At flock recently and at todays meeting we discussed plans moving
The first of those is to check our membership and make sure everyone is
still wanting to be involved. We have currently 1 open seat, but others
may wish to bow out now if they don't have time or desire.
So, could all existing Server working group members chime in on list if
they are interested in staying in the group or giving up their seat?
We will give two weeks (until the 2016-08-23 meeting) and then look to
fill empty seats with interested parties. Once the seats are filled we
will look at setting a new meeting time.
Sgallagh is going to try and clean up and present on list a new PRD for
us to review over the next week. Members can use that discussion to see
if they are still wishing to participate.
We need to discuss our new plans going forward for the Fedora Server Edition.
Fedora has started adopting the Kellogg Logic Model for planning purposes. At
this year's Flock in Kraków, Poland, we started this process with the people who
I put together a Kanban board as an imperfect visualization of this
discussion so we can use that. In a proper visualization of the model, all of
the cards would have arrows pointing to the column to their right to indicate
which card they were implementing.
I'd like to continue this exercise with the larger Server SIG audience on this
mailing list. We will go through five unique phases, each corresponding to the
five columns in the Kellogg Logic Model from right to left (so we will start at
the rightmost column, "Impact" and then proceed through "Outcomes", "Outputs",
"Activities", and "Resources in order).
This thread will begin with "Impact". In the Kellogg Logic Model, this is the
high-level change we want to effect in the world. An analogue to this in our
original PRD would be our Mission and Vision statements.
At Flock, we summarized (and shortened) the Vision and Mission as follows:
* Vision: Admins default to choosing Fedora Server as a deployment platform for
apps and services
* Mission: Produce common base and server roles (best practices in a can)
So let's start by discussing these. Do we like these as our long-term end goals?
Should they be more specific or more vague?
Stephen Smoogen asked me to go over the logic model with him a bit today, which
we did in #fedora-server. Since it may be useful to others, here's a log of our
conversation. I hope it helps disambiguate things.
(12:01:08 PM) sgallagh: OK, so let me start from a fairly high-level view.
(12:01:45 PM) sgallagh: We originally developed the Server PRD shortly after
Flock Charleston with a bunch of ideas we had for what the Server Edition should
(12:02:23 PM) sgallagh: It's been a couple years since then and the world has
changed, so instead of the small tweaks we made to the PRD last year, we decided
it might be time to take the design back to brass tacks
(12:02:50 PM) smooge: that is good. I am looking at using this process on EPEL
also (since it doesn't even have a PRD)
(12:03:03 PM) sgallagh: As part of that effort, I latched on to the Kellogg
Logic Model as a means to accomplish two things:
(12:03:18 PM) sgallagh: 1) Provide a structured brainstorming model for what we
want to get done
(12:03:30 PM) sgallagh: 2) Provide an organized, step-by-step guide for how to
(12:04:12 PM) sgallagh: Essentially, we will use this model to provide a
high-level view of what we want to achieve over the next couple years
(12:04:44 PM) sgallagh: The KLM is broken up into five main categories: three
that we can affect directly and two that are results or consequences from those.
(12:05:32 PM) sgallagh: Normally, planning is done from right to left; in
essence we start by describing what we want the end-state to look like and then
work backwards from there, ensuring that every step along the way ties itself
directly to those goals
(12:06:04 PM) sgallagh: Execution goes left to right.
(12:07:17 PM) sgallagh: I should mention the five categories (from left to
right): "Inputs", "Activities", "Outputs" and then "Outcomes" and "Impact"
(12:07:55 PM) sgallagh: So planning begins with "Impact". In other words, what
is the grand change that we want to see in the world/industry as a result of our
(12:08:32 PM) sgallagh: smooge: With me so far?
(12:09:01 PM) smooge: trying
(12:09:32 PM) sgallagh: smooge: OK, what can I do to make it clearer?
(12:10:08 PM) smooge: I don't understand how we can plan for stuff we don't have
(12:10:46 PM) smooge: So if we can work on Impacts, Activities and Outputs but
Outcomes and Impact are not ours to actually say about.. why do we start with
(12:11:02 PM) smooge: does that make sense?
(12:11:11 PM) sgallagh: I didn't say we had no affect over them
(12:11:22 PM) sgallagh: I said that we don't have *direct* affect.
