Status Meetings:2007-12-12:Log

[02:14am] You were promoted to operator by chanserv.
[02:14am] You changed the topic to "http://wiki.caminobrowser.org/Status_Meetings:2007-12-12:Agenda".
[02:16am] cl|zzz was promoted to operator by you.
[07:13am] ampcoder joined the chat room.
[09:25am] cl|zzz is now known as cl.
[12:01pm] smorgan joined the chat room.
[12:02pm] smorgan was promoted to operator by you.
[12:02pm] pinkerton joined the chat room.
[12:02pm] pinkerton was promoted to operator by chanserv.
[12:02pm] ardissone|away: we expecting mento?
[12:04pm] pinkerton: dunno
[12:05pm] ardissone|away: let me finish fixing this food realquick and then we can start, i guess
[12:06pm] pinkerton: i'll do the same
[12:10pm] ardissone|away: we ready?
[12:13pm] pinkerton: shrugs
[12:14pm] ardissone|away: ok
[12:14pm] ardissone|away: the big news is we released 1.6a1 yesterday
[12:14pm] ardissone|away: the world has not (yet) ended as a result of said release
[12:15pm] pinkerton: yay!
[12:15pm] ardissone|away: as usual, we're working to find ways to give it enough exposure but to prevent update sites from recommending it to everyone
[12:15pm] pinkerton: desmond said he liked seeing the scrolling tab stuff ship
[12:15pm] ardissone|away:
[12:16pm] smorgan: Although at least theoretically if they do, we won't have people still using it months later 
[12:16pm] ardissone|away: indeed
[12:16pm] cl: yeah, got my first swupdate notice last night
[12:16pm] cl: yay.
[12:16pm] ardissone|away: (more concerned about them all crashing on quit and giving up)
[12:16pm] ardissone|away: yay
[12:17pm] smorgan: ardissone|away: crashing on quit?
[12:17pm] smorgan: You mean my bug, that's in the stable version too? 
[12:17pm] ardissone|away: sam and I are working on some additional linkage from cbo to the preview
[12:17pm] ardissone|away: yes
[12:17pm] ardissone|away: i guess there's no difference there 
[12:17pm] cl: speaking of feedback on 1.6a1
[12:18pm] cl: the guy who wrote to feedback complaining about the way tabs were handled in 1.5.x says he really likes the new scrolling and menu
[12:18pm] ardissone|away: yay
[12:19pm] ardissone|away: sam promised we'd get talkback running today for 1.6a1
[12:19pm] pinkerton: cool
[12:19pm] ardissone|away: i did look at branch stats in general, and our two topcrashers there are Grammarian-related
[12:20pm] ardissone|away: and then smorgan 
[12:20pm] ardissone|away: 1.5.4 seems to be going well, aside from that one crash
[12:20pm] ardissone|away: particularly, i've heard no complaints on the forum this time 
[12:22pm] ardissone|away: sam and i were a little concerned that we're not getting enough branch testings (or something), since the site icon cleanup was on the branch for 1mo before we even spun 1.5.4
[12:22pm] ardissone|away: s/were/are/
[12:23pm] ardissone|away: i'm not really sure what to say about that
[12:24pm] ardissone|away: jumping back to 1.6b1
[12:24pm] ardissone|away: we need to start thinking about a date
[12:24pm] jeff joined the chat room.
[12:25pm] ardissone|away: and what features won't make b1/1.6
[12:25pm] ardissone|away: and which additional 10.5 changes, if any, we do want to take
[12:25pm] pinkerton: goes into a meeting
[12:25pm] smorgan: I'll be whipping up the breakpad reporter Real Soon Now
[12:26pm] ardissone|away: once we get a bit more space from the release-a-week land, i'll try to start looking at the nib/strings polish bugs
[12:26pm] smorgan: And doing the next round of reviews this week, which should help with the major 1.6b features
[12:26pm] ardissone|away:
[12:27pm] jeff was promoted to operator by you.
[12:27pm] ardissone|away: jeff is the only other person we can pick on today
[12:27pm] ardissone|away: for status update
[12:27pm] jeff: great ....
