IRC Log FDSCo 15 Nov 2005

[Stuart Ellis]

(21:13:54) quaid: <meeting>
(21:14:00) stickster: Oh, I was going to say Salma Hayek, but that's
*SO* much more interesting ;-D
(21:15:49) stickster: Hi everybody, including the recently deceased
(21:16:18) quaid:
http://fedoraproject.org/wiki/DocsProject/FedoraDocsSchedule
(21:17:54) quaid: sorry, back to back meetings for hours now
(21:18:05) quaid: the back-up of tasks is killing me :)
(21:18:33) quaid: so, pardon me if I just roll down the list of tasks
here ...
(21:18:47) quaid: I don't have any update on the Java needs, I have to
push some people on that
(21:20:43) G2: hI all
(21:21:15) quaid: howdy
(21:21:20) stickster: hi G2 
(21:21:38) G2: What's been happing over the past 2 weeks then
(21:22:11) quaid: stickster: Sopwith wants a single script he can run
that will build all of our stuff from CVS; is that easy enough? Can you
write that and put it in docs-common/bin ?
(21:23:05) quaid: that is what is holding up the DocsRawhide
(21:24:16) stickster: Hmm
(21:24:36) stickster: Well it would have to have some sort of
declaration at the top for which docs are to be built, but I suppose
that's not too hard
(21:24:59) Sopwith: I am working on getting shell accounts set up, so
you may be able to do the entire thing by yourself by the time I'm
finished.
(21:25:07) ***Sopwith works diligently to enable others to do the work
for him. ;)
(21:25:26) quaid: ah!
(21:25:26) stickster: commendable, sir
(21:25:31) quaid: that's what that was about
(21:26:47) quaid: let's see then ...
(21:27:20) quaid: Sopwith: well, our script would be in CVS, or are you
suggesting we can do the DocsRawhide stuff ourselves?
(21:27:51) Sopwith: I'm suggesting it's quite possible, yes, but let's
get the script written first and then we can touch base.
(21:27:56) quaid: ok
(21:28:13) quaid: stickster: sound ok?
(21:28:21) quaid: or shall we ask Tommy if he can?
(21:28:54) quaid: meanwhile ... moving on in parallel
(21:29:15) quaid: last week I got reminded of the power of personal
selling to make something successful.
(21:29:36) quaid: I'm wondering what you all think about reinvigorating
our one-to-one connections with people who have self-intro'd.
(21:29:56) quaid: considering the short hands today, perhaps a
discussion on-list?
(21:30:25) elliss: Yes.
(21:30:28) stickster: Yes, with the suggestion that everyone "claim"
some contributors by putting their mark next to the contributors' names
on a wiki page
(21:30:30) G2: Yes.
(21:30:39) G2: I think the list is up2date
(21:31:28) elliss: It may be a good idea to co-ordinate what we want to
say, though.
(21:31:38) stickster: G2: yes, you were pretty darn diligent about it,
and caught up quickly when needed
(21:31:50) stickster: there was something in the archives methinks
(21:32:11) stickster: It was floated, approved (or at least nothing
negative came up), and then promptly forgotten
(21:32:19) stickster: in the rush of other things certainly
(21:32:48) quaid: ok, I am working on a msg to the list to reinvigorate
(21:32:56) G2: Guys.
(21:32:57) quaid: I think we decided to see what an email blast would do
(21:33:06) G2: The list is BIG
(21:33:09) quaid: and the think I learned recently is, the email blast
is cheap to do, and everyone knows that.
(21:33:14) stickster: G2: ?
(21:33:20) G2: If we can get just a 1/4 of them to actually move on
their offer
(21:33:28) G2: We will have a huge team
(21:33:58) quaid: yep
(21:34:06) quaid: still though, trans kicks everyone's but
(21:34:13) quaid: Sarah said last week almost 3000 contributors
(21:34:22) elliss: I can't see anything in my fdsco archive on this
(21:34:46) quaid: we'll eventually need to get into the kinds of tools
they use in trans, e.g.
(21:34:54) elliss: Did we agree some text, or just to send a mail ?
(21:34:58) quaid: http://fedoraproject.org/wiki/DocsProject/Contributors
(21:35:06) quaid: elliss: I'll propose some text you can customize
(21:36:27) mrj left the room (quit: Remote closed the connection).
(21:36:49) quaid: ok, that will come to the list in a few minutes
(21:36:53) ***quaid stops writing and continues meeting
(21:37:03) BobJensen: lol
(21:37:10) quaid: ooh ooh
(21:37:13) quaid: the next one is cool
(21:37:32) ***quaid looks for URL
(21:38:38) quaid: http://wiki.linux.duke.edu/TestDB
(21:38:53) quaid: I'm also doing some tests at
http://wiki.linux.duke.edu/DocbookTest
(21:39:16) quaid: it's straight XML with a #format docbook header
