct 17, 2018 at 1:55 PM John Florian jflorian@doubledog.org wrote:
How does Discourse handle posts you've already read in a thread that's still active. With things like reddit or LWN, you get to read it over and over and over again if you really want to see whats new now.
Discourse handles this quite well actually, as long as you're logged in. It will keep track of the last post you've seen as well as show you counts of new messages in a particular thread. I wouldn't claim that it's as good as an email client but it's much better than other forums and comment sites.
How does it "know" I read a particular message over e-mail when I access it again over the web interface? I think I can answer that without checking: it doesn't, so you are forced to re-read or at least click through the messages you read already in your e-mail client.
If it had a login-only NNTP interface then it would be possible to sync "read" messages status between web and NNTP. And that would actually be pretty cool.
Regards, Dominik
On Wed, Oct 17, 2018 at 02:54:23PM -0400, John Florian wrote:
With things like reddit or LWN, you get to read it over and over and over again if you really want to see whats new now.
https://lwn.net/Comments/unread will show you comments posted afteryour last visit, with one post in grey to provide context. Requires subscription, though.
On 10/17/18 4:27 PM, Tomasz Torcz wrote:
On Wed, Oct 17, 2018 at 02:54:23PM -0400, John Florian wrote:
With things like reddit or LWN, you get to read it over and over and over again if you really want to see whats new now.
https://lwn.net/Comments/unread will show you comments posted afteryour last visit, with one post in grey to provide context. Requires subscription, though.
Ha! All these years later and now I learn this. Thanks Tomasz for sharing that. And my apologies to Corbet and friends for my lack of awareness.
This is exactly why I love being wrong.
On 2018-10-17 13:41, John Florian wrote:
How does Discourse handle posts you've already read in a thread that's still active. With things like reddit or LWN, you get to read it over and over and over again if you really want to see whats new now.
It handles it very well. You don't have to read it over and over; just click on the thread and you'll jump to the first unread message.
On Tue, Oct 16, 2018 at 01:35:01PM -0400, Simo Sorce wrote:
Ah nice side-thing I just found out, discussion.fedoraproject.org does not show *anything* unless I let my browser access this website http://discourse-cdn-sjc1.com
That's because we're using the hosted option. We could instead decide to host ourselves; we didn't want to do that for the initial experiment. But, actually, this is not nefarious. I encourage you to actually load that page and see what is there. Or, actually, it's short, so I'll copy it here:
What is this site? ==================
As defined in Best Practices for Speeding Up Your Web Site, this is a domain operated by discourse.org for quickly serving static assets that make up hosted Discourse forums:
1. We use a lightweight, minimal web server for serving simple static files as quickly as possible.
2. Using a different domain name increases the number of parallel downloads that the browser can perform.
3. Using a different domain name means browser cookies are not sent or attached to requests to this domain, reducing traffic and roundtrip times, and increasing cacheability.
4. The files are centralized (served from a single domain) and static (they do not change often), and thus gain maximum benefit from a CDN (content delivery network) which can send those files from the internal node geographically closest to the user requesting them.
What is served from this domain? ================================
Static assets necessary for a discourse.org forum to operate, such as site logos, JavaScript code, and images in posts. If you view source on the hosted Discourse forum page and search for "CDN", you'll see exactly which assets are served from this domain.
On Tue, Oct 16, 2018 at 10:32 AM Simo Sorce simo@redhat.com wrote:
On Tue, 2018-10-16 at 07:12 -0700, Gerald B. Cox wrote:
From the foreman post, in the comments: Email interface does not really work, there should be rather called “notifications”. You must visit the site eventually.
I personally find this a fatal flaw, I do not have time to waste visiting slow websites for each project I follow. If I have to switch away from my mail client I just end up unfollowing the project unless I am focusing specifically on it at that time.
It really is a much better solution.
Better for whom and for what purpose ?
I am sure it won't be better for me, all "web forums" I ever used made for my disappearence from the platform as it was too expensive (time- wise) for me to keep up.
Simo.
I believe that commenter was confusing notification mode with mail list
mode: https://meta.discourse.org/t/what-is-mailing-list-mode/46008/9
But with that said, he still closes with: '"Now, to the point. I think the change was for the good. Thanks for the report."