[12:28pm] ardissone|away:
[12:28pm] jeff: It should be close to 'finished' once bug 291160 lands
[12:29pm] ardissone|away: ?
[12:29pm] ardissone|away: Bug 291160 – breadcrumb extension seems to break if there's a slash in a page name
[12:31pm] ardissone|away: hello? are we still here?
[12:32pm] smorgan: is
[12:32pm] ardissone|away: whew
[12:32pm] jeff: just woke up
[12:32pm] ardissone|away: thought internet had died
[12:32pm] ardissone|away: aha 
[12:33pm] ardissone|away: suggests jeff try his status update again
[12:33pm] pinkerton: is in a meeting
[12:33pm] jeff: woops I meant 406510
[12:34pm] smorgan: An easy typo to make 
[12:34pm] ardissone|away: ah
[12:34pm] ardissone|away: ok, so we're blocked there on pink
[12:34pm] ardissone|away: but your code is already reworked for that patch, correct?
[12:35pm] jeff: yes, and now I am just finishing up some bugsadn
[12:35pm] jeff: s/bugsadn/bugs
[12:35pm] jeff: needs coffee
[12:35pm] ardissone|away: indeed 
[12:36pm] ardissone|away: the other thing to note is that our "major" landings typically produce many regressions, so we need to get our remaining features landed as soon as possible so we can start cleaning up after them
[12:37pm] ardissone|away: anyway
[12:38pm] ardissone|away: anything else on 1.6 features and planning?
[12:39pm] jeff: one thing that I *might* be able to squeeze in is animation when using scrolling tabs
[12:39pm] jeff: but seeing how I have been on getting tab DnD done .. its just a pipe dream at the moment
[12:41pm] ardissone|away: suspects scrolling animation would be a good 2.0 feature when we're 10.4 only 
[12:41pm] ardissone|away: on the tinderbox front
[12:42pm] ardissone|away: sam has been ordering minis to replace binus and maya
[12:42pm] ardissone|away: mini-binus is supposed to be here friday, and mini-mayaTrunk on tuesday, iirc
[12:44pm] ardissone|away: cb-mini has a low Ts timeout (15 secs) which is apparently the cause of random orange
[12:44pm] mento joined the chat room.
[12:44pm] mento was promoted to operator by chanserv.
[12:44pm] ardissone|away: heh
[12:44pm] ardissone|away: he heard us talking about tinderboxen 
[12:44pm] mento: oops
[12:44pm] cl: haha
[12:44pm] mento: i thought it was tuesday or thursday
[12:45pm] ardissone|away: yeah, i thought yesterday was monday 
[12:45pm] mento: glad it wasn't
[12:46pm] ardissone|away: mento: you still going to cvs the cb-mini configs?
[12:46pm] mento: right, yes
[12:46pm] mento: i just want to bump up the ts timeout a little bit
[12:46pm] ardissone|away: yeah
[12:46pm] mento: but it's stable, so i'll check 'em in soon, probably today
[12:46pm] ardissone|away: sam was supposed to talk to you aboyut that yesterday 
[12:46pm] ardissone|away: but that's good
[12:47pm] mento: 'k
[12:47pm] ardissone|away: the length of the r queue is suffering a bit more of late at the hands of mento-on-fire
[12:48pm] ardissone|away: which i suppose is better than no new patches joining it, so
[12:48pm] ardissone|away: half full, half empty, pick your glass
[12:48pm] mento: if i can retarget any of my r?s, let me know
[12:48pm] mento: i've been sending them to you and stuart
[12:48pm] mento: because i know you two always pay attention
[12:48pm] ardissone|away: yeah, i'll get to mine soon
[12:49pm] ardissone|away: the release-a-week syndrome hasn't been good to me 
[12:50pm] ardissone|away: pinkerton: are you still meeting, or can you listen to smorgan and i debate over one of your checkins from 2006? 
[12:52pm] cl: for the record, I'm in favour of doing what the bug/smorgan suggest
[12:52pm] cl: because it's more consistent.
