fluid-work IRC Logs-2008-12-09

[08:30:08 EST(-0500)] * Justin_o (n=Justin@142.150.154.101) has joined #fluid-work
[08:56:16 EST(-0500)] * jessm (n=Jess@c-24-34-214-137.hsd1.ma.comcast.net) has joined #fluid-work
[09:00:51 EST(-0500)] * colinclark (n=colin@bas1-toronto09-1279621226.dsl.bell.ca) has joined #fluid-work
[09:02:44 EST(-0500)] * Everett (n=Mirc4Lif@fctnnbsc16w-156034216002.pppoe-dynamic.nb.aliant.net) has joined #fluid-work
[09:03:01 EST(-0500)] <Everett> Morning
[09:09:48 EST(-0500)] <colinclark> Everett: Morning. I'm super overdue to send you an email with some links. I have been swamped getting ready for the Infusion 0.6 release, but I haven't forgotten.
[09:10:00 EST(-0500)] <colinclark> How are things going with you?
[09:11:10 EST(-0500)] <Everett> Quite well. I am currently installing Ubuntu 8.04 in Virtualbox on a WinXP host. I tried it with VMWare but responsiveness of Orca was poor at best. This seems like it will work really well. Once I make sure it works I'm going to install OpenSOlaris and use it to test ARIA widgets all day.
[09:11:32 EST(-0500)] <colinclark> Fantastic!
[09:11:39 EST(-0500)] <Everett> Accessible installers are great (ahem Microsoft).
[09:12:10 EST(-0500)] <colinclark> I'll send you a couple of links to Fluid components real quick, in case you find time to check them out.
[09:13:28 EST(-0500)] <Everett> Ok, via e-mail please.
[09:13:54 EST(-0500)] <Everett> When you guys test do you use the latest nightly of FF 3.1?
[09:15:06 EST(-0500)] * Bosmon (n=Antranig@ginger.caret.cam.ac.uk) has joined #fluid-work
[09:15:10 EST(-0500)] <colinclark> We haven't yet started testing on FF 3.1, no. We tend to wait for the betas to bake a little bit, and then switch over then.
[09:15:20 EST(-0500)] <Bosmon> What on earth are you doing up....
[09:15:22 EST(-0500)] <colinclark> Justin_o: Have you done any testing with FF 3.1 yet?
[09:15:34 EST(-0500)] <colinclark> Bosmon: It's 9 am!
[09:16:11 EST(-0500)] <colinclark> Though I am still making coffee.
[09:16:35 EST(-0500)] <Bosmon> "Using AJAX to develop sites, I've run into problems using IE 6.0 RTM
[09:16:35 EST(-0500)] <Bosmon> releases. For some strange reason, the GDR releases of IE don't exhibit the
[09:16:35 EST(-0500)] <Bosmon> same crash, Firefox and others work just fine of course.
[09:16:35 EST(-0500)] <Bosmon> Here's the burning question, how do we change IE from a RTM to a GDR
[09:16:35 EST(-0500)] <Bosmon> version?"
[09:16:37 EST(-0500)] <Justin_o> colinclark: in regards to accessibility testing mostly it has been using IE, only a bit in FF, can't remember off hand how much was in FF2 vs FF3
[09:17:21 EST(-0500)] <colinclark> Justin_o: In the end, our most fruitful testing for AT support will be on Firefox, since IE doesn't have any support for ARIA yet.
[09:17:25 EST(-0500)] <Justin_o> Bosmon: I read that page too... did you happen to take a snapshot of your vm
[09:17:33 EST(-0500)] <Bosmon> NO I DID NOT!
[09:17:35 EST(-0500)] <Bosmon> IT IS GONE FOREVER!
[09:17:49 EST(-0500)] <colinclark> Bosmon: What is this reference?
[09:17:50 EST(-0500)] <Bosmon> I never dreamed anything like this could happen
[09:17:58 EST(-0500)] <Bosmon> Just some bloody page...
[09:18:07 EST(-0500)] <Bosmon> Look at my comment on FLUID-1540
[09:18:18 EST(-0500)] <Bosmon> I installed VWD 2008 on my virtual machine
[09:18:33 EST(-0500)] <Bosmon> And it somehow "bumped" the IE6 on the machine to some other kind of IE6 which no longer exhibits the problem
[09:19:27 EST(-0500)] <Bosmon> http://blogs.conchango.com/markwilson/archive/2005/07/11/1788.aspx
[09:19:39 EST(-0500)] <Bosmon> This message suggests the actual version string itself may be irrelevant
[09:19:53 EST(-0500)] <Bosmon> The "function" of IE can simply change through applied patches....
[09:20:03 EST(-0500)] <Justin_o> colinclark: this is something I'm hoping to do more of in the coming year
[09:20:10 EST(-0500)] <colinclark> Cool.
[09:20:13 EST(-0500)] <Justin_o> sorry for the late reply... i've been buried
[09:20:25 EST(-0500)] <Justin_o> under lots of text
[09:20:55 EST(-0500)] <Bosmon> I'm sorry to say I don't feel it a very high priority to throw very many more resources at this ridiculous issue
[09:20:57 EST(-0500)] <Bosmon>
[09:21:38 EST(-0500)] <colinclark> Bigger fish to fry, I agree.