So, my opinion on email vs. web forum is that it is comes down to freedom vs. lock-in.
With mailing lists (and Usenet), messages are distributed in a more-or-less well-defined format, and users are able to choose clients, filtering, etc. to suit their use patterns. Sometimes people do novel things that help them handle volume, find and track topics they are interested in, etc.
With a web forum, users are limited to consuming content in the way the forum developer created, as chosen to be applied by the system managers. Rarely do users have any control over how they access the content, and they still only can control it in the way the developers thought of, implemented, tested, and continue to support. I've used web forums where a function I used regularly went away with an upgrade because the developers didn't think enough people used them and didn't want to support them anymore.
Now, that freedom to do as you please has kind of fallen by the wayside on the Internet in general; personal blogs went to hosted blogs (with more uniform interfaces) and then on to mass-hosted social media (with an interface decided entirely by the company running it). So maybe I'm just old-fashioned that way and not enough other people care.
I'll admit up front that I haven't checked out Discourse yet. However, I haven't seen before anything that handles the way I consume email, especially mailing lists. I'll keep selected messages of a thread around for later reference, flag messages so they'll be highlighted when I open a folder in the future, filter out certain keywords or posters on rare occasions, and more - and that's just off the top of my head.
Also, I follow a bunch of different mailing lists. If they all were web forums, I'd have a bunch of different interfaces to deal with, sites to visit, functionality to learn, etc., instead of my single mail client that I can tweak to my personal use patterns and can bring it all together.
On Sat, Oct 20, 2018 at 2:30 PM, Chris Adams linux@cmadams.net wrote:
So, my opinion on email vs. web forum is that it is comes down to freedom vs. lock-in.
Right. What does migration out of Discourse look like?
On Tue, 2018-10-16 at 07:12 -0700, Gerald B. Cox wrote:
On Fri, Oct 5, 2018 at 9:21 AM Matthew Miller < mattdm@fedoraproject.org> wrote: ...
That's why the general trend is *away* from email.
The Foreman community recently switched away from mailing lists in
this way,
and
https://theforeman.org/2018/07/discourse-6-months-on-impact-assesment.html
is really interesting and helpful read on the topic for those who
might have
some ... trepidation.
I'm not sayin' we are ready to shut this list down, but it's
honestly worth
considering if a different approach will be more effective.
Before the daggers come out.... please take some time to checkout Fedora Discourse https://discussion.fedoraproject.org/ and read the Foreman community link above.
I really hope not.
Discourse web interface is bloated and painful to work with.
As I had previously mentioned, Discourse allows for RSS feeds and email notifications. It also has a mailing list mode.
It really is a much better solution.
It isn't, even in mailing mode discourse sends HTML messages which makes email responses almost impossible and force you to use the bloated web interface.
On Thu, Oct 4, 2018 at 4:33 PM Matthew Miller mattdm@fedoraproject.org wrote:
The fact is, the world has moved away from quoted mail with inline replies. Top posting rules basically everywhere except hold-out old-school mailing lists. Gmail, both on the web and _especially_ on mobile, makes it almost impossible to do quoting and inline replies properly. I can't imagine other big consumer email providers are much better.
Yes, this sucks for having nuanced point-by-point discussions. But I don't think that's a tide we're really going to hold back. We should look at other tools that fit with the way people do asynchronous online discussion today. As Brian noted, we're trying out Discourse at https://discussion.fedoraproject.org/ — if you're interested in trying it out, come join the conversation there.
Matthew Miller mattdm@fedoraproject.org Fedora Project Leader
Please no. No more Discourse. Having already stopped keeping up with one major project (Chef.io) because of that switch.Far easier to live with how broken gmail is and possibly just stripping out html content, then trying force web forum software into being useful.
On Thu, Oct 4, 2018 at 2:33 PM Matthew Miller mattdm@fedoraproject.org wrote:
The fact is, the world has moved away from quoted mail with inline replies. Top posting rules basically everywhere except hold-out old-school mailing lists. Gmail, both on the web and _especially_ on mobile, makes it almost impossible to do quoting and inline replies properly. I can't imagine other big consumer email providers are much better.
