IRC log of htmlt on 2012-01-17

Timestamps are in UTC.

16:02:31 [RRSAgent]
RRSAgent has joined #htmlt
16:02:31 [RRSAgent]
logging to
16:03:10 [krisk]
If someone would like to dial into the conf call please speak up
16:04:55 [krisk]
16:06:21 [krisk]
Looks like no one else will be joining the session
16:06:59 [krisk]
Item #1 Bugs on Approved Tests
16:07:06 [krisk]
link ->
16:07:26 [krisk]
No new bugs opened
16:08:25 [krisk]
No updates on bug # 15041/15042
16:10:52 [krisk]
Agenda Item #2 New Test Submissions
16:12:23 [krisk]
A few more Bidi Tests have been submitted
16:12:43 [krisk]
16:15:18 [krisk]
Ms2Ger also submitted another dataset test
16:15:23 [krisk]
16:16:00 [krisk]
16:16:16 [krisk]
and fixed a bug in apples dataset tests (delete)
16:17:55 [krisk]
jgraham have you had a chance to look at the History api tests?
16:23:07 [jgraham]
No, the only thing of note I have done is made an experimental clone of the testsuite repo. on github
16:23:21 [jgraham]
To try out their code review facilities
16:23:36 [jgraham]
Worked OK for a little patch I made to testharness.js
16:23:48 [krisk]
16:23:58 [jgraham]
I haven't got the syncronisation working perfectly yet though
16:24:27 [krisk]
do you have any pointers to this experiment?
16:24:41 [jgraham]
16:24:58 [Ms2ger]
Speaking of which, setup({ "output": false });, right?
16:25:26 [jgraham]
16:32:35 [krisk]
How does the scynronisation error arise?
16:35:40 [jgraham]
Well the w3c server is hg and this is git and I have to keep them in sync using hg-git. And I accidentially pushed to git without pushing to hg. Then by the time I did push to hg there were more commits
16:35:56 [jgraham]
So I had to rebase (or I could have merged)
16:36:19 [jgraham]
Anyway, it is always possible to avoid this
16:36:27 [jgraham]
Just by having the right workflow
16:39:10 [jgraham]
(AIUI there are people that hack on mozilla using git locally and keeping in sync with the canonical copy in mercurial without substantial problems)
16:39:22 [jgraham]
(so it is clearly a solved problem)
16:40:27 [Ms2ger]
I don't think they actually land on the git tree
16:40:54 [krisk]
I have not used hg-git
16:44:32 [krisk]
any other items to discuss or shall we adjorn?
16:45:05 [Ms2ger]
Not from me
16:47:26 [jgraham]
No, I don't have any topics
16:48:28 [krisk]
Meeting Adjorned
16:48:36 [krisk]
RRSAgent, make logs public
16:52:59 [jgraham]
Ms2ger: Yeah, it looks like the right workflow is never to commit to master in the git repo, only to ever pull from hg to master
16:53:10 [Ms2ger]
Sounds right
16:53:21 [jgraham]
So you work with git on some branch (if you want). Then push to hg. Then pull to git.