[09:22:48 EST(-0500)] <Justin_o> I leave it on the bug parade, in case you guys get through all of the other bugs this week and need something else to do
[09:22:54 EST(-0500)] <Bosmon> ho ho
[09:23:05 EST(-0500)] <Bosmon> That bug by itself could take as long as a week
[09:23:31 EST(-0500)] <Justin_o> really... it's definitely not worth spending a week on
[09:24:13 EST(-0500)] <Justin_o> Bosmon: how is looking for pager and the updates to inline edit
[09:24:30 EST(-0500)] <Bosmon> Should be fine
[09:24:35 EST(-0500)] <Bosmon> I need to deal with "selectable" first
[09:24:39 EST(-0500)] <Bosmon> That is a blocker of blockers
[09:24:59 EST(-0500)] <Justin_o> ah yes... that one is important
[09:28:04 EST(-0500)] <Bosmon> IE6 users are still hovering around 20% of the market....
[09:28:49 EST(-0500)] <Justin_o> That's still a lot, do you know what FF is at?
[09:29:27 EST(-0500)] <Bosmon> FF is roughly the same
[09:29:30 EST(-0500)] <Bosmon> Counting all versions
[09:29:46 EST(-0500)] <Bosmon> IE7 has nearly 50% of the market
[09:29:47 EST(-0500)] <Everett> Does fluid have a grid control that would allow for col1 to have a label and col2 / col3 to have a checkbox?
[09:30:01 EST(-0500)] <Bosmon> Everett - yes
[09:30:06 EST(-0500)] <Bosmon> The renderer will do this
[09:30:14 EST(-0500)] <Bosmon> Unfortunately we have not had time to write samples for it yet
[09:30:14 EST(-0500)] <Justin_o> Bosmon: thanks for the info...
[09:30:24 EST(-0500)] <Bosmon> After 0.6 is out the door, we will concentrate on this
[09:31:32 EST(-0500)] <Bosmon> Opera still around a miserable 1%
[09:31:33 EST(-0500)] <Everett> bosmon: thanks, just curious. Theoretically could there be collapsable rows?
[09:31:40 EST(-0500)] <Bosmon> Yes
[09:31:43 EST(-0500)] <Bosmon> You can do what you want
[09:31:51 EST(-0500)] <Bosmon> The renderer will deal with any HTML/JS combination you throw at it
[09:32:11 EST(-0500)] <Bosmon> Chrome has already equalled Opera's userbase...
[09:32:19 EST(-0500)] <Everett> is there a url with useful information I could read about that?
[09:32:23 EST(-0500)] <Bosmon> Yes
[09:32:34 EST(-0500)] <Bosmon> http://wiki.fluidproject.org/display/fluid/Fluid+Renderer
[09:32:37 EST(-0500)] <Bosmon> It has rotted a little
[09:32:51 EST(-0500)] <Bosmon> But you can read it for a general flavour of what is possible
[09:32:59 EST(-0500)] <Everett> ok, thanks.
[09:32:59 EST(-0500)] <colinclark> Everett: One point to clarify about this. We have specifically not created a formalized grid component at all.
[09:33:06 EST(-0500)] <Bosmon> Anastasia has been working on some samples
[09:33:11 EST(-0500)] <Bosmon> But they are not yet written up
[09:33:21 EST(-0500)] <Everett> colin: understandable as the purpose of a grid is flexibility.
[09:33:30 EST(-0500)] <Bosmon> The majority of "hard info" about what the renderer does right now is sitting in the test cases....
[09:33:31 EST(-0500)] <colinclark> Exactly. We've found that every attempt to do this results in a bloated, awkward, and hard to customize thing.
[09:34:11 EST(-0500)] <colinclark> Rather, we've tried to address some of the data binding challenges inherent in the problem, and then allow the user to build their own markup and so on to build their own grid or other type of widget.
[09:34:12 EST(-0500)] <Everett> Is there anything partly complete that I may be able to peek at next week when I'm in TO? Or, is it not quite that far yet?
[09:34:24 EST(-0500)] <Bosmon> Oh, there is plenty to peek at
[09:34:25 EST(-0500)] <Bosmon>
[09:34:29 EST(-0500)] <colinclark> Bosmon: The baseball table example is close, isn't it?
[09:34:37 EST(-0500)] <Bosmon> There is also the pager sample I just wrote
[09:34:41 EST(-0500)] <colinclark> Ah, yes.
[09:34:48 EST(-0500)] <Bosmon> By next week, there may even be a CLE-12 implementation
[09:34:53 EST(-0500)] <Everett> Are you guys free at all on Mon - Tue (Dec 15 - 16)?
[09:35:05 EST(-0500)] <Bosmon> Well, I will be around... here that is
[09:35:06 EST(-0500)] <Bosmon>
[09:35:15 EST(-0500)] <colinclark> Bosmon lives in Cambridge, UK.
[09:35:42 EST(-0500)] <colinclark> But we can always bring him in over skype to chat with us.
[09:35:57 EST(-0500)] <colinclark> I'm free either of those days, Everett.
