Status Meetings:2007-02-28:Log

From Camino Wiki
Jump to navigation Jump to search
12:03  * ardissone|food waits on our fearless leader
12:04 <@ardissone|food> do we know it today was another scheduled no-pants day?
12:04 -!- pinkerton has joined #camino-mtg
12:05 <@ardissone|food> ok :)
12:05 <@mento> narrowly avoided foot-in-mouth disease there
12:05 <@ardissone|food> tell me about it :)
12:05 <@Wevah> hahaha
12:05  * ardissone|food is never using a mento-ism again
12:05 <@mento> hah
12:05 <@ardissone|food> ok, sam doesn't seem to be here-here
12:05  * smorgan smacks timeless in bugzilla
12:05 <@ardissone|food> and froodian is not
12:06 <@ardissone|food> heh
12:06 <@cl> neither one of sam is here, apparently
12:06 <@ardissone|food> so i guess it's my show today
12:06 <@ardissone|food> so
12:06 <@ardissone|food> everyone open 
               in their 
                        Camino 1.1b or 1.1b+ ;)
12:07 <@Wevah> what about my camino 1.2+?
12:07 <@ardissone|food> for those of you under rocks, or who cut out early on Friday, we released 
                        Beta Friday afternoon
12:07 <@pinkerton> yeah, i think right before i blogged about how it was coming soon ;)
12:07 <@mento> funny, i didn't see your blog :P
12:07 <@pinkerton> prolly for the best
12:07 <@ss> I saw it posted after we released. ;)
12:08 <@ss> So, looking through talkback, we've had quite a few reports
12:08 <@ss> ardissone has been investigating some of them
12:08 <@ardissone|food> mento: did jay say when reports would begin?
12:09 <@mento> he was in a firefox crunch yesterday, he said maybe last night but that didn't 
12:09 <@ss> I could ping him once today; we usually cross paths.
12:09 <@ss> There's still a few bugs left for 1.1, but we're looking really, really good
12:10 <@ardissone|food> just the hard ones
12:10 <@ss> One of our major crashers has a bug filed and that's being looked into.
12:10 <@ss> Hard smard. :)
12:10 <@mento> go go gadget smichaud
12:10 <@ss> Yeah, it was good to see him looking into it
12:10 <@pinkerton> heh
12:10 <@ss> Next up, we need to look at doing 1.0.4 soon.
12:10 <@ss> Very soon.
12:11 <@ss> Our last stable release on that branch was in September.
12:11 <@ardissone|food> mento: how's your week?
12:11 <@ardissone|food> :/
12:11 <@mento> 1.8.10 + minibranch + anything else?
12:11 <@ss> And there's a bunch of security releases that need to go out.
12:11 <@ardissone|food> the two fixes we always land on the minib, but everything else is in
12:11 <@ss> I don't think there's anything else.
12:11 <@ss> Yeah, everything should be in
12:12 <@mento> i guess i can spin it tomorrow.  slight chance that i can do it overnight.  we'll 
12:12 <@ss> Awesome
12:12 <@ss> The sooner the better. ;)
12:12 <@ss> tinderboxen suck (surprise)
12:12 <@ardissone|food> we should target "sometime this week" for release, so that marcello can 
                        get ML done
12:12 <@mento> yeah, i don't want to lag too much on this, there are a lot of gecko security fixes
12:13 <@ss> Yeah
12:13 <@ss> Going to bookmark loss... I don't think there's much new there.
12:13 <@mento> someone should look at the possible leak that's making pawn suck more than usual
12:14 <@ardissone|food> is it in tests, then?
12:14 <@ardissone|food> since boxset is fine just building?
12:14 <@mento> yeah, i'd put money on it happening during tp
12:14 <@mento> not every tp cycle, though, and some are worse than others.
12:14 <@ardissone|food> should we file a bug somewhere to track it?
12:15 <@mento> i don't know, how's "yes?" sound?
12:15 <@ardissone|food> where :)
12:16 <@mento> camino.general
12:16 <@ardissone|food> k
12:16 <@ss> Alright, moving right along...
12:16 <@ss> Our bug queues don't look too bad. The review queue is slowly filling up
12:16 <@ss> But not as bad as it's been
12:17 <@ss> And the sr queue is empty. :)
12:17 <@ardissone|food> or was last night
12:17 <@ss> It is now
12:17 <@ardissone|food> i think there's one bug
12:17 <@ss> I just looked.
12:17 <@ardissone|food> oh, pink got to it then  :)
12:17 <@ss> :)
12:17 <@ss> Now, the fun thing for the week.
12:17 <@smorgan> someone can steal that latest tracking rect patch review from josh if they want 
                 it, btw
12:17 <@ss> Everyone open
12:18 <@ss> We've gotten initial approval to use the MoCo headquarters for a meetup
12:19 <@ss> They'll cover a few of our expenses and also provide connectivity (and, iirc, a 
            conference line for us to use).
