05:25:06 mstone: incidentally...
05:25:07 mstone: https://www.redhat.com/archives/fedora-olpc-list/2009-October/msg00000.html
05:25:37 mstone: (since I know you care about hearing when people get
diverted from actual contribution but Dumb Administrivia)
05:28:45 mchua: Ah, thanks! Yeah, I'll get out my Dumb Administrivia bludgeon
and see what we can do.
05:29:23 mstone: :)
05:29:54 mstone: it's not a huge deal, but I thought you'd appreciate having
something concrete to point at.
05:30:41 mstone: mchua: though, come to think of it, I didn't even mention
the hoops I had to jump through before hitting that wall
just in order to try...
05:30:57 mstone: I just jumped, because I knew that I wouldn't block on any
humans.
05:31:18 mchua: I do - thanks! Yeah, hacking people-systems is like hacking
code in that tracebacks are helpful for bug advocacy.
05:31:41 mchua: hrm. Is it a process I should try to walk through and document
the pain points of?
05:31:42 mstone: but, in fact, I had to track down an old password for rhbz,
generate a new one, search for related tickets [which was so
slow that I gave up and just filed], filed the ticket,
05:31:47 mchua: (how bad/common do you reckon they'd be?)
05:31:56 * mchua listens
05:31:57 mstone: I'll get to that in a minute
05:32:14 mstone: then, after I filed the ticket, I thought to myself: "gee, I
know this is really simple to fix. why don't I just go fix
it?"
05:32:39 mstone: so I had to figure out what packages to install (remember, I
don't have fedora installed on my machine!)...
05:32:51 mstone: turned out that I needed "fedora-packager", which actually
worked great once I found it.
05:33:35 mstone: then I had to find the "updating cvs instructions" [which at
least I knew about!], renew my koji cert, find the
appropriate ssh key, do the cvs checkout, write the patch...
05:33:43 mstone: and then I couldn't commit it. :)
05:34:24 mstone: most people, I imagine, use their packaging credentials
regularly and so don't need to renew/find them all at once
like this.
05:34:44 mstone: also, most people who do fedora packaging probably do it on
fedora machines, rather than in fedora chroots on debian
machines.
05:35:08 mstone: but still -- this is an awful lot of work just to provide a
4-line patch.
05:35:22 mchua: I've been banging my head against the Barrier To Packaging And
Patching lately.
05:35:42 mstone: (and then, of course, there was the test environment
generation and the actual diagnosis, solution, and testing
before this!)
05:36:00 mchua: I've come to believe that most people who do fedora packaging
probably do it on a regular basis because the people who don't
have given up because there's just so much *stuff* to do.
05:36:12 mchua: s/stuff/overhead - that's a better word.
05:36:37 mstone: mchua: git, bzr, and mercurial have a pretty good idea in
this department, which is that they make it *trivial* to
send patches via email
05:36:59 mstone: git format-patch -1 | git send-email --to foo@bar.com :)
05:37:08 mstone: no credentials required.
05:37:54 mstone: now, I'm definitely not going to claim that this is the
right solution for packaging, but it should at least be
thought-provoking.
05:39:11 mchua: +1. mind if I share this convo with some folks who might be
able to Make It Better?
05:39:22 mstone: oh, please feel free.
05:39:25 mchua: both your sugar patch in specific, and the packaging
experience in general.
05:39:30 mchua: ok. I'll cc you on that.
05:39:39 mstone: sounds good.
05:40:16 mstone: as I said -- I don't want to make a big deal out of it; I
simply want to make the minimal amount of noise necessary to
give people who care about this sort of thing one more
"spark" to add to their measurements.