Yes, this sucks for having nuanced point-by-point discussions. But I don't think that's a tide we're really going to hold back. We should look at other tools that fit with the way people do asynchronous online discussion today. As Brian noted, we're trying out Discourse at https://discussion.fedoraproject.org/ — if you're interested in trying it out, come join the conversation there.
Completely agree - and discourse is a great solution. It allows for RSS feeds and email notifications = participation is a click away. Your email inbox is no longer cluttered with hard to follow threads, nor do you have to worry about using space to archive conversations which are important to you... just use the bookmark feature.
On Tue, Oct 09, 2018 at 06:58:27AM -0700, Gerald B. Cox wrote:
Completely agree - and discourse is a great solution. It allows for RSS feeds and email notifications = participation is a click away. Your email inbox is no longer cluttered with hard to follow threads, nor do you have to worry about using space to archive conversations which are important to you... just use the bookmark feature.
One click is too much for my terminal email client via SSH on my phone. My email client already supports filtering into separate mailboxes for each list and also supports threads shown in a hierarchy. If Fedora lists go away in favor of a web forum, I will probably just not participate anymore.
On Tue, Oct 9, 2018 at 7:06 AM Anderson, Charles R cra@wpi.edu wrote:
One click is too much for my terminal email client via SSH on my phone. My email client already supports filtering into separate mailboxes for each list and also supports threads shown in a hierarchy. If Fedora lists go away in favor of a web forum, I will probably just not participate anymore.
I just tried it out and found it incredibly easy to use. Not only that, noticed I had made an error in my post and was able to simply edit it. Easily setup a rss feed for notifications. Love it.
On 10/09/2018 04:06 PM, Anderson, Charles R wrote:
One click is too much for my terminal email client via SSH on my phone. My email client already supports filtering into separate mailboxes for each list and also supports threads shown in a hierarchy. If Fedora lists go away in favor of a web forum, I will probably just not participate anymore.
That's more or less the same for me. My fetchmail, procmail and thunderbird setup gives me a way to keep an eye on this (and many other mailing lists) without much effort. I will not consider switching to a forum based website, since my experience with projects using them has been discouraging. Any "mailing list mode" will be a second class citizen, with limitations and constant babysitting, and certainly due to be removed in the future because "most of our users are using the web UI".
I'll probably not follow Fedora anymore. My folders contain all fedora and fedora-devel mails since my subscription, on 24/12/2005 (that is, Fedora 4), not sure if there will be a 2019 folder.
Regards.
On 04 Oct 2018, at 23:32, Matthew Miller mattdm@fedoraproject.org wrote:
The fact is, the world has moved away from quoted mail with inline replies. Top posting rules basically everywhere except hold-out old-school mailing lists. Gmail, both on the web and _especially_ on mobile, makes it almost impossible to do quoting and inline replies properly. I can't imagine other big consumer email providers are much better.
Top posting rules where people talk at each other. This is common in the corporate world where email is treated like letter writing.
When people are having a conversation with each other, you reply like this. In the corporate world, I often see people struggle putting their responses in red text, or in caps, instead of quoting properly. Quoting properly isn’t “old school” or “hold out”, it’s basic courtesy.
I don’t find quoting on mobile particularly difficult (speaking for iPhone).
Regards, Graham —
I suggest that if we think this is a serious issue and that we want to get the maximum possible help to those who are forced to pay by the bit that we stop sending them mail they don't need all together. Lets move this mailing list (and others) to discussion.fedoraproject.org.
Why switch, when we already have
https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org
? it does good quoting already, you get to click what you want, and i'm guessing this reply i'm sending will show up plaintext.
Ray
On Thu, Oct 4, 2018 at 7:27 PM Ray Strode rstrode@redhat.com wrote:
I suggest that if we think this is a serious issue and that we want to get the maximum possible help to those who are forced to pay by the bit that we stop sending them mail they don't need all together. Lets move this mailing list (and others) to discussion.fedoraproject.org.
Why switch, when we already have
https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org
? it does good quoting already, you get to click what you want, and i'm guessing this reply i'm sending will show up plaintext.