[09:36:14 EST(-0500)] <Everett> Ok, I'll get finalization (haha) of my meeting times later this week and perhaps we can coordinate something.
[09:44:44 EST(-0500)] <colinclark> Everett: Sounds great.
[09:46:34 EST(-0500)] <Everett> Colin: thanks, talk to you later.
[09:46:37 EST(-0500)] * Everett (n=Mirc4Lif@fctnnbsc16w-156034216002.pppoe-dynamic.nb.aliant.net) has left #fluid-work
[09:49:09 EST(-0500)] * fj4000 (n=Jacob@142.150.154.106) has joined #fluid-work
[09:50:13 EST(-0500)] <colinclark> fj4000 is back!
[09:50:34 EST(-0500)] <fj4000> hey hey
[09:51:16 EST(-0500)] * fj4000 is thinking expired tuna + long subway ride dont mix
[09:51:17 EST(-0500)] * clown (n=clown@guiseppi.atrc.utoronto.ca) has joined #fluid-work
[10:06:25 EST(-0500)] * anastasiac (n=team@142.150.154.160) has joined #fluid-work
[10:06:25 EST(-0500)] <colinclark> fj4000: eeeww
[10:06:28 EST(-0500)] <colinclark> You ok?
[10:07:24 EST(-0500)] <fj4000> yeah.....lets just say, I will be thinking twice before I ever eat tuna I dont make myself
[10:10:06 EST(-0500)] <Bosmon> !!!??!?!!
[10:15:19 EST(-0500)] * colinclark (n=colin@bas1-toronto09-1279621226.dsl.bell.ca) has joined #fluid-work
[10:19:02 EST(-0500)] <Bosmon> colinclark:
[10:19:08 EST(-0500)] <Bosmon> We have an interesting "type" issue
[10:19:27 EST(-0500)] <Bosmon> In "forcing" every bridged function to return "this", we may be going a bridge too far....
[10:24:07 EST(-0500)] <colinclark> Bosmon: eek
[10:24:07 EST(-0500)] <colinclark> yes
[10:24:32 EST(-0500)] <Justin_o> I have made a small update to the commit policy during code freeze and bug parade
[10:24:34 EST(-0500)] <Justin_o> http://wiki.fluidproject.org/display/fluid/Coding+and+Commit+Standards
[10:24:59 EST(-0500)] <Justin_o> basically when code is reviewed the result needs to be indicated on the jira
[10:26:08 EST(-0500)] <Bosmon> OK
[10:26:14 EST(-0500)] <Bosmon> What shall we do?
[10:26:14 EST(-0500)] <colinclark> Justin_o: Ok, good to know.
[10:26:22 EST(-0500)] <Bosmon> For example - the function formerly known as "tabindex"
[10:26:28 EST(-0500)] <Bosmon> Do we follow the letter of the law?
[10:26:38 EST(-0500)] <Bosmon> And insist that no genuine JQuery UI plugin can return an integer?
[10:26:47 EST(-0500)] <colinclark> no, that's ridiculous
[10:26:53 EST(-0500)] <Bosmon> Is it?
[10:26:56 EST(-0500)] <colinclark> yes!
[10:27:04 EST(-0500)] <Bosmon> We have "Guidelines"!
[10:27:05 EST(-0500)] <colinclark> i mean, think about methods like .attr()
[10:27:10 EST(-0500)] <Bosmon> OK
[10:27:12 EST(-0500)] <colinclark> in jQuery
[10:27:12 EST(-0500)] <Bosmon> Well
[10:27:23 EST(-0500)] <Bosmon> The question is, how do we then distinguish these methods?
[10:27:42 EST(-0500)] <Bosmon> We could say, that anything that returns a primitive type, will continue to return a primitive type
[10:28:06 EST(-0500)] <colinclark> I suppose that's a reasonable start...
[10:28:13 EST(-0500)] <colinclark> I suspect you'll find other exceptions to the rule.
[10:28:28 EST(-0500)] <Bosmon> Well, then we will struggle
[10:28:45 EST(-0500)] <colinclark> Try it in the keyboard plugin first.
[10:28:48 EST(-0500)] <Bosmon> ok
[10:28:49 EST(-0500)] <colinclark> You will probably succeed there.
[10:28:51 EST(-0500)] <Bosmon>
[10:28:55 EST(-0500)] <colinclark> And that's all we really need for now.
[10:29:33 EST(-0500)] <colinclark> Today is going to be a rough day for me with grant proposal writing... so you will see me disappear a lot.
[10:29:47 EST(-0500)] <Bosmon> ok
[10:29:51 EST(-0500)] <Bosmon> Which one are you writing today?
[10:29:59 EST(-0500)] <colinclark> Engagement.
[10:30:04 EST(-0500)] <Bosmon> cool
[10:30:14 EST(-0500)] <colinclark> It's very exciting, yes.
[10:30:23 EST(-0500)] <Bosmon> Will we get to detect all sorts of circular and elliptical bar-codes?
[10:30:31 EST(-0500)] <colinclark> Possibly, yes.
