fluid-work IRC Logs-2008-07-23

Owned by Fluid IRC Log Bot
Last updated: Jul 23, 2008Version comment
20 min read

[02:04:58 EDT(-0400)] * phiggins (n=dante@12.157.240.2) has joined #fluid-work
[08:01:42 EDT(-0400)] * Justin_o (n=Justin@142.150.154.235) has joined #fluid-work
[08:48:08 EDT(-0400)] * jhung (n=Jon@142.150.154.211) has joined #fluid-work
[09:11:45 EDT(-0400)] * anastasiac (n=team@142.150.154.160) has joined #fluid-work
[09:17:29 EDT(-0400)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined #fluid-work
[09:53:27 EDT(-0400)] * theclown (n=theclown@guiseppi.atrc.utoronto.ca) has joined #fluid-work
[10:00:09 EDT(-0400)] <Bosmon> I'm not sure I "like" the JQuery plugin model...
[10:24:28 EDT(-0400)] * colinclark (n=colin@142.150.154.101) has joined #fluid-work
[10:52:54 EDT(-0400)] <Bosmon> Everyone who might be pointing at keyboard-a11y note that I am just committing a fairly substantial refactoring that changes its external API and semantics quite a bit...
[10:53:02 EDT(-0400)] <Bosmon> http://issues.fluidproject.org/browse/FLUID-993
[10:53:14 EDT(-0400)] <Bosmon> I'll make the docs catch up later in the day....
[11:08:42 EDT(-0400)] <colinclark> Bosmon just asked about the callbacks in the accessibility plugin.
[11:08:47 EDT(-0400)] <colinclark> I will respond here in the channel.
[11:08:58 EDT(-0400)] <colinclark> (14:42:05) Antranig: I am wondering if you have any historical recollection of why we chose to name our keyboard handlers "willSelect" and "willUnselect" in favour of the more traditional "onSelect" etc.?
[11:09:49 EDT(-0400)] <colinclark> There are a couple of reasons I chose this terminology, but I'm open to suggestions.
[11:10:05 EDT(-0400)] <colinclark> First, I wanted to distinguish these callbacks from ordinary event handlers...
[11:11:12 EDT(-0400)] <colinclark> They happen as a result of a very convenient abstraction. The keyboard-a11y plugin takes care of all the event binding for you, so that you need to only be concerned with higher-level semantics like "select" and "unselect" rather than focus and blur and keydown and all of that stuff.
[11:11:35 EDT(-0400)] <colinclark> I've also been thinking about creating a standard convention for making the timing of callbacks clear:
[11:12:07 EDT(-0400)] <colinclark> willSelect() suggests that the callback is called right before selection happens. Similarly with willUnselect.
[11:12:17 EDT(-0400)] <Bosmon> Yes
[11:12:21 EDT(-0400)] <colinclark> One can also imagine cases where you want a callback after things have happene.d
[11:12:34 EDT(-0400)] <colinclark> Maybe not as relevant for selection, but this would provide consistency.
[11:12:39 EDT(-0400)] <colinclark> So, didSelect(), etc.
[11:13:11 EDT(-0400)] <colinclark> The point Antranig raises is whether or not this is an added bit of confusion from the standard on[Event] style.
[11:13:16 EDT(-0400)] <Bosmon> Well, I think for these cases the "onXxxx" and "afterXxxx" are more recognisable by the world at large.....
[11:13:28 EDT(-0400)] <colinclark> "on" doesn't distinguish between before and after.
[11:13:34 EDT(-0400)] <Bosmon> I find it a bit unsettling to implement a function which is named "willXxxxx"
[11:13:36 EDT(-0400)] <colinclark> Perhaps, yes.
[11:13:44 EDT(-0400)] <Bosmon> Since it really makes no sense in ones own mind (tongue)
[11:13:53 EDT(-0400)] <Bosmon> If it "will" happen, why on earth am I implementing it now (tongue)
[11:13:53 EDT(-0400)] <colinclark> I'm taking my cue from the Cocoa delegation model.
[11:14:04 EDT(-0400)] <Bosmon> Ah!
[11:14:05 EDT(-0400)] <colinclark> I am showing my lineage. (wink)
[11:14:25 EDT(-0400)] <Bosmon> Well, I believe the standard Javascript semantic for "on" is essentially a "before"
[11:14:26 EDT(-0400)] <colinclark> I'm willing to move to onEvent and afterEvent if we think it's more apparent.
[11:14:43 EDT(-0400)] <Bosmon> Since one always has the option to "return false" from a handler to thus curtail the effect of handling
[11:14:48 EDT(-0400)] <colinclark> The full picture of delegation would actually incude predicates:
[11:14:50 EDT(-0400)] <colinclark> include
[11:14:53 EDT(-0400)] <colinclark> shouldSelect
[11:14:54 EDT(-0400)] <Bosmon> Which could not be the case if this handling were really "after"
[11:14:56 EDT(-0400)] <colinclark> willSelect
[11:14:58 EDT(-0400)] <colinclark> didSelect
[11:15:57 EDT(-0400)] <Bosmon> To be honest, there is no real need for an explicit "before" and "after" once you have a properly functional "on"
[11:16:09 EDT(-0400)] <Bosmon> It is a bit like the "half-AOP" models that give you both kinds of "half-sided pointcuts"
[11:16:42 EDT(-0400)] <Bosmon> I propose a model based on GLOBAL PERMANENT WARFARE
[11:17:01 EDT(-0400)] <Bosmon> Where we only ever deal in "full-sided" interception functions (tongue)
[11:19:24 EDT(-0400)] * colinclark laughs.
[11:20:07 EDT(-0400)] <colinclark> Out of sheer curiosity, I'm going to check if jQuery UI has a standard convention for this.
[11:24:23 EDT(-0400)] <Bosmon> Is there some kind of "natural size limitation" in the default Uploader sample?
[11:24:36 EDT(-0400)] <Bosmon> I don't seem to be able to add files much bigger than about 15Mb
[11:24:39 EDT(-0400)] <colinclark> Bosmon: Size of file, you mean?
[11:25:07 EDT(-0400)] <colinclark> It seems unlikely, since it mocks out the upload process when a server is absent. A question for Eli, I think.