You are correct. It worked exactly like that. However, it seems like there just simply wasn't much in the way of promoting the HyperKitty interface after the initial growing pains of deploying it and converting everything over.
On Thu, Oct 04, 2018 at 11:27:02PM -0000, Ray Strode wrote:
Why switch, when we already have https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org ? it does good quoting already, you get to click what you want, and i'm guessing this reply i'm sending will show up plaintext.
Unfortunately, all is not rosy there. See this thread on the users' list from this fall about confusion with hyperkitty quoting: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org/... (Look back at your message I'm replying to here, and note that the standard attribution line is missing. And that's not the only issue identified.)
Additionally, when one does reply, it's just using the standard web browser text box, and to my knowledge that doesn't have an easy "delete by line" keyboard command, let alone things like smart reflow with quoting. So that's not super-ideal for inline replies. Hyperkitty's threaded view also is rudimentary compared to a good mail client.
All of this stuff would be something we could invest development resources in and make better, but we don't really _have_ those resources, and no significant outside-of-fedora hyperkitty development community ever developed. We're left with some pretty awful things like the prominent "Sign Up" button on every Fedora list leading to a big, ugly "Sign Up Closed: We are sorry, but the sign up is currently closed" screen — which is not very inviting, to say the least.
On Fri, Oct 5, 2018 at 11:11 AM Matthew Miller mattdm@fedoraproject.org wrote:
On Thu, Oct 04, 2018 at 11:27:02PM -0000, Ray Strode wrote:
Why switch, when we already have https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org ? it does good quoting already, you get to click what you want, and i'm guessing this reply i'm sending will show up plaintext.
Unfortunately, all is not rosy there. See this thread on the users' list from this fall about confusion with hyperkitty quoting: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org/... (Look back at your message I'm replying to here, and note that the standard attribution line is missing. And that's not the only issue identified.)
Additionally, when one does reply, it's just using the standard web browser text box, and to my knowledge that doesn't have an easy "delete by line" keyboard command, let alone things like smart reflow with quoting. So that's not super-ideal for inline replies. Hyperkitty's threaded view also is rudimentary compared to a good mail client.
All of this stuff would be something we could invest development resources in and make better, but we don't really _have_ those resources, and no significant outside-of-fedora hyperkitty development community ever developed. We're left with some pretty awful things like the prominent "Sign Up" button on every Fedora list leading to a big, ugly "Sign Up Closed: We are sorry, but the sign up is currently closed" screen — which is not very inviting, to say the least.
Python is migrating to Mailman 3 now: https://mail.python.org/mm3/mailman3/lists/
And there's interest in the openSUSE community to switch from mlmmj to Mailman 3 with HyperKitty.
So the community *is* developing, it's just much later than probably expected.
On Thu, Oct 04, 2018 at 11:27:02PM -0000, Ray Strode wrote:
Unfortunately, all is not rosy there. See this thread on the users' list from this fall about confusion with hyperkitty quoting: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.o...
It's hilarious reading this thread in Hyperkitty and having absolutely no issue understanding which quote is from whom.
(Look back at your message I'm replying to here, and note that the standard attribution line is missing. And that's not the only issue identified.)
Additionally, when one does reply, it's just using the standard web browser text box, and to my knowledge that doesn't have an easy "delete by line" keyboard command, let alone things like smart reflow with quoting.
These sound like good enhancement features that are likely not essential for the core target user.
So that's
not super-ideal for inline replies. Hyperkitty's threaded view also is rudimentary compared to a good mail client.
All of this stuff would be something we could invest development resources in and make better, but we don't really _have_ those resources,
But we can file bugs against Discourse and they will be magically and quickly fixed to our satisfaction, yes?
We can invest money and resources into Discourse deployment, and time and effort trying to convert each ML over to it one by one, dealing with all of the existing docs and pointers to MLs and updating them, no sweat?
Threading is something I could fix on the front end with some backend help. Is anybody handy with Django who is interested?
Is that the biggest issue to tackle?
I'm concerned that those proposing Discourse seem to not have used Hyperkitty at length.
and no
significant outside-of-fedora hyperkitty development community ever developed. We're left with some pretty awful things like the prominent "Sign Up" button on every Fedora list leading to a big, ugly "Sign Up Closed: We are sorry, but the sign up is currently closed" screen — which is not very inviting, to say the least.