(12:11:45 PM) sgallagh: For a simple example: I want my employees to be healthier.
(12:11:59 PM) sgallagh: So I build a gymnasium in the office.
(12:12:13 PM) sgallagh: I can't *directly* make my employees healthy, but I can
provide a path to it.
(12:13:03 PM) sgallagh: Actually, let's take that as an example to work through
(12:13:19 PM) smooge: ok sounds good
(12:13:58 PM) sgallagh: The Impact I want to have is "Improved employee retention"
(12:14:19 PM) sgallagh: An Outcome that I might seek is "Employees spend their
PTO on vacation instead of sick leave"
(12:15:04 PM) sgallagh: So one of many Outputs I might include would be building
a gym for them to exercise before/after work.
(12:15:24 PM) sgallagh: smooge: Is that a little easier to follow?
(12:15:37 PM) smooge: yes
(12:16:18 PM) smooge: though my pedantic brain tries to rebel at connecting the
(12:16:24 PM) sgallagh: So then we'd go down to Activities which would be things
like "Purchase gym equipment" and then futher to Inputs which would be capital
(12:16:40 PM) sgallagh: Well, that's actually the most important part
(12:16:48 PM) sgallagh: And yes, this is a contrived example
(12:17:20 PM) sgallagh: The most important aspect of this is that nothing
appears to the left in the model without directly serving at least one element
to the right (except Impact, of course)
(12:18:06 PM) sgallagh: Among other things, this helps trim down on work that
doesn't directly advance the mission
(12:19:30 PM) sgallagh: So my first goal for this process is to get everyone in
the WG on the same page wrt our "mission"
(12:19:47 PM) sgallagh: *How* do we want to change the world
(12:20:13 PM) sgallagh: And then break it down until we have a path there
(12:20:56 PM) ***sgallagh leaves room for comments
(12:21:18 PM) smooge: ok that makes a lot more sense than the web pages I was
(12:22:03 PM) sgallagh: Oh, good
(12:22:56 PM) sgallagh: It's really not anything that there aren't a hundred
other frameworks for, but since we weren't using *any* of them, going with the
one that other parts of the Fedora Project latched on to seemed sensible
(12:23:01 PM) smooge: so when I looked at the taiga board I felt the impacts
were too large to be achievable. I see "Improved employee retention" as
something that is achievable because it can be measured against
(12:23:58 PM) sgallagh: Well, this model differentiates between "capable of
being measured" and "we have a way to measure it".
(12:24:19 PM) smooge: I am trying to figure out how much of that inability to
see due to inbuilt blinders
(12:24:21 PM) sgallagh: But when we get to that phase, we may indeed change
what's there to make it more possible
(12:24:48 PM) smooge: since I have a ton of inbuilt blinders :)
(12:25:38 PM) sgallagh: Please also be aware that the Taiga board is the result
of 20 minutes of brainstorming at Flock. They're a guidepost
(12:25:39 PM) smooge: the other part that didn't come up in the articles I read
was how we move left to right in the KLM.
(12:25:49 PM) sgallagh: I expect them to change markedly once the whole WG is
hacking at it
(12:26:41 PM) sgallagh: smooge: Sorry, could you rephrase that? I'm not sure
what statement you are making.
(12:26:58 PM) smooge: as in if the Impact was "End World Hunger" but all the
items on the left were unable to make that happen.. how do you go with "Ok I
have N inputs, what can I do with that to and how do I change End World Hunger
to End Hunger on my street"
(12:27:39 PM) sgallagh: I'm still not quite following.
(12:27:49 PM) sgallagh: Inputs isn't "what I have". It's "what I need"
(12:28:51 PM) smooge: ok. so I normally deal with "what I have" because I never
have "what I need"
(12:28:52 PM) sgallagh: If you don't have what you need when you get down to
that point, then it's time to consider new Outputs, probably.
(12:29:28 PM) sgallagh: Right, and that's probably the easiest place to get
(12:30:12 PM) sgallagh: in this model, planning always moves right to left. If
you get to the end and discover that "what I need" doesn't match "what I have",
you either need to acquire what you don't have, reconsider what you need, or
change what you plan to deliver.
(12:30:22 PM) sgallagh: The model will iterate in that case
(12:32:37 PM) sgallagh: smooge: Do you understand, or am I being obtuse?