(21:39:26) quaid: immediately I ran into challenges
(21:39:27) quaid: entities
(21:39:44) quaid: the xslt processor barfs when it can't expand the
entities
(21:40:07) quaid: I am trying to figure out how to do it elegantly with
the entities as separate Wiki files that get called in, but I don't have
this figure out yet.
(21:40:17) quaid: but it's a further step, anyway
(21:40:40) stickster: Pretty cool... so DB could be imported straight to
the Wiki and ACL'd?
(21:40:53) stickster: Or not
(21:41:02) quaid: yeah
(21:41:08) quaid: it's the two-way street that I want to figure out
(21:41:35) quaid: we need this functionality internally, and I'm trying
to convince all the right people to join efforts with fp.org
(21:41:41) elliss: You may have to twiddle the Python code to get
multi-page documents
(21:41:52) quaid: it's not as quick and dirty as the Doc-Book Wiki
stuff, but it's the Right Way
(21:42:15) quaid: elliss: yes, I expect we'll have to patch further and
often run ahead of the upstream.
(21:42:46) G2: You guys heard of AxKit
(21:42:51) quaid: nope
(21:43:08) G2: http://axkit.org/
(21:43:15) G2: Just had a new release
(21:43:29) stickster: AxKit relies on XML source, right?
(21:43:48) G2: aye.
(21:43:52) stickster: I thought we were aiming for people being able to
contribute in wikispeak, and have some conversion done *to* XML, maybe
before storage
(21:44:07) stickster: Or let the user set a pref for which way they
prefer to write?
(21:44:08) stickster: Dunno
(21:44:17) quaid: flexibility is good
(21:44:27) quaid: but contribute in wikispeak, yeah, that is key
(21:44:43) stickster: Right, trying to remember LBE (low barrier to
entry)
(21:44:54) G2: yeah. i tmight be just a case of sacrificing some
formatting just now
(21:44:59) quaid: I didn't discourage skvidal however :)
(21:45:24) elliss: G2: Yes.
(21:45:40) quaid: the Wiki2XML works, in one direction only, and now
it's parsing direction from XML to Wiki, so that's another piece.
(21:45:51) elliss: MoinMoin wikispeak won't map to DocBook
(21:45:53) quaid: Wiki is not our formatted output
(21:46:04) elliss: Except at a very simplistic level
(21:46:28) BobJensen: Wiki scares a lot of people
(21:46:38) quaid: elliss: are we going to lose something for every
translation, even if people only change content and not markup details?
(21:46:49) quaid: BobJensen: jiminey! what more can we do? :)
(21:47:05) stickster: Yeah, we've been through this exhaustively already
(21:47:16) quaid: right now we want writers who can use a Wiki
(21:47:17) stickster: time to pull a gregdek and say, let's move forward
(21:47:20) BobJensen: quaid: the whole wiki formatting is just odd but
it works
(21:47:52) quaid: BobJensen: I'm not a big fan at'all, well documented
fact that, but it's what the people seem to want.
(21:48:03) elliss: I don't think that we can't produce a complete doc in
Wikispeak, just the text
(21:48:07) quaid: and if we can get developers contributing that much
more because it's easier ... so be it.
(21:48:18) BobJensen: quaid: I agree, did not mean to derail things
(21:48:32) quaid: well,it's a point :)
(21:48:42) quaid: elliss: how about for 'just edits'
(21:48:56) quaid: like, a feature that lets you pull a single page and
it's entities to edit,then puts it back
(21:49:07) quaid: v. having an entire book exist as a real, all-the-time
Wiki structure
(21:49:17) elliss: The problem is that formatting won't map, nor links
(21:49:43) quaid: I really need to get an instance of D-B Wiki running
to see how it handles this stuff.
(21:49:44) BobJensen: Using the CMS system I have for FedoraLinks.org
the user if presented with a OOo writer interface
(21:49:53) BobJensen: s/if/is
(21:50:15) quaid: OO.org has some traction, yes, and should be another
choice
(21:50:44) G2: openoffice?
(21:50:55) elliss: Yes. 
(21:51:04) elliss: There's a page on the Wiki
(21:51:33) quaid: some sort of Web based lock file in module, download,
edit, upload, unlock
(21:51:34) elliss: But like Wiki you have to beat up the results to get
DocBook as we would like it
(21:51:43) quaid: edit = use your fave tool, OO.org, Emacs, etc.
(21:51:45) BobJensen: Sonar_Guy and are of the opinon, get us the
content and we will help submit it no matter how big or small
(21:51:49) elliss: E.g. words replaced by entities
(21:52:39) elliss: Bob: our big problem is getting two-way
(21:53:06) BobJensen: elliss: I understand that