[11:25:23 EDT(-0400)] <Bosmon> Yes, file size
[11:25:44 EDT(-0400)] <Bosmon> Secondly, I do see that keyboard controls do actually seem to work, for tabbing and arrowing down the "X" boxes
[11:25:52 EDT(-0400)] <Bosmon> But I guess this is being done manually right now?
[11:26:50 EDT(-0400)] <colinclark> Bosmon: I wouldn't think arrowing should work. Tabbing, yes.
[11:27:10 EDT(-0400)] <Bosmon> Oh, yes, I see now
[11:44:38 EDT(-0400)] <colinclark> anastasiac is looking for a copy of Bullshot...
[11:44:39 EDT(-0400)] <colinclark> http://www.imdb.com/title/tt0085279/
[11:44:52 EDT(-0400)] <colinclark> As a tool to decode Bosmon's strange affinity for global warfare.
[11:44:55 EDT(-0400)] <colinclark> (wink)
[11:54:53 EDT(-0400)] * ecochran (n=ecochran@dhcp-169-229-212-100.LIPS.Berkeley.EDU) has joined #fluid-work
[11:55:04 EDT(-0400)] <Bosmon> Hi Eli (smile)
[12:07:10 EDT(-0400)] <ecochran> Justin_o: look at Fluid-822
[12:07:14 EDT(-0400)] <ecochran> might be your resume bug
[12:07:30 EDT(-0400)] <Justin_o> okay.. i'm going to check it now
[12:07:32 EDT(-0400)] * phiggins (n=dante@conference/oscon/x-ff350bb6e77e23b3) has joined #fluid-work
[12:08:59 EDT(-0400)] <Bosmon> Eli - I saw an odd thing when I added "big files"?
[12:09:13 EDT(-0400)] <Bosmon> I assume there is some kind of limit configured into the thingummy... that makes them go away
[12:09:15 EDT(-0400)] <ecochran> shoot
[12:09:19 EDT(-0400)] <ecochran> yes
[12:09:26 EDT(-0400)] <ecochran> I need to add feedback for that
[12:09:26 EDT(-0400)] <Bosmon> But there also seems to be some support in the code for error feedback in this case
[12:09:31 EDT(-0400)] <Bosmon> But I don't believe I actually see it (tongue)
[12:09:32 EDT(-0400)] <Bosmon> Oh, ok
[12:09:33 EDT(-0400)] <Bosmon> cool
[12:09:45 EDT(-0400)] <Justin_o> ecochran: yes that does appear to be the resume bug..
[12:09:55 EDT(-0400)] <ecochran> there is a JIRA related to the big files issue
[12:10:09 EDT(-0400)] <ecochran> Justin_o: there is not a short-term fix for that bug
[12:10:26 EDT(-0400)] <ecochran> I may have two workarounds but neither is pretty
[12:10:40 EDT(-0400)] <ecochran> 1 > only pause after the current upload completes
[12:11:09 EDT(-0400)] <ecochran> 2 > If the swf doesn't respond, just skip that file and move on
[12:11:30 EDT(-0400)] <ecochran> that actually might not be too bad, but I'd need some kind of smart error
[12:12:43 EDT(-0400)] <Justin_o> currenlty for errors, does it skip the file and continue or just halt the entire upload process?
[12:12:59 EDT(-0400)] <Bosmon> This is famously one of those issues which causes developers to throw tomatoes at each other, but should we do something about Uploader.js being a somewhat unholy mixture of tabs and spaces? (tongue)
[12:18:06 EDT(-0400)] * phiggins (n=dante@conference/oscon/x-6831e3672a43f7f7) has joined #fluid-work
[12:21:44 EDT(-0400)] <anastasiac> I thought that all of our tomatoes had been thrown, and cleaned up already... didn't we agree to spaces? 4 spaces?
[12:22:06 EDT(-0400)] <ecochran> Justin_o: some errors stop the upload, other continue depending on the evil of the error
[12:22:22 EDT(-0400)] <ecochran> Probably my fault on the spaces and tabs
[12:22:25 EDT(-0400)] <ecochran> I'll clean it up
[12:22:30 EDT(-0400)] <ecochran> I like tabs
[12:23:12 EDT(-0400)] <Justin_o> ecochran: thanks for the info...
[12:24:34 EDT(-0400)] <Bosmon> !
[12:24:41 EDT(-0400)] <Bosmon> I must have missed the tomato session...
[12:24:51 EDT(-0400)] <ecochran> I wasn't there either
[12:24:59 EDT(-0400)] <ecochran> it must have been a UT
[12:25:01 EDT(-0400)] <ecochran> thing
[12:25:25 EDT(-0400)] <Bosmon> I'd be happy with the spaces, so long as there was some even number of them
[12:25:38 EDT(-0400)] <Bosmon> Through some mysterious "compromise" with Aaron, I found a lot of our code committed to 3 spaces....
[12:26:36 EDT(-0400)] <ecochran> OK, warning to anyone working in Uploader.js... I just checked in the file with tabs converted to spaces and... oh... 80% of the lines changed
[12:26:38 EDT(-0400)] <ecochran> ha
[12:26:50 EDT(-0400)] <ecochran> the next merge will be hell!
[12:27:14 EDT(-0400)] <Bosmon> ooo
[12:27:20 EDT(-0400)] <Bosmon> I guess I will update right away (tongue)
[12:28:33 EDT(-0400)] <Bosmon> aaah.... lovely
[12:28:54 EDT(-0400)] <ecochran> I fixed my setting in Aptana to use spaces!
[12:29:29 EDT(-0400)] <Bosmon> Eli - what sort of thing might be in a row that is not "ready"?
[12:29:43 EDT(-0400)] <Bosmon> My current guess is the set of things we want to be "selectable" is the set of things with the class "ready"?
[12:29:57 EDT(-0400)] <ecochran> Window > Preferences > Aptana > Editors > Tab Insertion
[12:30:19 EDT(-0400)] <ecochran> Bosmon: yes
[12:31:52 EDT(-0400)] <ecochran> ready is the default state, as opposed to uploaded or error
[12:32:55 EDT(-0400)] <colinclark> Tomatoes are being thrown?
[12:33:27 EDT(-0400)] <anastasiac> no, no tomatoes. We're all enjoying spaghetti sauce
[12:34:06 EDT(-0400)] <colinclark> yum, sauce
[12:34:24 EDT(-0400)] <colinclark> Here's the official decree on code conventions:
[12:34:34 EDT(-0400)] <colinclark> We all equally disagree by choosing someone else's conventions.