(12:32:53 PM) smooge: no I believe I understand.
(12:33:32 PM) smooge: I am mostly writing and then rewriting responses because
my brain is picking apart the framework for not covering all the counterpoints
it keeps coming up with
(12:33:48 PM) smooge: and I am telling myself that a response isn't needed to do
(12:34:18 PM) smooge: it is just a framework to get discussion going not a
framework like ruby on rails that I am depending my website on
(12:34:45 PM) sgallagh: yes, exactly.
(12:35:02 PM) sgallagh: It's something we can point to later every time someone
says "What if we did X?"
(12:35:18 PM) smooge: so my silence is me typing stuff and then going and saying
that sentence was not going to be helpful.. retype, retype
(12:35:56 PM) smooge: versus me not paying attention.
(12:36:26 PM) sgallagh: I wasn't making any accusations. I was mostly just
hoping I wasn't talking over you :)
(12:36:54 PM) smooge: soryr I didn't feel like you were... I was being quiet a
lot when you were waiting for things
(12:37:10 PM) sgallagh: What responses were you thinking, though?
(12:37:34 PM) menantea is now known as menantea_away
(12:37:42 PM) sgallagh: I mean, your last point is correct; it's meant for
framing the discussion, not for deciding specific technical actions.
(12:38:13 PM) sgallagh: smooge: For another example, mattdm used this one at
(12:39:01 PM) smooge: well most of problems with it are the lack of feeding
stuff you find from the left back into the right. You waterfall into the top and
work down to the bottom over and over again
(12:40:11 PM) sgallagh: Right, but that's very intentional.
(12:40:30 PM) sgallagh: If the stuff you find in the left doesn't already match
something that you're trying to achieve, it's *probably* noise
(12:40:57 PM) sgallagh: And if it's revealing something you *should* be trying
to achieve, then it makes sense to go through the revision process to include it
(12:41:00 PM) sgallagh: (IMHO)
(12:43:47 PM) smooge: school calling. i am on phone
(12:44:47 PM) sgallagh: smooge: Unless you have other questions, I think we're
done for now.
(12:45:00 PM) sgallagh: Do you mind if I post this discussion to the server@
list for anyone else who is interested?
(12:45:19 PM) smooge: no problems on my part
(12:45:35 PM) sgallagh: OK, thanks
I was hoping we'd have more movement on the PRD/KLM discussion here on the list,
but since it hasn't happened, I think we should move the Logic Model discussion
over to the meeting tomorrow.
Any other topics for this week's meeting?
Reminder: the meeting time has moved five hours later in the day, to 4pm US/Eastern.
We've talked about this for a while, but let's make it formal. The plan
is to transition from Cloud as a Fedora Edition to Something Container
Clustery (see https://fedoraproject.org/wiki/Objectives/ProjectFAO).
But, we still need cloud as a _deploy target_. The FAO-container-thing
will continue to have cloud image deploy targets (as well as bare
metal). I think it makes sense to _also_ have Fedora Server as a cloud
This could possibly be both a Fedora Server Minimal Cloud Image and
Fedora Server Batteries Included Image — but that'd be up to Server WG,
Overall, I'm proposing:
1. Dissolve Cloud WG (See below; don't panic)
2. Form new Atomic WG or FAO WG (name to be bikeshedded)
(a lot of overlap in membership with current Cloud WG, of course!)
3. _Keep_ Cloud SIG as a gathering point around cloud technology and
covering shared underlying technology (fedimg, koji cloud image
production, autocloud). Think of this as analogous in some ways
to something like the ARM SIG.
4. Change https://getfedora.org/cloud/ to
https://getfedora.org/atomic/ or https://getfedora.org/fao/
5. Create new http://cloud.fedoraproject.org/ in the same style as
6. New Atomic/FAO WG produces Whatever New Deliverable (starting
with Two Week Atomic)
7. Cloud Base Image becomes base (uh, see what I did there?) for new
Fedora Server cloud image (or images).