12:19 <@cl> does "snacks and beverages" include adult beverages? :-p
12:19 <@ss> Unfortunately, no. ;)
12:19 <@ss> I had "snacks and drinks" but thought I should change it. ;)
12:19  * cl wants some Mozilla Maibock
12:20 <@kreeger> im guessing that means a run to the local watering hole is in order
12:20 <@ss> The one thing we need to think about and make a decision on in the next week or two, 
            is a date.
12:20 <@ss> I initially suggested Memorial Day weekend.
12:20 <@ss> But I'm not sure how good that is for everyone else.
12:21 <@kreeger> i think desmond mentioned that he would be in CA, USA the first week or so in 
12:21 <@Wevah> i might actually have money then!
12:21 <@Wevah> it'll be awesome
12:21 <@pinkerton> maybe have it around wwdc
12:21 <@pinkerton> that seems most logical
12:21 <@pinkerton> i forgot it was so close
12:22 <@pinkerton> i doubt i could fly out at the end of may, then again out a week later for wwdc
12:22 <@mento> we could do it on the east coast
12:22 <@kreeger> WWDC is June 11-15
12:22 <@ss> Heh
12:22 <@cl> Yeah, Memorial Day weekend is probably a bad travel time for me
12:22 <@kreeger> or in KC
12:22 <@mento> :)
12:22 <@cl> Or Miami.
12:22 <@cl> Lots of fun things in Miami.
12:22 <@ss> So how about the weekend of the 8th
12:22 <@ss> It'd be the 8th through the 10th or so
12:23 <@pinkerton> it's multiple days?
12:23 <@ardissone|food> will pre-wwdc be really bad for smfr?
12:23 <@pinkerton> we can't have that much to talk about ;)
12:23 <@ss> pinkerton: We'd probably meet for two days, yeah.
12:23 <@ss> Really?
12:23 <@ardissone|food> (as opposed to post?)
12:23 <@ss> Planning out features?
12:23 <@cl> Other than "avoid big travel periods", I'm equally open (or not) pretty much whenever
12:23 <@cl> because I have absolutely no idea what my schedule might or might not be, or if I 
            could even be there.
12:24 <@kreeger> either of those dates would work for me (memorial or June)
12:24 <@pinkerton> we can do that weekend, or the weekend after wwdc
12:24 <@ardissone|food> pre might also have more hotels full?
12:24 <@pinkerton> i'll be in cali prolly for 2 weeks
12:24 <@ss> 16th and 17th?
12:24 <@ss> I'm not sure which weekend would be worse.
12:24 <@ss> (For hotels)
12:24 <@ss> But we also wouldn't be *in* SF, where WWDC is taking place.
12:25 <@pinkerton> true
12:25 <@ardissone|food> ah
12:25 <@ss> So... pick a weekend!
12:25 <@ss> Play some darts
12:26 <@pinkerton> i say after
12:26 <@pinkerton> it'll give us a chance to talk about what we see in leopard
12:26 <@pinkerton> and what we learn at the conf
12:26 <@kreeger> yeah, would be cool to hear some post-WWDC topics
12:26 <@ss> Sounds good
12:26 <@ardissone|food> yeah
12:26 <@pinkerton> yeah, that's why we're doing after at google for our mac team convergence. 
12:26 <@ss> So, we're tenatively planned for the 16th and 17th of June
12:26 <@pinkerton> sounds good
12:26 <@ss> I'll let them know so they can firm it up.
12:28 <@ss> Next up, some individual bugs
12:28 ! ss invited thebot into the channel.
12:28 -!- thebot has joined #camino-mtg
12:28 <@ss> bug 350331
12:28 < thebot> ss: Bug maj, --, Camino1.1, 
      , NEW, Pages sometimes fail to render (blank content) when 
                loaded; resizing the window fixes display
12:29 <@ss> I'm actually not sure what specific action we need there.
12:29 <@smorgan> Hopefully we can use the other version of the bug to find out what isn't hooked 
                 up right, and then work backward from there
12:29 <@ss> Ah...
12:29 <@ardissone|food> that's my hope, too
12:29 <@ardissone|food> that bug was much better when i ran the backout patch
12:29 <@smorgan> The specific action it needs is some understanding of what is happening
12:30 <@ardissone|food> for this week, i just dumped all the big/hard/etc bugs in here so we can 
                        get re-acquainted with them
12:31 <@ardissone|food> after our switch to strings stuff the last 2 weeks
12:31 <@ss> Ah, alright
12:31 <@ardissone|food> we can (mostly) run through them quickly
12:31 <@ss> bug 370721
12:31 < thebot> ss: Bug maj, --, Camino1.1, 
      , NEW, Image dragged from the Desktop to the content area fails 
                to render
12:32 <@ardissone|food> this is "the other version"
12:32 <@ss> Right
12:32 <@ss> Just posting them in order. :)
12:32 <@ss> bug 350732
12:32 < thebot> ss: Bug tri, --, Camino1.1, 
      , NEW, Going back with popup bar on screen truncates scroll view
12:33 <@ardissone|food> mento: what's your calendar look like for getting into 370721?