[12:34:42 EDT(-0400)] <colinclark> So, four spaces are it.
[12:34:51 EDT(-0400)] <colinclark> I am a tabbist myself, so I hate my own decree.
[12:34:55 EDT(-0400)] <colinclark> But it stands nonetheless.
[12:34:57 EDT(-0400)] <colinclark> (wink)
[12:35:24 EDT(-0400)] <ecochran> do we have a place to document these "conventions"
[12:35:52 EDT(-0400)] <colinclark> We're not wholly consistent, but the standards are these: http://javascript.crockford.com/code.html
[12:36:09 EDT(-0400)] <colinclark> The benefit of choosing these particular standards is that they can be verified using JSLint for us.
[12:36:28 EDT(-0400)] <colinclark> So, in regards to your tomatoes, throw them Crockford's way. (tongue)
[12:38:13 EDT(-0400)] <colinclark> This is the "blame someone we don't work with" approach.
[12:38:58 EDT(-0400)] <ecochran> as opposed to the usual blame Colin approach? (wink)
[12:40:34 EDT(-0400)] <colinclark> lolk
[12:40:39 EDT(-0400)] <colinclark> that is preferable, yes
[12:40:39 EDT(-0400)] <ecochran> sorry that came out mean... and I meant it affectionately
[12:40:49 EDT(-0400)] <colinclark> no, no, it was hilarious.
[12:41:03 EDT(-0400)] * jessm (n=Jess@rrcs-70-63-141-143.midsouth.biz.rr.com) has joined #fluid-work
[12:41:08 EDT(-0400)] <Bosmon> Yes
[12:41:15 EDT(-0400)] <Bosmon> I find that blaming Colin is perpetually hilarious
[12:41:17 EDT(-0400)] <Bosmon> (tongue)
[12:41:27 EDT(-0400)] <Bosmon> Clearly I mean that affectionately too... (tongue)
[12:41:39 EDT(-0400)] * colinclark dodges tomatoes.
[12:43:00 EDT(-0400)] <Bosmon> Ah, talking of Colinism, I recently created a Colinist set:
[12:43:05 EDT(-0400)] <Bosmon> http://www.flickr.com/photos/8129850@N02/sets/72157606324059055/
[12:43:33 EDT(-0400)] <colinclark> Darcie and the cat are particularly impressive.
[12:43:35 EDT(-0400)] <Bosmon> This one is particularly magisterial: http://www.flickr.com/photos/8129850@N02/2694491306/in/set-72157606324059055/
[12:43:54 EDT(-0400)] <Bosmon> Note the Giant Knees, Watson...
[12:44:20 EDT(-0400)] <colinclark> What is your magnificent lens, Bosmon? jhung is curious.
[12:44:31 EDT(-0400)] <anastasiac> nice pics!
[12:45:02 EDT(-0400)] <colinclark> In this one, you can see the gifts we bestowed on our host: http://www.flickr.com/photos/8129850@N02/2693685607/in/set-72157606324059055/
[12:45:14 EDT(-0400)] <Bosmon> It is the magnificent Canon 10-22mm
[12:45:25 EDT(-0400)] <Bosmon> http://www.kenrockwell.com/canon/1022.htm
[12:45:32 EDT(-0400)] <ecochran> you gave him a cat?
[12:45:43 EDT(-0400)] <jhung> ahhh
[12:45:45 EDT(-0400)] <colinclark> A copy of Crockford, whose Chapter 5 sealed the deal on my preference for that-ism.
[12:45:50 EDT(-0400)] <Bosmon> He also brought over numerous square metres of turf...
[12:46:07 EDT(-0400)] * anastasiac is confused
[12:46:08 EDT(-0400)] <Bosmon> Yes, around 3.30am that night Colin received a mysterious text that consisted of only the word... "that-ism!!!!"
[12:46:13 EDT(-0400)] <jessm> and jessm's cell phone!
[12:46:13 EDT(-0400)] <ecochran> those Canadians never leave home with out a very nice house gift!
[12:46:19 EDT(-0400)] <colinclark> The lawn was challenging to get through security.
[12:46:22 EDT(-0400)] <Bosmon> (smile)
[12:46:37 EDT(-0400)] <Bosmon> Yes, it had to be sterilised first... thus somewhat reducing its qualifications as turf...
[12:46:41 EDT(-0400)] * anastasiac is considering going to find a sane chat room
[12:47:00 EDT(-0400)] <colinclark> The bottle of Amarone in the picture is arguably the only wine that is "robust" enough for Bosmon and I to agree upon.
[12:47:33 EDT(-0400)] * jhung marvels at 10-22 lens, but is saddened it's Canon.
[12:47:43 EDT(-0400)] <Bosmon> I sold all my Nikon kit last month (smile)
[12:48:05 EDT(-0400)] <ecochran> yes, and here's Darcie, after 3 hours of tech talk, seeing if she can sneak over the fence unnoticed
[12:48:05 EDT(-0400)] <ecochran> http://www.flickr.com/photos/8129850@N02/2694487684/in/set-72157606324059055/
[12:48:19 EDT(-0400)] <Bosmon> Unlike Rockwell I am not looking back...
[12:48:22 EDT(-0400)] <colinclark> oh dear
[12:48:33 EDT(-0400)] <colinclark> someone seems to like my wife a bit too much.
[12:48:33 EDT(-0400)] <Bosmon> urk
[12:48:43 EDT(-0400)] <Bosmon> We seem to have attracted a spammer...
[12:48:49 EDT(-0400)] <jhung> lol
[12:48:55 EDT(-0400)] <Bosmon> I can delete it if you want (tongue)
[12:49:07 EDT(-0400)] <jessm> it adds colour
[12:49:22 EDT(-0400)] <colinclark> It's hard to argue with, I will admit. (tongue)
[12:49:31 EDT(-0400)] <colinclark> Okay, I'm getting pizza.
[12:49:37 EDT(-0400)] <colinclark> jhung: Hungry?
[12:49:44 EDT(-0400)] <jhung> yep. Let's go.