[10:30:49 EST(-0500)] <colinclark> Our colleague Jorge and his team have also been working on some fairly compelling wifi-based positioning.
[10:31:06 EST(-0500)] <colinclark> It may well be suitable as well. He thinks it only requires about 3 routers in a space to get sufficient resolution.
[10:31:22 EST(-0500)] <colinclark> http://scampus.blogspot.com/
[10:32:12 EST(-0500)] <colinclark> Anyway, I'm going to hop on the subway and get into work. I was going to work at home but would rather be in the office, close to the grant frenzy.
[10:54:12 EST(-0500)] <fj4000> colinclark got a q for u
[10:54:24 EST(-0500)] <fj4000> dang
[10:54:34 EST(-0500)] <fj4000> just missed him.....by 22 mins
[10:55:01 EST(-0500)] <fj4000> Bosmon, can I throw a puzzle your way?
[10:55:37 EST(-0500)] <Bosmon> oh yes
[10:55:45 EST(-0500)] <Bosmon> As if I don't have enough
[10:55:45 EST(-0500)] <fj4000> its regarding that.locate....
[10:55:48 EST(-0500)] <fj4000>
[10:56:03 EST(-0500)] <fj4000> so in UI Options, I need to fix bug 1925
[10:56:32 EST(-0500)] <fj4000> scoping the preview window to an iframe works well, BUT
[10:56:52 EST(-0500)] <fj4000> now I need to modify the selector set that deals with managing the preview window
[10:57:43 EST(-0500)] <fj4000> what I see in the defaults is simple a selector ".preview", but I need to get that.locate to find ".preview" within the iframe
[10:58:07 EST(-0500)] <Bosmon> Well....
[10:58:10 EST(-0500)] <fj4000> so I thought I could change the default to $('iframe').contents().find('.preview')
[10:58:29 EST(-0500)] <fj4000> but, this isnt flexible, in that if there is no/multiple iframes, it blows up
[10:58:35 EST(-0500)] <Bosmon> I see
[10:58:55 EST(-0500)] <fj4000> so, how would you recommend the default selectors look when it comes to this?
[10:58:58 EST(-0500)] <Bosmon> Why isn't the "container" for the component already correctly localised inside the iframe?
[10:59:51 EST(-0500)] <fj4000> im not sure I understand what you mean by localised....it is scoped within the component, yes
[10:59:53 EST(-0500)] <fj4000> but
[11:00:11 EST(-0500)] <fj4000> there isnt any method in that.locate to normalize iframe content
[11:00:18 EST(-0500)] <fj4000> at least I didnt see any
[11:00:21 EST(-0500)] <fj4000> so im not sure
[11:00:24 EST(-0500)] <Bosmon> What?
[11:00:27 EST(-0500)] <Bosmon> I said the opposite
[11:00:45 EST(-0500)] <Bosmon> Why isn't the container, to which this DOM binder is bound, inside the right iframe?
[11:01:21 EST(-0500)] <fj4000> it shouldnt be in the iframe...only a peice of the component needs to be in an iframe
[11:01:31 EST(-0500)] <fj4000> just the real-time preview part
[11:01:36 EST(-0500)] <fj4000> not the options parts
[11:02:24 EST(-0500)] <fj4000> basically, is it possible now to use that.locate to pluck nodes from an iframe?
[11:02:42 EST(-0500)] <fj4000> or is that not part of its capabilities right now?
[11:03:15 EST(-0500)] <Bosmon> !?!?!?!
[11:03:30 EST(-0500)] <Bosmon> Well ok, then you can pass the "scoping component" as the 2nd argument to "locate"
[11:03:50 EST(-0500)] <fj4000> ok, so for an iframe, would the scope looke like
[11:03:50 EST(-0500)] <Bosmon> That scoping component can be some separately provided configuration to the component
[11:03:56 EST(-0500)] <fj4000> $('iframe').contents() ?
[11:04:20 EST(-0500)] <Bosmon> I am not sure, I have never searched for things across iframes
[11:04:29 EST(-0500)] <Bosmon> It is the sort of thing Justin_o might know though
[11:04:38 EST(-0500)] <Bosmon> He had to do a lot of frames-based stuff for the doh.robot
[11:04:44 EST(-0500)] <fj4000> yes, he showed me how jQuery does it
[11:04:58 EST(-0500)] <fj4000> $('iframe').contents().find('body.preview').doWhatever()
[11:05:08 EST(-0500)] <Bosmon> ok
[11:05:23 EST(-0500)] * Everett (n=Mirc4Lif@fctnnbsc16w-156034216002.pppoe-dynamic.nb.aliant.net) has joined #fluid-work
[11:05:28 EST(-0500)] <Bosmon> Then yes, it looks like the correct node will be $('iframe').contents() as you say
[11:05:38 EST(-0500)] <Everett> Does someone have a second to discuss expected behaviour for inline edit?
[11:05:49 EST(-0500)] <Bosmon> Hi Everett, yes
[11:07:09 EST(-0500)] <Everett> bosmon: 1. Using JAWS 10. 2. Press enter on button 3. find edit control and enter forms mode. 4. make changes to text and press enter. 5. press enter on button again to commit changes.