I have no idea what that is, but it sounds like a minor bug that probably isn't difficult to fix.
But sure let's split our community over that.
~m
Il giorno ven 19 ott 2018 alle 5:53, =?iso-8859-1?b?TeFpcu1u?= Duffy duffy@fedoraproject.org ha scritto:
I'm concerned that those proposing Discourse seem to not have used Hyperkitty at length.
I'm one of those. I've never used Hyperkitty before, but I'll give it a try now.
On Fri, Oct 19, 2018 at 03:53:06AM -0000, Máirín Duffy wrote:
But we can file bugs against Discourse and they will be magically and quickly fixed to our satisfaction, yes?
Of course not. However: development is very active. https://github.com/discourse/discourse/commits/master
I'm concerned that those proposing Discourse seem to not have used Hyperkitty at length.
As you know, I was very excited about HyperKitty. I *did* try it very seriously at first, but not at length, because it quickly became apparent that it wasn't really up to the task of being my primary interface to email. I just couldn't use it for day-to-day communication. Not necessarily any single thing, but lots and lots of fundamentals. How do I get a list of new threads? How do I get a list of threads I've read but which have new responses, and ideally show only the new responses? How can I mute a thread I don't want to be alerted on? How do I get to the next thread from the *bottom* of a thread I just read? How can I search... usefully at all? My point isn't to rag on HyperKitty, but I could definitely go on.
I tried for a while to file suggestions and bug reports, but especially after the extra two years it took to even get deployed, it was *very* clear there were no resources for ongoing development from Red Hat, no significant non-RH Fedora development, and no meaningful outside development either. Basic things like https://gitlab.com/mailman/hyperkitty/issues/64 didn't even get *responses*. So, I stuck with my previous email client setup.
And the thing is, it's *not just me*. Take a look at https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/ right now. Every single thread shows a "meh" face and "+0/-0". It says
Most popular discussions No vote has been cast this month (yet).
It is the 19th of the month. Not a single vote on our most busy mailing list. The same is true for every other list I looked at. People just aren't using this.
I *really* think HyperKitty has potential. But we can't run on potential. Discourse is a pure open source project that is *really catching on and successful*. It's not perfect either, of course, but we're way better off aligning with something with momentum.
On Fri, Oct 19, 2018 at 4:26 AM Matthew Miller mattdm@fedoraproject.org wrote:
On Fri, Oct 19, 2018 at 03:53:06AM -0000, Máirín Duffy wrote:
But we can file bugs against Discourse and they will be magically and quickly fixed to our satisfaction, yes?
Of course not. However: development is very active. https://github.com/discourse/discourse/commits/master
I'm concerned that those proposing Discourse seem to not have used Hyperkitty at length.
As you know, I was very excited about HyperKitty. I *did* try it very seriously at first, but not at length, because it quickly became apparent that it wasn't really up to the task of being my primary interface to email. I just couldn't use it for day-to-day communication. Not necessarily any single thing, but lots and lots of fundamentals. How do I get a list of new threads? How do I get a list of threads I've read but which have new responses, and ideally show only the new responses? How can I mute a thread I don't want to be alerted on? How do I get to the next thread from the *bottom* of a thread I just read? How can I search... usefully at all? My point isn't to rag on HyperKitty, but I could definitely go on.
I tried for a while to file suggestions and bug reports, but especially after the extra two years it took to even get deployed, it was *very* clear there were no resources for ongoing development from Red Hat, no significant non-RH Fedora development, and no meaningful outside development either. Basic things like https://gitlab.com/mailman/hyperkitty/issues/64 didn't even get *responses*. So, I stuck with my previous email client setup.
And the thing is, it's *not just me*. Take a look at https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/ right now. Every single thread shows a "meh" face and "+0/-0". It says
Most popular discussions No vote has been cast this month (yet).
It is the 19th of the month. Not a single vote on our most busy mailing list. The same is true for every other list I looked at. People just aren't using this.
I *really* think HyperKitty has potential. But we can't run on potential. Discourse is a pure open source project that is *really catching on and successful*. It's not perfect either, of course, but we're way better off aligning with something with momentum.