[12:49:51 EDT(-0400)] <Bosmon> The CATTT attitude is particularly amusing, if you zoom in to it (tongue)
[12:50:13 EDT(-0400)] <ecochran> Here are my pictures from Paris: http://www.flickr.com/photos/fathermojo/sets/72157606316555700/
[12:50:35 EDT(-0400)] <jessm> ecochran: let's not advertise those too far and wide, eh?
[12:50:44 EDT(-0400)] <ecochran> just in case, you all needed something to do while eating pizza
[12:50:54 EDT(-0400)] <Bosmon> They are great
[12:50:56 EDT(-0400)] <Bosmon> Very colourful (tongue)
[12:50:57 EDT(-0400)] <ecochran> jessm: ok
[12:51:13 EDT(-0400)] * jessm ate too much cheese in Paris
[12:51:31 EDT(-0400)] <jessm> (wink)
[12:51:45 EDT(-0400)] <Bosmon> http://www.flickr.com/photos/fathermojo/2692828445/in/set-72157606316555700/
[12:51:49 EDT(-0400)] <Bosmon> This one is really super
[12:51:55 EDT(-0400)] <Bosmon> Do you travel everywhere with a tripod, Eli? (tongue)
[12:52:25 EDT(-0400)] <Bosmon> !!!!!!!
[12:52:27 EDT(-0400)] <Bosmon> http://www.flickr.com/photos/fathermojo/2693638918/in/set-72157606316555700/
[12:52:34 EDT(-0400)] <Bosmon> Humanity has ALREADY been imprisoned in concrete eggcups!
[13:09:32 EDT(-0400)] <ecochran> now that we've updated to the latest jQuery and jQuery UI, I'm seeing some very odd formatting around the jQuery UI dialog
[13:09:40 EDT(-0400)] <ecochran> in FF the padding is off
[13:09:53 EDT(-0400)] <ecochran> in Safari the dialog no longer resizes when adding content
[13:09:57 EDT(-0400)] <ecochran> grumph!
[13:12:13 EDT(-0400)] <Bosmon> I am starting to boggle how precisely to bind to the "delete" key in JS....
[13:12:24 EDT(-0400)] <Bosmon> It seems like classically one of those things that might require "event normalisation by a library"....
[13:12:52 EDT(-0400)] <Bosmon> Although many of them at least agree that the keycode is 46....
[13:58:42 EDT(-0400)] <Bosmon> OK, so uploader rows are deletable and arrow-selectable....
[13:58:54 EDT(-0400)] <Bosmon> I am stumbling still on the "Apple" modifier
[13:58:55 EDT(-0400)] <Bosmon> http://unixpapa.com/js/key.html
[13:59:04 EDT(-0400)] <Bosmon> This page suggests that we cannot really get at this portably....
[13:59:38 EDT(-0400)] <Bosmon> Suggests that the modifier is 224 in Mozilla, 17 in Opera and 91 or 93 in Safari
[13:59:40 EDT(-0400)] <Bosmon> "lovely"
[14:01:19 EDT(-0400)] * phiggins (n=dante@conference/oscon/x-de981efe4c5beacf) has joined #fluid-work
[14:20:42 EDT(-0400)] * phiggins (n=dante@conference/oscon/x-df952e351686f85f) has joined #fluid-work
[14:26:23 EDT(-0400)] <colinclark> Bosmon: Sorry for the delay. I have had repeated distractions today.
[14:26:28 EDT(-0400)] <colinclark> Let me just read over your various issues.
[14:29:35 EDT(-0400)] <colinclark> ecochran: Do you think these padding and styling issues are addressable for 0.4?
[14:29:54 EDT(-0400)] <colinclark> And have you filed bugs about them. Justin says he filed one in regards to the bottom padding of the dialog...
[14:30:06 EDT(-0400)] <colinclark> Justin_o: Do you remember the bug number of the one you filed?
[14:30:49 EDT(-0400)] <Justin_o> http://issues.fluidproject.org/browse/FLUID-915
[14:30:56 EDT(-0400)] <colinclark> Thanks!
[14:31:06 EDT(-0400)] <Justin_o> np
[14:31:07 EDT(-0400)] <colinclark> Bosmon: I'm just reading over the link you send along.
[14:31:10 EDT(-0400)] <ecochran> it appears to be a jQuery dialog bug
[14:31:35 EDT(-0400)] <ecochran> they added some height settings to the dialog content
[14:32:34 EDT(-0400)] <ecochran> what is odd is that the Safari problem is intermittent, which indicates that it might be a Safari 3 bug interacting with a change in jQuery UI
[14:33:04 EDT(-0400)] <colinclark> ecochran: Interesting. Good to know.
[14:33:04 EDT(-0400)] <ecochran> anyway, if I comment out one line in ui.dialog.js, everything is happy
[14:33:11 EDT(-0400)] <ecochran> but do we want to do that?
[14:33:25 EDT(-0400)] <ecochran> > also I haven't tested it in IE yet
[14:33:25 EDT(-0400)] <colinclark> ecochran: No! (smile)
[14:33:50 EDT(-0400)] <colinclark> It might be good to coordinate with Justin on this one.
[14:34:10 EDT(-0400)] <colinclark> Have we looked at jQuery UI's Trac to see if they are aware of it and have a workaround?
[14:34:32 EDT(-0400)] <colinclark> ecochran: What is the line in ui.dialog.js that is the culprit?
[14:34:47 EDT(-0400)] <colinclark> Bosmon: Everyone is impressed with your keyboard-a11y work on Uploader.
[14:34:57 EDT(-0400)] * anastasiac nods emphatically
[14:35:03 EDT(-0400)] <ecochran> size: function() {
[14:35:03 EDT(-0400)] <ecochran> var container = this.uiDialogContainer,
[14:35:03 EDT(-0400)] <ecochran> titlebar = this.uiDialogTitlebar,
[14:35:03 EDT(-0400)] <ecochran> content = this.element,
[14:35:03 EDT(-0400)] <ecochran> tbMargin = parseInt(content.css('margin-top')) + parseInt(content.css('margin-bottom')),
[14:35:04 EDT(-0400)] <ecochran> lrMargin = parseInt(content.css('margin-left')) + parseInt(content.css('margin-right'));
[14:35:05 EDT(-0400)] <ecochran> //content.height(container.height() - titlebar.outerHeight() - tbMargin);
[14:35:07 EDT(-0400)] <ecochran> content.width(container.width() - lrMargin);
[14:35:09 EDT(-0400)] <ecochran> },
[14:35:16 EDT(-0400)] <colinclark> ecochran: A line number would help. (smile)
[14:35:28 EDT(-0400)] <anastasiac> maybe pastebin?