[11:08:32 EST(-0500)] <Everett> I think my two concerns with this are: 1. most users will be confused by this being announced as a button. 2. Having to find the button again to commit changes seems non-intuitive, however taking into consideration multi-line edits there may not be a better way.
[11:09:13 EST(-0500)] <Everett> Could the text on the button change to "Save changes" or something more intuitive once the edit control has been activated?
[11:21:48 EST(-0500)] <Bosmon> Sorry, I lost attention on this window
[11:22:04 EST(-0500)] <Bosmon> I have to say I am using a client that doesn't notify me of IRC things, so if I seem to be somnolent you should just PM me
[11:22:06 EST(-0500)] <Bosmon> I will see that
[11:22:38 EST(-0500)] <Bosmon> Hi Everett - the button is not actually part of the InlineEdit control itself at all
[11:22:42 EST(-0500)] <Bosmon> But just part of that particular sample
[11:22:51 EST(-0500)] * colinclark (n=colin@142.150.154.101) has joined #fluid-work
[11:23:01 EST(-0500)] <Bosmon> If you can come up with something you think is more appropriate, it would be enthusiastically accepted
[11:29:02 EST(-0500)] <Everett> Is the Bosmon: is the caret placed in that example where a sighted user clicks?
[11:36:13 EST(-0500)] <Bosmon> Can you be more specific?
[11:37:21 EST(-0500)] <Everett> bosmon: If the words "the brown dog" are editable and a user clicks on the first "n" will the caret be placed in the edit control at the beginning of the text, or before the n?
[11:38:38 EST(-0500)] <Bosmon> Ah
[11:38:48 EST(-0500)] <Bosmon> The position of the caret is part of the options to the InlineEdit instance
[11:38:59 EST(-0500)] <Bosmon> But we do not support moving to the position of the target character
[11:39:06 EST(-0500)] <Bosmon> Only at either the start or the end of the field
[11:39:13 EST(-0500)] <Bosmon> And believe me, even that was quite trouble enough to implement
[11:39:59 EST(-0500)] <Everett> bosmon: Ok, just curious.
[11:40:11 EST(-0500)] <Bosmon> Everett: In terms of the "button to commit", the button really only does appear in the rich text versions as you say
[11:40:51 EST(-0500)] <Bosmon> The "simple field" inline-edits will generally "commit on blur"
[11:40:54 EST(-0500)] <Bosmon> I think
[11:40:54 EST(-0500)] <Everett> bosmon: should focus automatically be placed on the edit field when entering edit mode?
[11:41:01 EST(-0500)] <Bosmon> Yes, it should
[11:41:38 EST(-0500)] <Bosmon> At least, to the best of my knowledge
[11:41:48 EST(-0500)] <anastasiac> Bosmon, quick Q: should the stuff in ...sample-code/pager/sakai-site-setting actually do anything?
[11:42:09 EST(-0500)] <Bosmon> I am but a humble developer, and any opinions or recommendations you would have would be happily taken in
[11:42:13 EST(-0500)] <Bosmon> anastasiac: Yes, that is broken
[11:42:25 EST(-0500)] <Bosmon> And I guess will remain so for the release, I guess we should just "back it out" or hide it somehow
[11:42:30 EST(-0500)] <Bosmon> Perhaps I can take it away into a branch
[11:42:54 EST(-0500)] <anastasiac> ok, good to know. Maybe Justin_o can make the call as to whether or not to move it to a branch
[11:43:03 EST(-0500)] <anastasiac> but I see little use to a sample that doesn't work
[11:43:10 EST(-0500)] <Bosmon>
[11:43:19 EST(-0500)] <Bosmon> Even to assemble that set of broken files was no little effort
[11:43:23 EST(-0500)] <Justin_o> Bosmon, anastasiac: I was just going to ask about that
[11:43:34 EST(-0500)] <Bosmon> But yes, it should clearly "go away"
[11:43:38 EST(-0500)] <Justin_o> does this mean that pager will not be updated in time for the release
[11:43:43 EST(-0500)] <Bosmon> I think there is little chance I can make it work for the release
[11:43:46 EST(-0500)] <Bosmon> Well, the other example does work
[11:43:49 EST(-0500)] <Bosmon> And so do the test cases
[11:43:56 EST(-0500)] <Everett> To the best of anyones knowledge are there any other AT dependent individuals testing fluid?
[11:44:16 EST(-0500)] <Bosmon> So if we possibly can, I would be happy if we could "review" the current state of it, to see if we consider it release-worthy
[11:44:23 EST(-0500)] <Bosmon> It has "at least" the functionality that it did for 0.5
[11:44:27 EST(-0500)] <anastasiac> Everett: "dependent", no, but Justin_o does to AT testing
[11:44:36 EST(-0500)] <Bosmon> Everett: No specific individuals that I am aware of
[11:44:50 EST(-0500)] <Bosmon> But sometimes the components are run by Mike Elledge, who i believe has access to some such users
[11:45:15 EST(-0500)] <anastasiac> yes - I'm not sure if Mike tests using accepted protocols or actual users, though
[11:45:21 EST(-0500)] <Bosmon> !!!