8. Vagrant image _probably_ the same — or maybe becomes its own
Fedora Project Leader
#fedora-meeting-1: Server Working Group Weekly Meeting (2016-08-23)
Meeting started by sgallagh at 15:00:55 UTC. The full logs are available at
* Roll Call (sgallagh, 15:01:15)
* Agenda (sgallagh, 15:05:12)
* Agenda Item: Open Seats (sgallagh, 15:05:12)
* Agenda Item: New Meeting Time (sgallagh, 15:05:12)
* Agenda Item: Fedora 25 Talking Points (sgallagh, 15:05:12)
* Agenda Item: Server SIG PRD (sgallagh, 15:05:12)
* Open Seats (sgallagh, 15:06:36)
* danofsatx and hanthana did not respond to queries to retain
their seat on the Server Working Group (sgallagh, 15:07:26)
* AGREED: smooge (Stephen Smoogen) is elected to serve on the Server
Working Group (+5, 0, -0) (sgallagh, 15:16:48)
* LINK: https://fedorahosted.org/dell-config-cluster/ and
https://fedorahosted.org/dell-satellite-sync/ though Dell
has now taken them both over (vvaldez, 15:21:02)
* LINK: https://github.com/vvaldez (vvaldez, 15:22:22)
* AGREED: vvaldez (Vinnie Valdez) is elected to serve on the
Server Working Group (+5, 0, -0) (sgallagh, 15:27:54)
* New Meeting Time (sgallagh, 15:28:04)
* Proposal: Server SIG meeting time moves to Tuesdays at 4pm
US/Eastern (following US DST) (sgallagh, 15:31:53)
* AGREED: Server SIG meeting time moves to Tuesdays at 4pm
US/Eastern (following US DST) (+6, 0, -0) (sgallagh,
* This meeting time goes into effect on 2016-08-30 (sgallagh,
* Fedora 25 Talking Points (sgallagh, 15:37:14)
* Talking Point: Fedora Server 25 now ships Node.js 6.x
says no (sgallagh, 15:43:25)
* Cockpit in Fedora 25 now has an SELinux Troubleshooter
module (sgallagh, 15:44:29)
* Cockpit in Fedora 25 can display host SSH keys in the system
dashboard (sgallagh, 15:44:54)
* Cockpit in Fedora 25 will include support for network
teaming, docker volume and storage management as well as
the creation of systemd timer units (sgallagh, 15:46:35)
* DOCKAH DOCKAH DOCKAH (adamw, 15:46:38)
* Cockpit will also support multi-step (incl. 2FA)
authentication in Fedora 25 (sgallagh, 15:47:05)
* LINK: http://dockah.dockah.dockah.com/ (mhayden, 15:47:14)
* Fedora 25 will ship with Docker 1.12 (sgallagh, 15:49:59)
* Non-server-specific: Fedora 25 will remove support for
weaker (1024-bit) certificate authorities (sgallagh,
* FreeIPA may or may not deliver version 4.4 in Fedora 25.
Needs discussion as it missed Alpha. (sgallagh, 15:51:25)
* Server SIG PRD (sgallagh, 15:56:47)
* Out of time in the meeting, will be discussed on the
server(a)lists.fp.o mailing list (sgallagh, 15:57:02)
* Open Floor (sgallagh, 15:57:06)
Meeting ended at 16:00:28 UTC.
Action Items, by person
People Present (lines said)
* sgallagh (102)
* zodbot (19)
* vvaldez (19)
* mhayden (15)
* adamw (14)
* mjwolf (9)
* nirik (8)
* bconoboy_ (3)
* smooge (0)
* dperpeet (0)
* jds2001 (0)
* danofsatx (0)
* hanthana (0)
Generated by `MeetBot`_ 0.1.4
.. _`MeetBot`: http://wiki.debian.org/MeetBot
First order of business will be addressing the open seats; as of right now,
neither Dan Mossor (danofsatx) nor Danishka Navin (hanthana) have replied to the
"continued interest" thread, so it is likely that both of their seats will be up
At present, we have at least one self-nomination for an open seat: Stephen
Smoogen (smooge). It was unclear to me whether any of the other respondents to
the WhenIsGood were self-nominating.
Once the seats are decided, we will pick our new meeting time. Right now, the
optimum time would appear to be 1500 US/Eastern (1900 UTC) on either Tuesdays or
Thursdays which would accommodate everyone who replied except for linuxmodder
I'll express a slight preference to taking the Tuesday slot, as it gives us a
last chance to discuss things before Go/No-Go meetings at the various Fedora
If anyone else would like to raise an agenda item for tomorrow's meeting, please