12:35 <@ardissone|food> smorgan and murph did a little digging into 350732 earlier
12:35 <@smorgan> I'll get to it again at some point
12:35 <@ardissone|food> but as i recall, the results were inconclusive
12:36 <@smorgan> I was trying to knock off my smaller bugs first, especially the keychain stuff 
                 that's sequential
12:36 <@ardissone|food> mhm
12:36 <@ss> Most of these bugs are still being looked at, right?
12:37 <@ardissone|food> not really
12:37 <@ardissone|food> that's why they're on the refresher list ;)
12:37 <@ss> The topcrash has an assignee
12:37 <@ss> Spelling UI does too
12:37 <@murph> I will take another look at 350732 again, along with smorgan.
12:38 -!- kreeger has quit [Quit: kreeger]
12:38 <@ss> 372003 has a patch
12:38 <@ardissone|food> yeah, no real reports of the download thing since smorgan's fix went in
12:38 <@ss> And that might fix 370954
12:39 <@ss> bug 371550 is the only one without any traction
12:39 < thebot> ss: Bug cri, --, ---, 
      , NEW, Crash or hang when scrolling a page while it's loading
12:39 <@ardissone|food> i've poked it to the extent of my ability to no avail
12:40 <@smorgan> People should just keep trying with zombies and such, I guess
12:41 <@ardissone|food> there are some talkbacks from 10.4.8 now, too
12:42 <@ardissone|food> bug 372020, then
12:42 < thebot> ardissone|food: Bug nor, --, 
                ---,, NEW, Need to prevent inadvertent bookmarking of 
                javascript: and data: URLs
12:43 -!- ardissone|food is now known as ardissone
12:43 <@ss> Yeah, I'd like to talk about this one.
12:43 <@ss> Core has proposed adding a dialog and I tend to agree with them.
12:43 <@ardissone> (refresh your agenda)
12:43 <@ss> If we're going to do that for 1.1, we need to have get the string in sooner rather 
            than later.
12:44 <@pinkerton> why would that block 1.1?
12:44 <@ss> s/string/nib
12:44 <@cl> "security"
12:44 <@ss> pinkerton: Security mainly.
12:44 <@smorgan> Why would there be a dialog?
12:44 <@pinkerton> we've gone this long....
12:44 <@ss> Because failing at bookmarking something without a dialog is shitty, I think.
12:45 <@pinkerton> is there a rash of people trying to explot this?
12:45 <@ss> pinkerton: Yeah, times are changin'
12:45 <@smorgan> We just don't enable the shortcut
12:45 <@smorgan> It's not a silent failure, it beeps
12:45 <@pinkerton> heh
12:45 <@ardissone> what about in a tab group/bookmark all tabs
12:45 <@ss> ^^
12:45 <@smorgan> Raise your hand if you've ever been on a data: or javascript: page that wasn't 
                 the exploit
12:46  * ardissone raises :p
12:46 <@pinkerton> raise your hand if you're ok throwing ardissone under the bus
12:46 <@ss> Well, sure, but I don't think that's the point.
12:46 <@ardissone> we don't want to disable Bookmark All if there's one of these bad things
12:46 <@ardissone> i agree we're ok with disabling bookmark this page
12:46 <@ss> A full, working exploit hadn't been made yet and no one knew it could be done.
12:47 <@ss> I'm okay with no dialog, but I agree with ardissone.
12:47 -!- kreeger has joined #camino-mtg
12:47 <@ss> We shouldn't disable bookmark all in those cases.
12:47 <@pinkerton> just skip it in the group
12:47 <@cl> ignoring the potentially problematic bookmark seems like a reasonable course of 
            action, yeah
12:47 <@ss> And don't alert the user?
12:47 <@smorgan> How is my point that no-one will ever see it not relevant to "we need to break 
                 loc freeze so that an action that basically no-one would ever legitimately see 
                 has a dialog"?
12:48 <@ss> You can guarantee that "no-one will ever see it"?
12:48 <@ss> Because if you can, you know far more about users than I do.
12:49 <@ss> I don't think you can make assumptions as to what users will and won't see.
12:49 <@smorgan> I have never, ever, seen a data: page
12:49 <@ss> I hate.