[11:45:25 EST(-0500)] <Bosmon> I was sure he had some actual users
[11:45:29 EST(-0500)] <Bosmon> If not accepted protocols
[11:45:37 EST(-0500)] <anastasiac> Everett, we're delighted to have your feedback and info on your experiences with the components
[11:45:54 EST(-0500)] <Justin_o> Everett: i try to do some AT testing, not nearly as much as I would like. I am actually hoping to see an AT dependent user in action to gain a better understanding of how our strategies of using the AT may differ
[11:46:13 EST(-0500)] <Bosmon> Heads-up to all, who weren't in the channel yesterday
[11:46:42 EST(-0500)] <Everett> Justin: where ar you located?
[11:46:43 EST(-0500)] <Bosmon> Our set of JQuery UI plugins, being selectable, tabindex, and activatable, are being completely blasted, probably as of today, and will be 100% API incompatible
[11:46:47 EST(-0500)] <Bosmon> "boo"
[11:47:04 EST(-0500)] <Justin_o> Everett: I'm in Toronto in the fluid room
[11:47:28 EST(-0500)] <Everett> Ok, so apparently I was using J9 before. So, with J10 here is the trick with the inline edit, which otherwise is pretty good.
[11:47:31 EST(-0500)] <Bosmon> The upside is that this makes way for a beatific new approach whereby the entirety of Fluid becomes available for invocation in JQuery UI plugin style automatically, where appropriate
[11:47:49 EST(-0500)] <Bosmon> And the JQuery UI plugin code itself is being "returned" to appearing in an overtly "that-ist" form, in line with the rest of the framework
[11:47:54 EST(-0500)] <Everett> Justin: perhaps I can play with some things Mon-Tue if you're around.
[11:48:17 EST(-0500)] <Bosmon> Everett - what is J9 and 10?
[11:48:56 EST(-0500)] * Topic is 'fluid.thatistBridge = function (name, peer)' set by Bosmon on 2008-12-09 11:48:56 EST(-0500)
[11:49:06 EST(-0500)] <Everett> Inline edit: 1. activate editable text by pressing enter. 2. Focus is moved to edit control and changes can be made. 3. pressing enter will save changes. 4. there appears to still be an edit control detected by JAWS that is empty and which cannot be navigated away from with auto-forms mode.
[11:49:19 EST(-0500)] <Everett> JAWS9 / 10
[11:49:27 EST(-0500)] <Bosmon> Ah, thanks
[11:50:05 EST(-0500)] <Everett> This happens for all examples that I've tested (examples 1 & 2).
[11:50:27 EST(-0500)] <Bosmon> Ah, that is interesting
[11:50:37 EST(-0500)] <Bosmon> The control has a CSS hidden style
[11:50:50 EST(-0500)] <Bosmon> Which would be very unfortunate if it is not enough to cause it to become invisible to JAWS....
[11:51:06 EST(-0500)] <Everett> Knowing very little about ARIA currently... is there not a way to set the editable texts role to something other than button, perhaps "Editable"?
[11:51:28 EST(-0500)] <Justin_o> Everett: I would definitely like to meet with you those days. I'm going to be doing acceptance testing for our release and your input would be helpful
[11:51:29 EST(-0500)] <Bosmon> The text itself's role is "button"??!
[11:51:46 EST(-0500)] <Everett> Bosmon: the control is not visible to JAWS before I activate the edit, only after I commit the changes by pressing enter inside the edit field.
[11:51:59 EST(-0500)] <Bosmon> Outrageous
[11:52:44 EST(-0500)] <Everett> Bosmon: when I navigate to the editable text it is identified as "button"
[11:52:51 EST(-0500)] <Bosmon> Rotten
[11:52:53 EST(-0500)] <Justin_o> Bosmon, Everett: are you referring to the rich text inline edit or all variants of it
[11:52:59 EST(-0500)] <Bosmon> Throw tomatoes
[11:53:53 EST(-0500)] <Everett> http://build.fluidproject.org/fluid/fluid-components/html/InlineEdit.html example 1 is what I'm testing specifically.
[11:54:17 EST(-0500)] <Justin_o> ah that's interesting...
[11:54:22 EST(-0500)] <Bosmon> Ah
[11:54:37 EST(-0500)] <Bosmon> Let me see what is happening in the DOM there...
[11:54:47 EST(-0500)] <Justin_o> do you have the same experience hererhttp://build.fluidproject.org/fluid/sample-code/inline-edit/announcements/announcements.html
[11:54:53 EST(-0500)] <Justin_o> sorry... i ruined the link
[11:54:55 EST(-0500)] <Justin_o> http://build.fluidproject.org/fluid/sample-code/inline-edit/announcements/announcements.html
[11:55:00 EST(-0500)] <Justin_o> try that one
[11:55:47 EST(-0500)] <Bosmon> Well, all I see in the DOM is the <input> sitting inside the hidden span
[11:55:54 EST(-0500)] <Bosmon> Perhaps JAWS is somehow "caching" the visibility of it
[11:56:01 EST(-0500)] <Bosmon> Having once seen it, it refuses to believe it has gone...
