Sorry for the last message, here's the proper thread title for archive
12:13 < quaid> ok
12:13 * stickster just updated page with his item
12:13 < stickster>
12:13 < stickster> crud, sorry
12:13 < quaid> the first is easy
12:13 < stickster>
12:13 < quaid> there isn't a GSoC update; aiui the deadline for entries was
extended to tomorrow
12:13 < stickster> Ah OK
12:13 < quaid> and I don't have a view yet into what has been proposed
12:14 < glezos> and I don't have a decision yet whether i'll participate as
12:14 < quaid> but it does look like, except for FDP, Fedora didn't get as many
ideas up on pages for people to propose against
12:14 < quaid> kind of our usual, "How about you dream up something you'd
like to do?"
12:14 < stickster> *sigh
12:15 < stickster> I think the Board will need to start proactively dealing with
this better in 2008
12:15 < quaid> hey, you know, if all we had to do was put out GNOME or something,
then maybe we'd have time for niceties :)
12:15 < stickster> And to do that, we need a better planning system. I think there
was an unfortunately short thread on a better project-wide calendaring system in one of
12:15 < glezos> stickster, right
12:16 < quaid> ok, we can move on
12:16 < stickster> All right, my item next, yes?
12:16 < stickster> So, obviously if you look at f-docs-commits you'll see
I've spent a bunch of time in Makefile.common recently
12:17 < stickster> We are now set up so people can check out single documents as a
module, and automatically get docs-common included.
12:17 < quaid> but this is docs-common at ,/ v. ../?
12:17 < stickster> Things should Just Work(TM) regardless of whether you do that, or
mirror the whole repository down
12:17 < stickster> gotcha!
12:17 < quaid> :)
12:18 < quaid> ok, so you dealt with the concern about multiple docs-common copies/
12:18 < stickster> You'll see the "find-common-dir" near the top (or
some name like that, can't recall off the top of my head)
12:18 < stickster> It looks in current dir first, then parents up, then looks for a
system dir in case we ever get around to packaging it
12:18 < quaid> ok
12:18 * quaid kicks CVS for not supporting symlinks
12:18 < stickster> Boo-ya!
12:18 < stickster> Nah, symlinks bad for people using Windowsw
12:18 < stickster> s/sw$/s/
12:19 < quaid> ?
12:19 < quaid> and the rest of it works on Windows?
12:19 < glezos> stickster, :)
12:19 < stickster> Well, cygwin
12:19 < quaid> well, as long as you got it to work
12:19 < stickster> OK, and...
12:20 < quaid> If it had to be a choice, cygwin would lose out
12:20 < stickster> sure
12:20 < stickster> But I think symlinks are generally a bad way of doing things
12:20 < stickster> fragile
12:20 < stickster> So...
12:20 < stickster> 2. entities
12:20 < stickster> The entities files originally in docs-common/common/ are now gone
12:21 < stickster> They were useless anyway, a pain for translators, and mostly
12:21 < stickster> All system wide entities have migrated to
docs-common/common/entities/ using PO, the way that God intended
12:21 < stickster> Next up, I'm going to do the same for docs-common/common/
snippets like bugreporting, legalnotice, etc.
12:21 < quaid> ah, good, ok
12:22 < stickster> hang on , brb
12:22 * quaid saw cruft and didn't know which was cruft and which was gold
12:29 * quaid gets distracted in EPEL SIG meeting
12:29 < quaid> I got the new CVS bits and didn't have any questions so far
12:29 < quaid> but this is good because it simplifies the l10n process
12:30 < quaid> should we have the QSTG updated?
12:31 < quaid> or maybe wait until things settle down some more?
12:32 < glezos> dunno
12:33 < stickster> back
12:33 < glezos> quaid, I'd say that we could wait to get it done right
12:33 < glezos> It's 2 months more
12:34 < stickster> It won't be nearly that long before I have everything done
12:34 < stickster> In the next few days, after I'm done, I'll update it
12:34 < stickster> *: I also want to bring up the whole "xml2po -e" vs.
"xml2po -k" thing
12:34 < stickster> I think we're better off with the latter.
12:35 < stickster> We tried it before but it didn't work because of the stupid
stuff in docs-common/common/ (i.e. snippet inclusion without XInclude)
12:35 < stickster> I think we can safely go back to keeping entities in the POT/PO
12:35 < stickster> That means that translators will simply translate, e.g.
