On Thu, 2004-08-12 at 11:14, Dave Pawson wrote:
> Have you already looked at my updfstab-tutorial? I think it
> better technical solution that works with the existing subsystems
> (hotplug, kudzu) rather than creating a need for another one (autofs in
> this case). I like autofs too, but I normally use it only for things
> like NFS or SMB mounts, since it doesn't have great support in the
> spatial browser. Mounts supported by updfstab will appear in the
> "Computer" browser, which makes the user experience a little more
> Please check out the current version at my site, http://docs.frields.org
> ... there's a HTML-browseable copy for convenience. I've been working on
> intergrating Tim Waugh's guidelines for troubleshooting hotplug problems
> such as missing USB vendor/product information.
1. I'm in no position to judge either as being better / worse.
2. Any reason to have more than one alternative?
Are you (is anyone) in a position to say this is better/easier if
3. I do actually need / want such doco myself, since I use a flash card
in a reader :-) (And I did it a.n.otherway :-)
Anyway Paul, I was first to get it to the list :-) So there
<grin/> For 1 & 2, I guess I meant "better" in the sense that the
hotplug/updfstab method (1) is a way that both makes a more consistent
user interface (see comments above), (2) does not depend on the hardware
configuration remaining constant between insertions (i.e. what happens
if next time you have one or more other USB storage devices plugged in
already?), and (3) allows the user to contribute back to the community
by entering a bug against either hotplug or kudzu for their particular
USB device, so others can benefit.
As for your flash card, I'm genuinely curious: Does your card reader not
allow the card to report its identity using "lsusb -v" or dmesg? If it
does, the process in updfstab-tutorial will work. If not, I'd love it if
you would try out the process in the updfstab-tutorial and let me know
where it breaks so I can fix it. :-)
Q.v. my BZ entry from 8/1:
Having said all that, there's always more than one way to skin a cat!
Perhaps we could combine these in such a way as to make the whole thing
palatable to a broader audience? It seems to me a single document that
gives multiple methods is better than finding several partly or wholly
inconsistent documents, each professing to be the One True Method(TM).
Paul W. Frields, RHCE