[11:56:22 EST(-0500)] <anastasiac> Bosmon, Everett: regarding the role of button: the ARIA spec has a certain number of roles to choose from
[11:56:33 EST(-0500)] <Bosmon> The state of the DOM is essentially identical to when the edit begins
[11:56:38 EST(-0500)] <anastasiac> the use of button was to signify that the item is actionable
[11:56:44 EST(-0500)] <anastasiac> clearly, though, this doesn't make sense to you
[11:56:57 EST(-0500)] <Everett> So, a better description: 1. The page loads. 2. I navigate to the first example. 3. when JAWS reads the editable text it appends the word "Button" (I assume this is the role). 4. I press enter and am placed in the edit control and can make changes. 5. Pressing enter commits the changes. 6. there is now the modified text and a edit control beneath it that has no content and which I cannot navigate out of using
[11:56:59 EST(-0500)] <anastasiac> so we should at least investigate what other options might make sense
[11:57:22 EST(-0500)] <Bosmon> I wonder how we can make the "phantom edit control" go away
[11:57:38 EST(-0500)] <Bosmon> Perhaps JAWS will be mollified if we belt-and-braces set the hidden style on the input specifically, as well as its container
[11:57:41 EST(-0500)] <Everett> I can get out of forms mode using JAWSKey + semicolon, but if I navigate back to it using up / down arrow it sticks me back in there
[11:57:43 EST(-0500)] <Bosmon> But all the same, I feel it is being rather unfair
[11:58:26 EST(-0500)] <Justin_o> this may have to do with a bug that places a caret on the screen when you push the arrow up or down... i'll see if I can find it
[11:59:21 EST(-0500)] <Justin_o> http://issues.fluidproject.org/browse/FLUID-1641
[11:59:46 EST(-0500)] * ecochran (n=ecochran@dhcp-169-229-212-44.LIPS.Berkeley.EDU) has joined #fluid-work
[11:59:49 EST(-0500)] <Everett> Ok, if I refresh the window (Ins + Escape) the mystery edit control disappears.
[11:59:56 EST(-0500)] <Bosmon> JAWS is a meanie!
[12:00:17 EST(-0500)] <Justin_o> Bosmon: do you think that may have any relation to the errors
[12:00:26 EST(-0500)] <Justin_o> fluid-1641 that is
[12:00:31 EST(-0500)] <Bosmon> No, I don't really
[12:00:43 EST(-0500)] <Everett> So, I think JAWS is caching the control as being there and not realizing that it has disappeared.
[12:00:54 EST(-0500)] <Bosmon> The sheer existence of the phantom control is unconnected with any particular keystrokes
[12:01:09 EST(-0500)] <Bosmon> In that it is seen to exist, before Everett hits any arrow keys
[12:01:14 EST(-0500)] <Bosmon> In an attempt to dispel it
[12:01:35 EST(-0500)] <Bosmon> The only thing I can think of is to set display: none on the input control it self as part of the "hide" operation
[12:01:44 EST(-0500)] <Bosmon> We can try to make a special "souped-up" version for Everett to test
[12:01:50 EST(-0500)] <Bosmon> Although I'm not quite sure how to get it to him
[12:02:13 EST(-0500)] <Everett> Has anyone else tested this widget with JAWS10?
[12:02:57 EST(-0500)] <Justin_o> Bosmon: that's good, i was afraid i'd have to add it to the bug parade
[12:03:11 EST(-0500)] <Everett> I'm loading Ubuntu with Orca 2.2x... to test. I forget which version is loaded. [12:05:55 EST(-0500)] <Everett> Ok, I have to rant again... this virtualbox thing is great! [12:06:01 EST(-0500)] <Justin_o> ecochran: are you around for some quick uploader questions [12:06:16 EST(-0500)] <ecochran> yes [12:06:17 EST(-0500)] <ecochran> shoot [12:06:28 EST(-0500)] <ecochran> Justin_o: ^ [12:07:01 EST(-0500)] <Bosmon> virtualbox? [12:07:04 EST(-0500)] <Justin_o> I just wanted to check in on the state of the uploader (i.e. differentiate what will be in the 0.6 release vs 0.8) [12:07:04 EST(-0500)] <Bosmon> Something like VMWare? [12:07:12 EST(-0500)] <Justin_o> I'm updating my test plans [12:07:23 EST(-0500)] <Justin_o> ecochran: ^ [12:07:59 EST(-0500)] <ecochran> Justin_o: well... things are looking pretty good. How specific do you want me to be? [12:08:05 EST(-0500)] <ecochran> We won't have a dialog example [12:08:20 EST(-0500)] <Justin_o> I'm assuming that the uploader is in release state as is, only needing the bug fixes [12:08:28 EST(-0500)] <Justin_o> no more features to be added [12:08:42 EST(-0500)] <Justin_o> is that what your view is as well [12:09:28 EST(-0500)] <Bosmon> OK [12:09:43 EST(-0500)] <Bosmon> I am basically going through the entire codebase now, updating it for FLUID-1616 [12:09:46 EST(-0500)] <ecochran> I'm going to be making some tweaks to which buttons display at what points in the workflow as part of the work on 1936 (yes, stretching the "BUG" definition a bit [12:09:48 EST(-0500)] <Bosmon> So