&DISTRO; as &DISTRO;
12:36 < stickster> If we update one of them, like the Fedora release number, the
translation doesn't immediately catch "the fuzzies"
12:36 < glezos> stickster, whatever you believe it's better
12:36 < stickster> Yeah, we tried to do it this way before, and fell back because we
couldn't make it work
12:37 < stickster> Now we have more XML savvy and fewer crazy things we're
trying to do, so things like that should now be possible again
12:38 < quaid> ok, cool
12:38 < quaid> stickster: good work, padewan
12:38 < stickster> Oh, one more thing
12:39 < stickster> I'm putting per-branch aliases in CVSROOT/modules, so you can
now check out, e.g. "install-guide-devel" to get just the devel branch. It cuts
down on bandwidth as well
12:39 < stickster> So, next up:
12:39 < stickster> 1. fixing snippets => XML+PO
12:39 < stickster> 2. fixing graphics => SVG+[tools]
12:40 < quaid> GuGu: we're meeting here for the next 20 minutes (or so); do you
want to save your question(s) until then, then we can talk here?
12:40 < quaid> stickster: sorry, what are snippets?
12:41 < stickster> quaid: bugreporting-*.xml, legalnotice-*.xml, etc.
12:42 < stickster> all found in docs-common/common/ right now
12:42 < quaid> ah, right, those
12:42 * quaid forgot the special term 'snippet'
12:42 < quaid> ;-D
12:43 < stickster> OK, I'm done too
12:43 < stickster> couf_away is still away, so we can safely skip his item
12:44 < glezos> I have a suggestion
12:45 < glezos> that might urge translators get more Docs translated for F7
12:45 < glezos> if we had on the web some l10n statistics for cvs.fp.org
, that might
urge them to push their translation too
12:46 < quaid> hmm
12:46 < quaid> can we do that semi-manually?
12:46 < stickster> boo, hiss
12:46 < quaid> that is, write some Perl that dumps stats to SVG?
12:46 < stickster> heh
12:46 < quaid> err
12:46 < glezos> Sonar_Gal, would it be a good idea to ask FIP for a place to run the
12:46 < quaid> Python I meant
12:46 < stickster> Poor Sonar_Gal
12:47 < quaid> her fault, if she chose Sonar_Gyrl, then Sonar_Guy would get all the
12:47 < glezos> quaid, we could do a cron job to run this script and publish the
results.. they are not pretty, but they are usable... Especially if you run them through a
search-n-replace script that transforms them into an HTML table...
12:47 < quaid> +1
12:47 < quaid> at least that
12:48 < quaid> maybe an update hourly (or at :20 or something)
12:48 < glezos> quaid, should we ask FIP for l10n.fedoraproject.org?
12:48 < stickster> And translation.fp.o
12:48 < quaid> as a CNAME?
12:48 < stickster> whatever
12:49 * stickster doesn't trust people to guess "l10n"
12:49 < quaid> ok, whatever, I mena, as a pointer v. stand alone host
12:49 < stickster> Oh yeah
12:49 < quaid> then I want documentation.fp.o !
12:49 < quaid> :D
12:49 < glezos> ok.. we could standarize on translations.fedoraproject.org
it from there.. I guess this is something -trans-list should decide
12:49 < stickster> +1!
12:49 * quaid is all about the "docs is unprofesional, spell it out" :)
12:50 < quaid> glezos: yes, propose and see what they prefer, just as we did for
docs.fp.o and then we'll put up what we have as we hae it
12:52 < glezos> k. then I'll just send an email to FIP
12:53 < quaid> cool, or file a ticket
12:53 < quaid> if there is nothing to discuss, just request
12:54 < quaid> (http://admin.fedoraproject.org/tickets
12:54 < quaid> and there we go
12:54 < quaid> anything more here?
12:55 < glezos> nah
12:55 < stickster> nope
12:55 < quaid> heh, no quorum even
12:55 < quaid> rockin', thx getns
12:55 < quaid> gents even
12:55 < quaid> </meeting>
Paul W. Frields, RHCE http://paul.frields.org/
gpg fingerprint: 3DA6 A0AC 6D58 FEC4 0233 5906 ACDB C937 BD11 3717
Fedora Project: http://fedoraproject.org/wiki/PaulWFrields
: stickster @ #fedora-docs, #fedora-devel, #fredlug