(21:53:51) quaid: oh, one more thing to discuss before we go
(21:54:08) quaid: some marketing folks are seriously redoing all the
content on f.r.c, and cross-comparing to fp.o
(21:54:09) quaid: this week
(21:54:34) quaid: this may accelerate the move from f.r.c to fp.o,
running the same CVS to PHP back-end somehow.
(21:54:36) stickster: I assume that's one of the reasons for the
DB->wiki presentation thingie
(21:54:38) quaid: I need to talk with Sopwith about that
(21:55:11) quaid: anyway, I'm helping get content changes posted to
f.r.c than passing all that off for trans.
(21:55:35) quaid: this work can happen in parallel whilst we discuss
moving from f.r.c to fp.o
(21:57:01) stickster: quaid: Abour relnotes publishing
(21:57:26) G2: I take it we've reviewed:
http://meta.wikimedia.org/wiki/DocBook_XML_export and
http://www.tldp.org/wt2db/
(21:57:31) stickster: quaid: How is this normally handled? Can I just
make a fc5/ dir, make html in CVS and then do the normal publishing
work?
(21:58:06) quaid: we are eschewing the past
(21:58:10) quaid: and using tagging
(21:58:15) quaid: so no, no new subdirectories
(21:58:25) quaid: all is in release-notes/ now and is tagged for trans
and release.
(21:58:39) quaid: or are you talking about f.r.c?\
(21:58:41) quaid: sorry, you are
(21:58:44) stickster: Yes, f.r.c
(21:58:53) quaid: never mind ...
(21:59:06) quaid: yeah, just the same thing as before, I reckon , unless
it needs improving.
(21:59:12) quaid: there are languages going on, too
(21:59:14) stickster: I'm looking at
fedora-cvs/web/html/doc/release-notes/
(21:59:18) BobJensen: My Xorg Beat is hitting rawhide tomorrow barring
any breakages where do I submit the release notes doc or did I miss the
deadline for test1?
(21:59:38) quaid: BobJensen: you are still good, just get them in asap
(21:59:46) BobJensen: quaid: will do
(21:59:52) G2: brb
(21:59:56) quaid: we're really holding for xorg right now across the
board
(22:00:10) BobJensen: that is what I thought
(22:00:18) stickster: so can I just make a fc5/ dir, and then ru/ zh_CN/
etc.?
(22:00:49) quaid: sorry, is that what we did for fc4?
(22:00:50) quaid: then, yeah
(22:01:10) quaid: no one has complained about the structure
(22:01:45) BobJensen:
http://mharris.ca/xorg-modular/xorg-modularization.html will vanish once
it reaches rawhide
(22:01:45) stickster: cool
(22:02:00) stickster: Just tell me what date we need to do the publish
(22:02:48) quaid: publish to web is with the release, right now 21 Nov.
(22:03:09) quaid: build freeze for ISO is supposed to be today, but
we're making it tomorrow or another day later.
(22:03:14) quaid: totally mushy right now, sorry
(22:03:35) quaid: stickster: let's save the trigger pull for tomorrow
night ... we still need to get changes from the Wiki, as well.
(22:05:19) stickster: quaid: do you mean you need help turning wiki beat
notes into relnotes?
(22:05:43) quaid: oh, no, not neccesarily ...
(22:05:51) quaid: I think nman64's stuff still works for now
(22:06:01) quaid: just letting you know when we need to have stuff ready
by for whatever, that's all
(22:06:13) quaid: and to say, if you were worried like I was about
freezing for ISO today ... don't be :)
(22:06:37) quaid: last items from me :
(22:06:56) quaid: I gave a verbal pledge to Sarah Wang on behalf of FDP
for Fedora Trans that our freezes will never defrost.
(22:07:04) stickster: I wasn't that worried... but I am out of it... I
have been very much overscheduled with other commitments and have had
very little time for FP work lately
(22:07:06) quaid: they get that all the time from devel, and we need to
stomp on that idea.
(22:07:39) quaid: otherwise, we're all doing OK with the current temp
process.
(22:07:46) quaid: and that's all of our priority 1 items
(22:07:55) quaid: AOB?
(22:07:56) quaid: or </> time?
(22:08:27) G2: I was trying to get my Sprog and supporting RPMs ready
for that freeze too. But Sprog has suddenly developed bugs
(22:09:44) stickster: Vote </>
(22:09:55) G2: yup
(22:09:56) elliss: </>
(22:10:20) quaid: </meeting>

-- 

Stuart Ellis

stuart at elsn.org

Fedora Documentation Project: http://fedora.redhat.com/projects/docs/

GPG key ID: 7098ABEA
GPG key fingerprint: 68B0 E291 FB19 C845 E60E  9569 292E E365 7098 ABEA
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://listman.redhat.com/archives/fedora-dsco-list/attachments/20051115/616dbdbd/attachment.sig>