[12:53pm] cl: and it'll finally give me reason to set that pref, because having it come forward on every download will reach my personal annoyance threshold.
[12:53pm] jeff: what bug are we talking about?
[12:54pm] smorgan: ardissone|away: what's the argument for the current way?
[12:54pm] ardissone|away: that we always want to show the manager when it's opened for the first time
[12:54pm] ardissone|away: but we want to prevent focus-stealing thereafter
[12:54pm] cl: jeff: bug 407883
[12:55pm] cl: right, that lack of consistency doesn't make much sense to my mind
[12:55pm] cl: and here's why:
[12:55pm] cl: either you keep the manager where you can see it
[12:55pm] smorgan: ardissone|away: I don't follow; that's not what the bug is asking
[12:55pm] cl: and therefore you weren't one of the people who was complaining about not knowing when downloads started to begin with
[12:55pm] smorgan: ardissone|away: If you open the manager and it's not open, it will come forward
[12:56pm] cl: *or* you keep the manager where you can't see it, meaning you probably *never* know when a download starts once the manager is open but hiding behind other windows
[12:56pm] ardissone|away: is completely confused now
[12:56pm] smorgan: ardissone|away: the question is whether starting a download should *open* the manager
[12:56pm] ardissone|away: oh
[12:56pm] smorgan: Not whether deliberately opening the manager should bring the window forward
[12:57pm] smorgan: (which clearly it should; that's how all windows work)
[12:57pm] ardissone|away: i think that should be a separate pref, actually
[12:57pm] smorgan: Why?
[12:57pm] ardissone|away: and not hijack the non-focus-stealing pref
[12:57pm] smorgan: It does steal focus though
[12:57pm] smorgan: It exactly steals focus
[12:58pm] ardissone|away: pink made the manager always steal focus on every download
[12:58pm] smorgan: Here's the question: should the expected behavior for people who don't want to see it depend on them remembering if a window they may well not be able to see is open or closed?
[12:58pm] ardissone|away: which is why opening/not opening should be a separate option
[12:58pm] smorgan: No, it's why it shouldn't
[12:59pm] smorgan: Maybe I'm not being clear, let me try an example:
[12:59pm] ardissone|away: if you never want to see it, you don't need the current hidden pref
[12:59pm] smorgan: huh?
[12:59pm] ardissone|away: if you want to see it but don't want each new download to steal focus
[12:59pm] smorgan: How do you not need the pref if you don't want to see it?
[1:00pm] cl: I never want to see it.
[1:00pm] ardissone|away: if we had a "don't open download manager" pref
[1:00pm] cl: Yet I see it all the motherloving time
[1:00pm] cl: I just never set the pref, which is why I still see it 
[1:00pm] ardissone|away: you don't need the "don't steal focus on 2nd and following downloads" hidden pref
[1:01pm] ardissone|away: the existing setup was not designed to support a "never show downloads" setup
[1:01pm] cl: wasn't that exactly what was supported before pink wrote the current behaviour?
[1:02pm] ardissone|away: no, it opened in the bg before pink's changes
[1:02pm] ardissone|away: opened all the time, every dl
[1:02pm] ardissone|away: just hidden
[1:02pm] ardissone|away: "hidden"
[1:02pm] ardissone|away: afaict, we've never supported not showing the dl window
[1:03pm] ardissone|away: we should, is my argument
[1:03pm] smorgan: okay, example of the point I was trying to make:
[1:03pm] ardissone|away: but not with the pref that keeps the window from stealing focus once opened on successive dls
[1:04pm] smorgan: You sit at your machine, which a big browser window open, and the pref is at default. You click a download. The download window will become visible, guaranteed.
[1:04pm] smorgan: Now, you sit at your machine, with a big browser window open, and the pref set. You click a download link. What happens?
[1:05pm] smorgan: You can't answer, unless you remember whether or not your download window was open
[1:05pm] ardissone|away: it comes to the front again and steals dfocus
[1:05pm] ardissone|away: if the pref was at default
[1:05pm] smorgan: Sorry, "set" was reversed
[1:05pm] ardissone|away: with the hidden pref not set, the behavior is completely predictable
[1:05pm] smorgan: "set to false"
[1:06pm] smorgan: is confused by "set" and "not set" now.