[14:35:34 EDT(-0400)] <colinclark> line number's easiest
[14:35:39 EDT(-0400)] <colinclark> since we all share the same file.
[14:35:48 EDT(-0400)] <ecochran> 240
[14:35:48 EDT(-0400)] <anastasiac> http://fluid.pastebin.com/
[14:35:55 EDT(-0400)] <colinclark> ecochran: As part of your work removing the checkbox from the tab order, did you also remove the remove box from the tab order?
[14:36:03 EDT(-0400)] <colinclark> The little x icon.
[14:36:11 EDT(-0400)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined #fluid-work
[14:36:34 EDT(-0400)] <ecochran> no
[14:36:52 EDT(-0400)] <ecochran> colinclark: you're referring to the dialog close button?
[14:36:59 EDT(-0400)] <Justin_o> can someone try openning uploader from this page http://build.fluidproject.org/fluid/sample-code/uploader/pop-up/index.html
[14:37:12 EDT(-0400)] <Justin_o> it doesn't seem to be working.. i want to see if it's just me
[14:37:20 EDT(-0400)] <anastasiac> broken on FF3 for me
[14:37:27 EDT(-0400)] <ecochran> not just you, won't work for me either in Safari
[14:37:51 EDT(-0400)] <anastasiac> hm.. let me check that - I updated the server recently, and I think I edited that file. give me a sec
[14:38:02 EDT(-0400)] <Justin_o> okay.. thanks
[14:38:19 EDT(-0400)] <ecochran> getting this error: container.ariaRole is not a function
[14:38:19 EDT(-0400)] <ecochran> http://build.fluidproject.org/fluid/fluid-components/js/jquery/jquery-1.2.6.js
[14:38:19 EDT(-0400)] <ecochran> Line 1
[14:38:52 EDT(-0400)] <anastasiac> ecochran - you're getting that error on the index.html??
[14:39:03 EDT(-0400)] <ecochran> yes
[14:39:06 EDT(-0400)] <ecochran> in FF3
[14:39:22 EDT(-0400)] <anastasiac> Doh!! Ok - my error.
[14:39:31 EDT(-0400)] <anastasiac> the file doesn't include the jARIA.js plugin
[14:39:42 EDT(-0400)] <anastasiac> I'll fix it now, and update the server... let you know when it's ready
[14:39:44 EDT(-0400)] <anastasiac> sorry!!
[14:39:55 EDT(-0400)] <Justin_o> thanks
[14:41:56 EDT(-0400)] <anastasiac> lesson learned: test more thoroughly!
[14:42:15 EDT(-0400)] <colinclark> anastasiac's face is a bit red, I think. (tongue)
[14:42:34 EDT(-0400)] * anastasiac slaps herself on the back of the hand
[14:42:44 EDT(-0400)] <colinclark> Actually, the real problem is cut-and-pasted markup.
[14:42:55 EDT(-0400)] <anastasiac> ok, server should be working now, Justin_o
[14:42:58 EDT(-0400)] <colinclark> I look forward to the future when we can have a single, shared bit of markup.
[14:43:10 EDT(-0400)] <Justin_o> thanks
[14:43:25 EDT(-0400)] <anastasiac> hm... this error was actually in the inclusions in the header...
[14:45:10 EDT(-0400)] <Bosmon> Sorry for inattention, as you know I get no notifications from this thing (tongue)
[14:45:19 EDT(-0400)] <Bosmon> Thanks for the "props" (tongue)
[14:45:31 EDT(-0400)] <Bosmon> I would like to chat now about "event attachment idempotency"....
[14:47:23 EDT(-0400)] <colinclark> Okay.
[14:47:30 EDT(-0400)] <colinclark> Shoot.
[14:47:43 EDT(-0400)] <Bosmon> I am wondering.... whether we actually in real fact ever want to attach "substantive" event handlers to anything at all....
[14:47:48 EDT(-0400)] <Bosmon> But simply to attach "delegational" ones
[14:47:55 EDT(-0400)] <colinclark> Tell me more.
[14:48:13 EDT(-0400)] <Bosmon> This thing occured to me since now our strategies in the Uploader for being i) activatable and ii) selectable are not quite the same
[14:48:40 EDT(-0400)] <Bosmon> We can now "happily" use the declarative "update it now" style for making things selectable
[14:48:51 EDT(-0400)] <Bosmon> But we can't do the same for activatable things, since the event handlers would simply pile up
[14:49:15 EDT(-0400)] <Bosmon> So, I am wondering whether we want to "similarly", but not quite in the same way revise our concept of how activatable works as well
[14:49:23 EDT(-0400)] <Bosmon> And, by extension, all of our event handlers...
[14:49:25 EDT(-0400)] <colinclark> How would this look?
[14:49:47 EDT(-0400)] <Bosmon> That is, to have every "physical" event handler we write, for a start, be namespaced so that only a single instance of it could ever be added to a DOM node
[14:49:53 EDT(-0400)] <colinclark> ecochran: Sorry, I missed your previous comment. Not I'm referring to the little red x icon next to each file row.
[14:50:07 EDT(-0400)] <Bosmon> And secondly, to have as its only activity, to look up a local "that" in the attached DOM node by name and do something with it
[14:50:14 EDT(-0400)] <Bosmon> Rather than attempt to do any real handling by itself...
[14:50:29 EDT(-0400)] <colinclark> Now that we've using the delete key to provide delete-ability, we don't even need to put that button in the tab order.
[14:50:38 EDT(-0400)] <colinclark> Since we've provided a more optimized style.
[14:51:01 EDT(-0400)] <ecochran> colinclark: I removed focus if the remove button has changed to the uploaded checkmark
[14:51:04 EDT(-0400)] <ecochran> otherwise no
[14:51:09 EDT(-0400)] <colinclark> Bosmon: On the surface, this seems rather bizarre.
[14:51:13 EDT(-0400)] <Bosmon> ha
[14:51:26 EDT(-0400)] <Bosmon> It is a Grand Conceptual Dawn, I tell you (tongue)
[14:51:36 EDT(-0400)] <colinclark> ecochran: Okay. Do you have any cycles to remove that altogether, or should someone else do it when they have a chance?