watch out [12:10:07 EST(-0500)] <ecochran> trying to match up with the workflow that Daphne and Erin are working on [12:10:23 EST(-0500)] <Justin_o> ecochran: yes, this is what I was wondering about [12:10:24 EST(-0500)] <ecochran> I will not be adding the error or feedback stuff that is currently being designed [12:10:42 EST(-0500)] <Bosmon> ecochran: I think you were not here when I gave the headsup that we are unfortunately having to make a 100% change in the API of keyboard-a11y [12:10:48 EST(-0500)] <Bosmon> Every user of it will need to be updated for the release.... [12:10:52 EST(-0500)] <Bosmon> "glarg" [12:11:13 EST(-0500)] <ecochran> Bosmon: grumph! [12:11:19 EST(-0500)] <Bosmon> Exactly [12:11:29 EST(-0500)] <ecochran> got docs? [12:11:33 EST(-0500)] <Bosmon> None! [12:11:53 EST(-0500)] <ecochran> if I read the transcript will I "get" it? [12:11:57 EST(-0500)] <Everett> bosmon: like vmware, but works better, at least for responsiveness of Orca. [12:12:05 EST(-0500)] <Bosmon> Not sure [12:12:18 EST(-0500)] <Bosmon> The bottom line is, we are moving in line with the latest recommendations for JQuery UI plugins [12:12:29 EST(-0500)] <Bosmon> As well as buying permanent insurance against people "nicking" our namespace again [12:12:38 EST(-0500)] <ecochran> Bosmon: can you or someone else write up a short description of what needs to change [12:12:43 EST(-0500)] <Bosmon> Yes [12:12:51 EST(-0500)] <Bosmon> We should have full docs ASAP [12:12:56 EST(-0500)] <Bosmon> Although I'm not sure which has higher priority.... [12:12:56 EST(-0500)] <jessm> ecochran: JQuery started using selectable [12:12:58 EST(-0500)] <Justin_o> ecochran: are the buttons as simple as a name change or are they actually different buttons [12:12:59 EST(-0500)] <ecochran> oh, I caught a little of this yesterday... funny conversation [12:13:02 EST(-0500)] <Bosmon> http://docs.jquery.com/UI/Guidelines [12:13:08 EST(-0500)] <Bosmon> These are the "Guidelines" [12:13:18 EST(-0500)] <jessm> Bosmon: which are your two that are competing for priority? [12:13:25 EST(-0500)] <Bosmon> I have hit upon a "cunning approach" whereby we can make our entire codebase satisfy them automatically [12:13:50 EST(-0500)] <Bosmon> jessm: Getting all of the codebase updated and the test cases passing again, and writing documentation explaining what the new API is [12:13:57 EST(-0500)] <ecochran> Justin_o: not completely sure yet. Are you getting crazy enough that you'd prefer if I backed off any big changes? Not that any of these changes are very big [12:14:16 EST(-0500)] <ecochran> Justin_o: I could keep it as simple as just getting the damn Upload button to work again after pause [12:14:26 EST(-0500)] <jessm> Bosmon: i think you've got the order right in that message ^ [12:15:08 EST(-0500)] <ecochran> Bosmon: if you can give me a good example, I'm happy to update the Uploader code and examples and the Inline Edit Section Info example [12:15:23 EST(-0500)] <Justin_o> ecochran: that's what i'm thinking for now... if we have time later we may be able to add the others. I want to make sure that the current bugs are fixed and that I'll have time to properly test the features [12:16:11 EST(-0500)] <ecochran> Justin_o: I'm working the priorities in Bug Parade first and foremost [12:16:27 EST(-0500)] <ecochran> I'll fix that Upload button and then call 1936 done [12:16:39 EST(-0500)] <Bosmon> ecochran: [12:16:41 EST(-0500)] <Bosmon> http://pastebin.com/me4be459 [12:16:51 EST(-0500)] <Bosmon> Here is the "current" version of keyboard-a116-tests.js [12:17:03 EST(-0500)] <Bosmon> You can see the new invocation style of most of the methods there [12:17:27 EST(-0500)] <Bosmon> We essentially have boiled everything down to one, gigantic, plugin, named "fluid" [12:17:37 EST(-0500)] <Bosmon> Which in effect exposes our entire framework [12:17:58 EST(-0500)] <Bosmon> The first argument is a string, the name of the thing you want to invoke - which must accept a JQueryable as the first argument [12:18:15 EST(-0500)] <ecochran> OK, I think that I have it [12:18:32 EST(-0500)] <Bosmon> The second argument is either the first target argument, or a list of arguments if it takes several arguments [12:18:52 EST(-0500)] <Bosmon> Or nothing [12:19:18 EST(-0500)] <ecochran> sorry, "first target argument"? [12:19:22 EST(-0500)] <Bosmon> Another important change is that all the things which used to be accessible under $.a11y are now under fluid.a11y [12:19:30 EST(-0500)] <Bosmon> All of those constants for keycodes and directions, etc. [12:19:39 EST(-0500)] <Bosmon> To ensure noone ever nicks stuff off us again [12:19:56 EST(-0500)]