Completely agree - well stated. It's quite obvious when using both tools which lends itself to better conversation and collaboration. I realize people are resistant to change - but once they start to use and learn about the new features and functionality - most people come around. Sometimes people need a nudge to change and adapt - myself included. It's just human nature - but as I get older, I try to keep an open mind and be adaptable.
I just couldn't use it for day-to-day communication. Not necessarily any single thing, but lots and lots of fundamentals. How do I get a list of new threads? How do I get a list of threads I've read but which have new responses, and ideally show only the new responses? How can I mute a thread I don't want to be alerted on? How do I get to the next thread from the *bottom* of a thread I just read? How can I search... usefully at all? My point isn't to rag on HyperKitty, but I could definitely go on.
Please do. Neal and I are starting up an effort. I reached out to Abhilash, the upstream lead, last night on the devel list and he was very responsive to this idea. He's already created a gitlab subproject for our efforts upstream.
I tried for a while to file suggestions and bug reports, but especially after the extra two years it took to even get deployed, it was *very* clear there were no resources for ongoing development from Red Hat, no significant non-RH Fedora development, and no meaningful outside development either. Basic things like https://gitlab.com/mailman/hyperkitty/issues/64 didn't even get *responses*. So, I stuck with my previous email client setup.
And the thing is, it's *not just me*. Take a look at
It is the 19th of the month. Not a single vote on our most busy mailing list. The same is true for every other list I looked at. People just aren't using this.
Matthew, the target user for Hyperkitty isn't a devel-list reader.
~m
On Sat, Oct 20, 2018 at 12:29 AM Máirín Duffy duffy@fedoraproject.org wrote:
I just couldn't use it for day-to-day communication. Not necessarily any single thing, but lots and lots of fundamentals. How do I get a list of new threads? How do I get a list of threads I've read but which have new responses, and ideally show only the new responses? How can I mute a thread I don't want to be alerted on? How do I get to the next thread from the *bottom* of a thread I just read? How can I search... usefully at all? My point isn't to rag on HyperKitty, but I could definitely go on.
Please do. Neal and I are starting up an effort. I reached out to Abhilash, the upstream lead, last night on the devel list and he was very responsive to this idea. He's already created a gitlab subproject for our efforts upstream.
I tried for a while to file suggestions and bug reports, but especially after the extra two years it took to even get deployed, it was *very* clear there were no resources for ongoing development from Red Hat, no significant non-RH Fedora development, and no meaningful outside development either. Basic things like https://gitlab.com/mailman/hyperkitty/issues/64 didn't even get *responses*. So, I stuck with my previous email client setup.
And the thing is, it's *not just me*. Take a look at
It is the 19th of the month. Not a single vote on our most busy mailing list. The same is true for every other list I looked at. People just aren't using this.
Matthew, the target user for Hyperkitty isn't a devel-list reader.
Can you share more information about the target audience? I read through the non-install docs at hyperkitty.readthedocs.io which is where the Gitlab project sent me. I could only find statements about the shortcomings with Pipermail the project was trying to fix. I did not see a target user statement or find one for Pipermail.
Thanks,
bex
~m _______________________________________________ devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-leave@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org
Check my blog, where there are even scans of the napkin sketches. Or Google "hyperkitty ux" my blog and my outreachy interns blog pop up.
~m
On Sat, Oct 20, 2018 at 4:19 PM Máirín Duffy duffy@fedoraproject.org wrote:
Check my blog, where there are even scans of the napkin sketches. Or Google "hyperkitty ux" my blog and my outreachy interns blog pop up.
Is this in response to my question? For some reason many of your emails have no quoting or context.
If this is, why has the project chosen to not document that in their code/docs? That would, it seems, help contributors stay focused on the goal.
regards,
bex
~m _______________________________________________ devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-leave@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org
-- Brian (bex) Exelbierd | bexelbie@redhat.com | bex@pobox.com Fedora Community Action & Impact Coordinator @bexelbie | http://www.winglemeyer.org
If this is, why has the project chosen to not document that in their code/docs? That would, it seems, help contributors stay focused on the goal.
https://wiki.list.org/DEV/Home?action=show&redirect=DEV
I went to list.org, clicked on developer wiki, it's on the front page.