[1:07pm] smorgan: In example two, you have set the pref to "No focus-stealing please"
[1:07pm] smorgan: It's no longer predictable
[1:07pm] smorgan: Why is that good?
[1:08pm] ardissone|away: because you only set the pref when you can see your manager?
[1:08pm] cl: ardissone|away: it doesn't come to the front and steal focus if it was already open and hidden behind another window
[1:08pm] smorgan: okay, I see what you mean now
[1:09pm] ardissone|away: i see your point, but i don't think full-window people are setting the pref
[1:09pm] ardissone|away: maybe i'm wrong
[1:09pm] ardissone|away: set = don't steal
[1:09pm] ardissone|away: if i were a full-window browser, i don't think i'd have it set
[1:09pm] ardissone|away: but i have a nice spot where i can the mgr at all times
[1:10pm] ardissone|away: i want it to open when i start a donwload, but never again steal focus, because i can look at it and have it not interrupt me
[1:10pm] ardissone|away: if people never want it to open, i think that's "reasonable" and we should support that
[1:10pm] cl: I want it never to open unless I explicitly invoke it.
[1:11pm] ardissone|away: just not using the pref designed to make it not steal focus when it's already open 
[1:11pm] cl: i am smart enough to know that my download has started without needing the manager to open and tell me
[1:11pm] cl: and if I want to see how it's progressing, I can go open the manager and look.
[1:11pm] ardissone|away: is happy for cl 
[1:12pm] smorgan: ardissone|away: the argument the other way is that it's a pref to turn off bringing the window forward, and it doesn't always prevent it from being brought forward
[1:12pm] ardissone|away: it's ill-named, yes
[1:13pm] ardissone|away: if we had named it dontStealFocusOnceOpen, it'd be better
[1:13pm] smorgan: What if we made it open and send-to-back with the pref set? I wonder if that would please everyone
[1:13pm] ardissone|away: but i believe we reused a name for a similar existing pref in xpfe
[1:13pm] cl: why don't we use the Core prefs for this?
[1:13pm] ardissone|away: not the old gf
[1:13pm] cl: yeah, and it looks like those prefs don't exist any more, at least as far as kb.mz.o is concerned
[1:14pm] ardissone|away: i dunno
[1:14pm] cl: they're all browser.download.manager.* now
[1:14pm] cl: see http://kb.mozillazine.org/About:config_entries#Browser..2A
[1:14pm] ardissone|away: kb.mz.o is also 99% firefox
[1:14pm] cl: I believe our current pref is analagous to browser. download. manager. focusWhenStarting
[1:15pm] cl: combined with browser. download. manager. showWhenStarting
[1:16pm] ardissone|away: if they do what they sound like, i imagine so
[1:16pm] smorgan: So we can either try my hybrid, or we could switch to both FF prefs
[1:16pm] ardissone|away: yeah
[1:16pm] cl: what the heck does browser.download.progressdnlddialog.keepAlive do?
[1:16pm] ardissone|away: er
[1:17pm] smorgan: close when done, I think
[1:17pm] cl: ah, right
[1:17pm] smorgan: any objection to trying the hybrid first?
[1:17pm] cl: new name: browser. download. manager. closeWhenDone
[1:17pm] ardissone|away: with your hybrid, true = open and send-to-back ?
[1:17pm] cl: (much more logical, IMO)
[1:17pm] smorgan: yes
[1:17pm] smorgan: Although having sane names does appeal to me
[1:18pm] cl: smorgan: does your hybrid allow a "don't ever show this unless I ask for it" option?
[1:18pm] ardissone|away: sane names +++
[1:18pm] smorgan: no, because it's never been requested AFAIK
[1:18pm] cl: consider this a request 
[1:18pm] ardissone|away: wait, isn't that what the bug is filed about?