12:49 <@ss> have*
12:49 <@smorgan> If we get a bunch of users complaining that it doesn't bookmark them, we release 
                 an update later
12:50 <@smorgan> Slightly degraded UE on a vanishingly small fraction of web pages is hardly a 
                 loc-freeze-breaking issue
12:51 <@smorgan> We have lots of bugs that would impact far more users that we wouldn't break 
                 freeze for
12:52 <@pinkerton> why not just disable the "bookmark item" and drop the tab from "bookmark all"
12:52 <@pinkerton> done.
12:52 <@smorgan> right
12:52 <@pinkerton> next?
12:52 <@ardissone> Bug 248160
12:52 < thebot> ardissone: Bug nor, --, 
                Camino1.2,, NEW, [l10n] Translate chrome strings on the fly
12:53 <@ardissone> kreeger wanted to talk about this quickly
12:53 <@kreeger> ...
12:54 <@kreeger> i have a patch up that uses the |nsIStringBundleOverride| interface
12:54 <@kreeger> basically the way it works is we look up a Localized string, if we don't define 
                 that one in our own Localization, we just use the gecko string
12:55 <@kreeger> the only trick to it is...
12:55 <@kreeger> the nsIStringBundle service gets started up before we register our own components
12:55 <@kreeger> so there is a core patch to add a new function call to the idl
12:55 <@kreeger> |void InitStringBundleOverrideService();|
12:56 <@kreeger> does this sound like the path we want to take?
12:56 <@pinkerton> why that over doing the same lookup in the front end?
12:57 <@pinkerton> oh, i guess it prevents the double lookup
12:57 <@kreeger> this is for strings that get processed in the background, like the HTTP password 
                 prompts and sheets
12:57 <@pinkerton> gotcha
12:57 <@pinkerton> seems reasonable to me
12:57 <@kreeger> there really isn't too many ways of getting notified that a string is getting 
                 used in gecko
12:58 <@kreeger> all we have to do is define the string for the gecko key in our localized string 
12:58 <@kreeger> ok cool, ill work on that a bit more
12:58 <@kreeger> (post 1.1ish)
12:59 <@ardissone> anyone have anything else?
12:59 <@smorgan> Any objection to bug 371480?
12:59 < thebot> smorgan: Bug nor, --, 
                Camino1.1,, NEW, Upgrade keychain entries at first 
                use, not first change
12:59 <@pinkerton> how does that not screw people that go back to 1.0x?
12:59 <@smorgan> It does
13:00 <@pinkerton> then i'm against it
13:00 <@smorgan> My point is that I don't think we care in 1.1
13:00 <@ardissone> who's gonna go back from 1.1 final?
13:00 <@pinkerton> i dunno, people do
13:00 <@ardissone> no one
13:00 <@ardissone> to an EOL release?
13:00 <@pinkerton> if it works for people and 1.1 for some reason crashes
13:00  * pinkerton shrugs
13:01  * pinkerton notes he's running an ancient version of colloquy because new versions mess up 
          on the mbp
13:01 <@ardissone> it's gonna screw up people at some point no matter what
13:02 <@ardissone> unless we never clean up
13:02 <@smorgan> How do the various pref upgrades we have done over time not screw upgrading 
13:03 <@pinkerton> those aren't as important as "shit, what's my Vanguard password, i need to 
                   dump all my APPL shares"
13:03 <@pinkerton> (aapl)
13:04  * pinkerton shrugs
13:04 <@pinkerton> i think users do this more than we think
13:04 <@pinkerton> they stick with what works if the latest and greatest goes south on them
13:04 <@pinkerton> i'm not putting my foot down, but i think we need to be careful
13:05 -thebot:#camino-mtg- Build 'MacOSX Darwin 8.6.0 maya Depend CmTrunk' has dropped from the 
          'Camino' tinderbox.
13:05 -!- thebot was kicked from #camino-mtg by ss
13:05 <@ardissone> um
13:05 <@pinkerton> yay?
13:05 <@ardissone> that's bad
13:05  * pinkerton patpats the bot
13:05 <@ss> I bet it's just still going.
13:06 <@ss> It's taken too long.
13:06 <@pinkerton> go cairo
13:06 <@ardissone> working on the trunk nightly
13:07 <@smorgan> Can we just relnote it?  "If you decide to live in the past after using 1.1 for 
                 a while, you can use Keychain Access to get your vanguard password"
13:07 <@smorgan> Or I can live with cruft until 1.2
13:07 <@pinkerton> i was going to suggest 1.2 as an option
13:07 <@smorgan> Well, until 2.0 I guess
13:09 <@pinkerton> i gotta jet
13:09 <@smorgan> I guess I just don't like the idea that we are making design decisions around 
                 the idea that people will dislike 1.1 and downgrade :(
13:09 <@ardissone> indeed :(