[14:51:55 EDT(-0400)] <Bosmon> I mean, you don't want users to have the affordance to attach multiple instances of a "component"/"event handler" to the same node, do you? (tongue)
[14:52:03 EDT(-0400)] <ecochran> colinclark: I'm confused
[14:52:06 EDT(-0400)] <ecochran> why remove it
[14:52:07 EDT(-0400)] <Bosmon> I'm sure you would prefer it were just impossible...
[14:52:08 EDT(-0400)] <ecochran> ?
[14:52:23 EDT(-0400)] <colinclark> Because files can be deleted using the keyboard directly.
[14:52:38 EDT(-0400)] <ecochran> ah
[14:52:50 EDT(-0400)] <colinclark> ecochran: For the keyboard user, it's slower to have to focus each of these little remove buttons.
[14:53:00 EDT(-0400)] <ecochran> colinclark: how would users know that they can hit delete?
[14:53:21 EDT(-0400)] <colinclark> ecochran: Great question. We have two possible strategies that need to be researched.
[14:53:42 EDT(-0400)] <colinclark> A bit of added text when the row is read, or perhaps if we're lucky there is an ARIA property for this.
[14:53:49 EDT(-0400)] <ecochran> colinclark: it's dead easy for me to remove the focus
[14:54:46 EDT(-0400)] <colinclark> ecochran: If you have the time, go for it.
[14:55:30 EDT(-0400)] <colinclark> Jacob1 is going to look into the necessary labeling tomorrow.
[14:56:31 EDT(-0400)] <colinclark> http://issues.fluidproject.org/browse/FLUID-985
[14:56:35 EDT(-0400)] <Bosmon> With of course the further ultimate goal that event handlers, devolving to have no "content" would essentially all become the same event handler
[14:56:55 EDT(-0400)] <colinclark> Bosmon: Catching up... one second.
[14:57:09 EDT(-0400)] <Bosmon> Therefore, perhaps all "activatable" or whatever requirements across an entire document would be handled by just a single event handler...
[14:57:37 EDT(-0400)] <Bosmon> Leading to magnificent economies!
[14:57:44 EDT(-0400)] <Bosmon> Improvements in efficiency, and safety
[14:58:30 EDT(-0400)] <Bosmon> After all, if we have gone to the trouble binding a scatter of localised "that's" across the entire DOM, why add insult to injury by adding a whole host of separate, but closely correlated and different event handlers to it too?
[14:59:07 EDT(-0400)] <ecochran> colinclark: Done
[14:59:09 EDT(-0400)] <colinclark> I hear we just hit 1000+ issues.
[14:59:11 EDT(-0400)] <colinclark> ecochran: You rule.
[14:59:22 EDT(-0400)] <Bosmon> This policy would mean then that, for example, having set off a "selectable domain" with a selector, like we do in the Uploader, we would not even ever need to update it at all
[14:59:28 EDT(-0400)] <colinclark> Okay, now I will focus on Bosmon's world domination plans...
[15:00:17 EDT(-0400)] <Bosmon> I mean ok, I guess I am really proposing two separate things here
[15:00:27 EDT(-0400)] <Bosmon> Which need not be done at the same time, but address closely related needs
[15:01:02 EDT(-0400)] <ecochran> gotta go get lunch and then go to a UIE seminar for an hour! see you in a bit
[15:01:13 EDT(-0400)] * ecochran (n=ecochran@dhcp-169-229-212-100.LIPS.Berkeley.EDU) has left #fluid-work
[15:01:17 EDT(-0400)] <Bosmon> i) turn "event handlers" into "things with empty content" which simply fish out a JQuery.data at the target node or one of its descendents, leading to ii) coalesce these event handlers into fewer, more global ones, ultimately simply ending up at top level in the document
[15:01:43 EDT(-0400)] <colinclark> Bosmon: I fear you are a madman.
[15:01:47 EDT(-0400)] <Bosmon> (smile) (smile)
[15:02:08 EDT(-0400)] <Bosmon> YET.... there are greater madmen in the world even than I (tongue)
[15:02:17 EDT(-0400)] <Bosmon> Did anastasiac manage to find a source yet? (tongue)
[15:02:33 EDT(-0400)] <Bosmon> http://cgi.ebay.co.uk/Bullshot-New-DVD_W0QQitemZ110273442617QQcmdZViewItem?hash=item110273442617&amp;_trksid=p3286.m14.l1318
[15:02:44 EDT(-0400)] <Bosmon> This one would do, I don't think there is anything more local...
[15:03:04 EDT(-0400)] <anastasiac> I tried zip.ca (the service my boyfriend subscribes to) but they don't have it
[15:03:18 EDT(-0400)] <anastasiac> I'm going to check a couple of rental places near me that have good collections
[15:03:25 EDT(-0400)] <Bosmon> Well
[15:03:32 EDT(-0400)] <Bosmon> I am sure it is the sort of thing you would want to watch again and again (smile)
[15:03:41 EDT(-0400)] * colinclark laughs!
[15:03:50 EDT(-0400)] <colinclark> Over and over and over again.
[15:04:11 EDT(-0400)] <Bosmon> Just look at the comments....
[15:04:16 EDT(-0400)] <Bosmon> Possibly one of the best movies ever made!, 7 October 2003
[15:04:16 EDT(-0400)] <Bosmon> Author: Ian_H from Buffalo, NY, USA
[15:04:16 EDT(-0400)] <Bosmon> I remember when this movie was first released. I loved it and saw it many many times. I bought it on video some years later and watched it over and over again, sharing it with friends as often as possible.
[15:04:41 EDT(-0400)] <colinclark> anastasiac: With the dollar conversation, that's like 8 billion Canadian dollars.
[15:04:42 EDT(-0400)] <colinclark> (tongue)
[15:04:47 EDT(-0400)] <colinclark> conversion
[15:06:07 EDT(-0400)] <colinclark> Sometime, you should all try Antranig's recipe for biscuits...
[15:06:22 EDT(-0400)] <Bosmon> ..... and witness the creation of one of the HARDEST SUBSTANCES KNOWN TO MAN?
[15:07:00 EDT(-0400)] <Bosmon> "One of the funniest films I've ever seen, 28 Dec 2006 By J. Bassett - See all my reviews
[15:07:00 EDT(-0400)] <Bosmon>
[15:07:00 EDT(-0400)] <Bosmon> My entire family love this film and quote it constantly."