This is veering a bit off topic now and I smell unnecessary snark. Something, btw, I'm sure happens on Discourse.
~m
On Sat, Oct 20, 2018 at 10:12 PM Máirín Duffy duffy@fedoraproject.org wrote:
If this is, why has the project chosen to not document that in their code/docs? That would, it seems, help contributors stay focused on the goal.
https://wiki.list.org/DEV/Home?action=show&redirect=DEV
I went to list.org, clicked on developer wiki, it's on the front page.
Interesting. I went to my search engine of choice and typed in hyperkitty. The first two links were:
https://gitlab.com/mailman/hyperkitty and https://hyperkitty.readthedocs.io/en/latest/
These felt fully authoritative and I did not notice a link to the url you cite.
This is veering a bit off topic now and I smell unnecessary snark. Something, btw, I'm sure happens on Discourse.
The question was genuine. Additionally, even at this site, the user profile doesn't seem to be present. I believe it can be gathered from reading every blog you have linked and all of the associated comments. This seems rather fragile as it is a long spin-up time and if your blog goes away, we have no user profile.
Unrelated, if you have edit rights on that page, the GitHub link is out of date and the page still refers to Fedora Hosted.
regards,
bex
~m _______________________________________________ devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-leave@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org
Brian (bex) Exelbierd writes:
The question was genuine. Additionally, even at this site, the user profile doesn't seem to be present.
I doubt there is a user profile chapter or page for HyperKitty at the Mailman project. If there is one, it would be in the HyperKitty docs themselves.
Unrelated, if you have edit rights on that page, the GitHub link is out of date and the page still refers to Fedora Hosted.
I hope you don't mean the Mailman wiki DEV/Home page? Couldn't find either reference there.
Steve
On Mon, Oct 22, 2018 at 6:56 PM Stephen J. Turnbull stephen@xemacs.org wrote:
Brian (bex) Exelbierd writes:
The question was genuine. Additionally, even at this site, the user profile doesn't seem to be present.
I doubt there is a user profile chapter or page for HyperKitty at the Mailman project. If there is one, it would be in the HyperKitty docs themselves.
As far as I know it is not and is only present as a set of links to a personal blog here: https://wiki.list.org/HyperKitty
Unrelated, if you have edit rights on that page, the GitHub link is out of date and the page still refers to Fedora Hosted.
I hope you don't mean the Mailman wiki DEV/Home page? Couldn't find either reference there.
This is hte page in question: https://wiki.list.org/HyperKitty
I believe my previous email contained the wrong link because of a cut and paste error.
regards,
bex
Steve _______________________________________________ devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-leave@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org
Oh and I should also point out - you shouldn't have to read all of that, the point was to make our mailing lists accessible to folks who aren't mailing list users, who are less technical, younger, etc., to be more conclusive. Same reason Discourse is being peddled here. Big diff is we keep the mail interface up for exisiting users rather than dropping them as a target!!
This is why folks are trying to point out here, you may get new users w discourse but they will be different and you may lose who you have now.
~m
On Fri, Oct 19, 2018 at 10:28:45PM -0000, Máirín Duffy wrote:
Please do. Neal and I are starting up an effort. I reached out to Abhilash, the upstream lead, last night on the devel list and he was very responsive to this idea. He's already created a gitlab subproject for our efforts upstream.
Okay; I'll be happy to. But I'll do it separately from this thread.
list. The same is true for every other list I looked at. People just aren't using this.
Matthew, the target user for Hyperkitty isn't a devel-list reader.
But devel is by far our most active and important list.
On Fri, Oct 19, 2018 at 03:53:06AM -0000, Máirín Duffy wrote:
On Thu, Oct 04, 2018 at 11:27:02PM -0000, Ray Strode wrote:
Unfortunately, all is not rosy there. See this thread on the users' list from this fall about confusion with hyperkitty quoting: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.o...
It's hilarious reading this thread in Hyperkitty and having absolutely no issue understanding which quote is from whom.
You're replying to an email from Matthew but it show here as coming from Ray Strode. This is the bug Matthew was referring to.
Pierre