[1:18pm] smorgan: ardissone|away: it's actually not clear
[1:18pm] ardissone|away: oh, good
[1:18pm] smorgan: The patch does that
[1:18pm] smorgan: But the description is about bringing forward
[1:18pm] cl: which leads me to believe that's what Ben wanted 
[1:19pm] smorgan: "IMO, the Download window should never jump forward if the preference is set to false"
[1:19pm] smorgan: which my hybrid does, exactly
[1:19pm] cl: i think someone should ask Ben what he meant
[1:19pm] smorgan: And the title says "should not come forward" not "should not open"
[1:20pm] cl: 'cause I would definitely be in favour of a solution that would prevent it from opening entirely unless I ask for it, which, IIRC, is how it used to work.
[1:20pm] smorgan: I'm fine with just tossing the extra pref in and migrating to the new names
[1:20pm] ardissone|away: i think that's cleaner
[1:20pm] smorgan: Adding a new hidden pref in one very narrow place seems worthwhile for the name sanity
[1:20pm] smorgan: pinkerton: do you object?
[1:20pm] smorgan: (say "no"  )
[1:21pm] ardissone|away: and i worry about the hybrid true bringing back the behavior that pink fixed initially due to complaints
[1:21pm] smorgan: ardissone|away: hm?
[1:21pm] smorgan: The default would still be the current behavior
[1:21pm] ardissone|away: oh
[1:21pm] ardissone|away: gotcha
[1:22pm] pinkerton: let me read back
[1:22pm] ardissone|away: you'll just continue to p.o. cl in that case 
[1:23pm] pinkerton: i'm confused
[1:23pm] cl: haha
[1:23pm] ardissone|away: welcome to the club
[1:23pm] smorgan: Aren't we all 
[1:23pm] smorgan: Okay, short version:
[1:23pm] smorgan: Base problem is that when the manager is closed, it opens and comes forward even with the "don't come forward" pref
[1:23pm] smorgan: Two possible fixes on the table:
[1:24pm] pinkerton: we don't have a pref ui for that do we? or is that trunk only?
[1:24pm] smorgan: 1) Keep our current prefs, but change the behavior for "don't steal focus" to send a newly-opened-due-to-download manager to the back
[1:24pm] pinkerton: is running 1.5
[1:25pm] smorgan: 2) Adopt one more pref, and switch everything to the saner Firefox names for these prefs
[1:25pm] ardissone|away: the pref is hidden
[1:26pm] smorgan: pinkerton: the current pref has two states: default = open and bring forward the window when it's not open, bring forward when it is, and other) open and bring forward the window when it's not open, but don't bring it forward when it is
[1:26pm] pinkerton: ah, i see
[1:26pm] smorgan: if you want "never ever bring it forward, whether I happened to leave the window open last night or not", you are currently SOL
[1:27pm] pinkerton: right
[1:27pm] pinkerton: i like saner names, but will migrating the prefs cause cruft migration code?
[1:27pm] smorgan: Yes, it will
[1:27pm] smorgan: but that's hardly new
[1:28pm] ardissone|away: we're better at cleaning it out now, too
[1:28pm] pinkerton: i take it you folks prefer (2)?
[1:28pm] smorgan: Yeah, I got rid of some Chimera 0.5 pref migrators recently 
[1:28pm] smorgan: I prefer 2 just because the current names make me cry when I see them
[1:28pm] pinkerton: that's fine. make it so
[1:28pm] ardissone|away: they lead to confusing bugs being filed, too 
[1:29pm] smorgan: can someone list the prefs for me in that bug? I'll make a new patch
[1:29pm] ardissone|away: which in turn make meetings run over 30 mins
[1:29pm] ardissone|away: yeah
[1:29pm] ardissone|away: just those two, or sanifying all the dl ones?
[1:31pm] smorgan: If there are corresponding prefs for all, lets do all of them
[1:31pm] ardissone|away: ok 
[1:31pm] ardissone|away: anything else today?
[1:32pm] ardissone|away: ok, let's all have a good week 
[1:33pm] ardissone|away: happy hacking/reviewing
[1:33pm] ardissone|away: don't forget to try 1.6a1