[15:07:47 EDT(-0400)] <Bosmon> Anyway, yes
[15:07:54 EDT(-0400)] <Bosmon> Event handler application coalescence?
[15:08:24 EDT(-0400)] <Bosmon> Essentially there would just be one event handler in the world for each type of "thing"
[15:09:11 EDT(-0400)] <Bosmon> colinclark: Any further ideas re. the Apple key?
[15:10:09 EDT(-0400)] <colinclark> Bosmon: I am constantly being asked questions. Give me another moment. (smile)
[15:11:24 EDT(-0400)] <Bosmon> Hmm... even after appointing Jess, we still need 2 Colins...
[15:11:41 EDT(-0400)] <Bosmon> I'm sure there is some rule of numbers at work here...
[15:15:30 EDT(-0400)] * phiggins (n=dante@conference/oscon/x-8f4a5d67e8827ba7) has joined #fluid-work
[15:24:17 EDT(-0400)] <colinclark> Bosmon: Let me contemplate your idea about event binding for awhile.
[15:24:42 EDT(-0400)] <colinclark> In the meantime, we should definitely determine how to deal with apparent inconsistencies in browser support for the meta key. (sad)
[15:25:14 EDT(-0400)] <colinclark> Bosmon: It should be something jQuery deals with for us.
[15:25:28 EDT(-0400)] <Bosmon> Yes
[15:25:35 EDT(-0400)] <Bosmon> I looked at the code, and unfortunately there is just one line there
[15:25:38 EDT(-0400)] <colinclark> But key abstraction tends to be a weak point.
[15:26:41 EDT(-0400)] <colinclark> I'm going to whip up a quick test.
[15:29:55 EDT(-0400)] <Bosmon> OK, I have updated the Keyboard a11y API page to reflect the new API...
[15:31:23 EDT(-0400)] * ecochran (n=ecochran@dhcp-169-229-212-100.LIPS.Berkeley.EDU) has joined #fluid-work
[15:31:36 EDT(-0400)] <ecochran> I'm back... too much to do
[15:43:24 EDT(-0400)] * davidb (n=davidb@142.150.154.101) has joined #fluid-work
[15:43:25 EDT(-0400)] <jessm> Bosmon: we will evermore need 2 Colins
[15:43:38 EDT(-0400)] <davidb> anastasiac, Jacob1 http://simon.html5.org/test/aria/
[15:44:37 EDT(-0400)] <davidb> for testing on windows
[15:47:32 EDT(-0400)] <davidb> anastasiac: Jacob1: aria-describedby is for 'more detailed information'... so should work... maybe have to tweak screen reader verbosity settings
[15:49:00 EDT(-0400)] <davidb> anastasiac: Jacob1 : aria-describedby should be converted into MSAA AccDescription
[15:49:30 EDT(-0400)] <davidb> so "green light"
[15:49:38 EDT(-0400)] * davidb talks to himself
[15:50:21 EDT(-0400)] <Jacob1> sorry bout that!
[15:50:41 EDT(-0400)] <colinclark> jessm: I don't think I could deal with two of me. (tongue)
[15:50:42 EDT(-0400)] <Jacob1> ok, awesome
[15:50:58 EDT(-0400)] <colinclark> "blah, blah, blah" all day long.
[15:50:59 EDT(-0400)] <davidb> colinclark: not many people could deal with another of themselves.
[15:51:03 EDT(-0400)] <Jacob1> does the MSAA conversion happen automatically?
[15:51:14 EDT(-0400)] <davidb> colinclark: heck i get scared everytime i look in the mirror
[15:51:25 EDT(-0400)] * colinclark laughs.
[15:51:36 EDT(-0400)] <davidb> colinclark: seriously sometimes i wonder who is looking back at me
[15:51:47 EDT(-0400)] <davidb> but he looks annoying
[15:52:20 EDT(-0400)] <ecochran> colinclark: fixed the padding and clipping bugs using a slightly evil (but legal) CSS technique of marking a bit of style !important
[15:53:08 EDT(-0400)] <colinclark> ecochran: Did you ever give me a line number into the ui.dialog.js source/
[15:53:08 EDT(-0400)] <davidb> Jacob1: sorry missed that... yes automatically
[15:53:09 EDT(-0400)] <colinclark> ?
[15:53:10 EDT(-0400)] <colinclark> I forget
[15:53:15 EDT(-0400)] <ecochran> yes I did
[15:53:30 EDT(-0400)] <ecochran> 240
[15:53:36 EDT(-0400)] <davidb> Jacob1: so, FF3 converts aria-describedby into an MSAA AccDescription on windows
[15:53:42 EDT(-0400)] <ecochran> colinclark: 240
[15:53:56 EDT(-0400)] <colinclark> Cool, thanks.
[15:54:03 EDT(-0400)] <colinclark> I will take a peek.
[15:54:40 EDT(-0400)] <colinclark> Bosmon: I made a little test file for the meta key...
[15:54:49 EDT(-0400)] <davidb> Jacob1: this is one place i would say testing with jaws may be premature... (depends on goals again)
[15:54:57 EDT(-0400)] <colinclark> Justin is assembling a table of the thoroughly inconsistent results he is seeing.
[15:55:04 EDT(-0400)] <colinclark> Since he has every browser under the sun.
[15:55:11 EDT(-0400)] <davidb> but jaws sometimes overrides accessible name and description... tries to be too clever.
[15:55:17 EDT(-0400)] <davidb> Jacob1: ^
[15:55:22 EDT(-0400)] <ecochran> has Bosmon gone to sleep?
[15:55:33 EDT(-0400)] <davidb> Jacob1: Window-Eyes will be fine though.
[15:55:44 EDT(-0400)] <colinclark> ecochran: He doesn't get notifications when people use his nick.
[15:55:52 EDT(-0400)] <Jacob1> ok
[15:55:55 EDT(-0400)] <colinclark> He'll notice eventually. (smile)
[15:56:56 EDT(-0400)] <ecochran> Anyone know why I'm suddenly getting this error in the live version of the Uploader?
[15:56:57 EDT(-0400)] <ecochran> container.ariaRole is not a function
[15:56:57 EDT(-0400)] <ecochran> http://localhost:8080/sakai-imagegallery2-web/fluid-components/js/jquery/jquery-1.2.6.js
[15:56:57 EDT(-0400)] <ecochran> Line 2101
[15:57:56 EDT(-0400)] <anastasiac> ecochran, we added aria support. This error indicates that someone (sad) forgot to add the jARIA.js file to an html file
[15:58:02 EDT(-0400)] <anastasiac> which live link were you trying to use?
[15:58:08 EDT(-0400)] <ecochran> my own
[15:58:24 EDT(-0400)] <ecochran> that was the first thing that I checked for
[15:58:27 EDT(-0400)] <ecochran> it is referenced
[15:58:33 EDT(-0400)] <ecochran> I'm confused
[15:58:39 EDT(-0400)] <anastasiac> I'm also confused
[15:58:43 EDT(-0400)] <ecochran> maybe I need to do a clean install
[15:58:48 EDT(-0400)] <ecochran> I'll get back to you
[15:58:50 EDT(-0400)] <anastasiac> is your path right?
[15:58:57 EDT(-0400)] <ecochran> hmm, should be
[15:58:59 EDT(-0400)] <anastasiac> the path to jARIA.js
[15:59:15 EDT(-0400)] <anastasiac> it's in the fluid-components/js/jquery folder
[15:59:22 EDT(-0400)] <ecochran> yep
[15:59:31 EDT(-0400)] <anastasiac> browser cache emptied?
[15:59:59 EDT(-0400)] <ecochran> I always run with the cache disabled when testing
[16:00:23 EDT(-0400)] <anastasiac> ah, tomcat - sometimes deploying is funky, and I've had to actually remove the sakai tool from the webapps folder before redeploying
[16:00:31 EDT(-0400)] <anastasiac> never figured out why
[16:02:54 EDT(-0400)] <Bosmon> Yes, it is funky
[16:03:05 EDT(-0400)] <Bosmon> It is all due to the exact timings it looks at different datestamps...
[16:03:49 EDT(-0400)] <Bosmon> There is nothing in Maven/unzip that makes it necessarily write to web.xml before it writes to any of the other files, so you are always taking "pot luck" that it will manage to deploy the entire app before the container notices (tongue)
[16:03:58 EDT(-0400)] <Bosmon> These problems seem a lot better in Tomcat 6...
[16:04:44 EDT(-0400)] <Bosmon> Yes, if you want to get my attention, best to send me an individual IM (tongue)
[16:05:07 EDT(-0400)] <Bosmon> colinclark: What shall I do with this test you have created? (tongue)
[16:05:52 EDT(-0400)] <colinclark> Bosmon: I'd suggest we're going to have to do some browser normalization ourselves for this. (sad)
[16:06:08 EDT(-0400)] <Bosmon> colinclark: Isn't it even worse than that? (tongue)
[16:06:21 EDT(-0400)] <colinclark> I'm not certain.
[16:06:24 EDT(-0400)] <ecochran> anastasiac: my mistake
[16:06:28 EDT(-0400)] <ecochran> all fixed
[16:06:48 EDT(-0400)] <anastasiac> ecochran, glad it's fixed
[16:26:43 EDT(-0400)] <anastasiac> ecochran, question about the swfupload object and the uploadUrl:
[16:26:53 EDT(-0400)] <ecochran> shoot
[16:27:05 EDT(-0400)] <anastasiac> does the swfupload object call the script at that url once per file, or once after all files are uploaded?
[16:27:53 EDT(-0400)] <ecochran> hmm
[16:29:25 EDT(-0400)] <ecochran> anastasiac: don't completely know... but I would assume that it would have to use that URL for each file (in fact for every communication with the server)
[16:29:49 EDT(-0400)] <ecochran> it is used internally for communication btwn the swf and server
[16:30:00 EDT(-0400)] <anastasiac> yes, that's what I understood
[16:30:35 EDT(-0400)] <anastasiac> do we have any docs about what the communication is? is it just an http post? what info is communicated?
[16:31:22 EDT(-0400)] * anastasiac looks at http://www.swfupload.org/documentation
[16:34:00 EDT(-0400)] <anastasiac> http://www.swfupload.org/documentation/howitworks says:
[16:34:14 EDT(-0400)] <anastasiac> 7) The server receives each upload one at a time. From the server's point of view each upload appear to have come from a standard upload page. There is nothing extra to install or configure on the server in order to use SWFUpload.
[16:34:28 EDT(-0400)] <anastasiac> Sounds to me like it's one communication per file
[16:35:04 EDT(-0400)] <davidb> Jacob1, anastasiac your question led me to updating http://developer.mozilla.org/en/docs/ARIA_User_Agent_Implementors_Guide#11.3.10_States_and_Object_Properties so thanks!
[16:35:26 EDT(-0400)] * Justin_o (n=Justin@142.150.154.235) has left #fluid-work
[17:05:12 EDT(-0400)] <ecochran> anastasiac: yes
[17:34:42 EDT(-0400)] * colinclark (n=colin@142.150.154.101) has joined #fluid-work
[17:35:09 EDT(-0400)] * theclown (n=theclown@guiseppi.atrc.utoronto.ca) has left #fluid-work
[17:46:13 EDT(-0400)] * phiggins (n=dante@conference/oscon/x-ac733d1a178e7345) has joined #fluid-work
[17:57:30 EDT(-0400)] * phiggins (n=dante@conference/oscon/x-e42b4c4a4e887cbc) has joined #fluid-work
[18:04:06 EDT(-0400)] * apetro-_ (n=apetro@12.164.139.7) has joined #fluid-work
[18:06:17 EDT(-0400)] * Bosmon (n=Antranig@ginger.caret.cam.ac.uk) has joined #fluid-work
[18:08:21 EDT(-0400)] * phiggins (n=dante@conference/oscon/x-513e309f2bceca67) has joined #fluid-work
[18:21:36 EDT(-0400)] * ecochran (n=ecochran@dhcp-169-229-212-100.LIPS.Berkeley.EDU) has left #fluid-work
[18:57:40 EDT(-0400)] * apetro-- (n=apetro@wsip-98-174-242-39.ph.ph.cox.net) has joined #fluid-work
[20:31:28 EDT(-0400)] * phiggins (n=dante@71-34-102-251.ptld.qwest.net) has joined #fluid-work