Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 57 Next »

[09:56:18 CDT(-0500)] <span style="color: black"> <Justin_o> yura1: for setting priorities in IoC, can i have a negative priority</span>

[09:56:20 CDT(-0500)] <span style="color: black"> <Justin_o> ?</span>

[09:57:08 CDT(-0500)] <span style="color: black"> <yura1> i think so</span>

[09:57:15 CDT(-0500)] <span style="color: black"> <Justin_o> thanks.. can i mix that with "last"</span>

[09:57:25 CDT(-0500)] <span style="color: black"> <yura1> i would step into code though to see what takes precedence</span>

[10:06:22 CDT(-0500)] <span style="color: black"> <Justin_o> yura1: thanjks</span>

[10:06:24 CDT(-0500)] <span style="color: black"> <Justin_o> thanks</span>

[10:10:00 CDT(-0500)] <span style="color: black"> <Justin_o> yura1: i didn't step through the code, but a quick test looks like last wins</span>

[10:10:07 CDT(-0500)] <span style="color: black"> <yura1> cools</span>

[11:49:05 CDT(-0500)] <span style="color: black"> <avtar> hey kasper</span>

[11:49:44 CDT(-0500)] <span style="color: black"> <avtar> i've made the required changes to the gpii wiki</span>

[12:58:15 CDT(-0500)] <span style="color: black"> <kasper> avtar - you're a hero!!!</span>

[12:58:23 CDT(-0500)] <span style="color: black"> <kasper> thanks a lot</span>

[13:04:22 CDT(-0500)] <span style="color: black"> <colinclark> michelled: Here's a fairly concrete example of schema.org's approach to micro data: <a href="http://schema.org/Person" class="external-link" rel="nofollow">http://schema.org/Person</a></span>

[13:05:11 CDT(-0500)] <span style="color: black"> <colinclark> Looks to me like the general approach is based on a proposed W3C Microdata spec: <a href="http://www.w3.org/TR/html5/microdata.html" class="external-link" rel="nofollow">http://www.w3.org/TR/html5/microdata.html</a></span>

[13:06:56 CDT(-0500)] <span style="color: green">* clown toots own horn: <a href="http://library.gnome.org/misc/release-notes/3.4/#rna11y.zoom-settings" class="external-link" rel="nofollow">http://library.gnome.org/misc/release-notes/3.4/#rna11y.zoom-settings</a></span>

[13:12:29 CDT(-0500)] <span style="color: black"> <colinclark> clown: Yay!</span>

[13:12:36 CDT(-0500)] <span style="color: black"> <colinclark> I enjoyed seeing the dialog in Fedora 17 alpha</span>

[13:12:41 CDT(-0500)] <span style="color: black"> <colinclark> very nice</span>

[13:15:03 CDT(-0500)] <span style="color: black"> <michelled> thx colinclark</span>

[13:40:52 CDT(-0500)] <span style="color: black"> <avtar> kasper: np</span>

[13:52:02 CDT(-0500)] <span style="color: black"> <Justin_o> fluid-everyone: the community meetings is on now.. yura is talking about node.js</span>

[15:09:20 CDT(-0500)] <span style="color: black"> <travis_84> New to an irc but saw this project at GSoC and wanted to introduce myself</span>

[15:12:43 CDT(-0500)] <span style="color: black"> <Bosmon> Hi travis_84, welcome</span>

[15:13:07 CDT(-0500)] <span style="color: black"> <Bosmon> Which project(s) are you interested in?</span>

[15:14:53 CDT(-0500)] <span style="color: black"> <travis_84> Hi. Well the whole program itself is awesome, it hits close to home more and more</span>

[15:15:24 CDT(-0500)] <span style="color: black"> <Bosmon> Glad you like it <ac:emoticon ac:name="smile" /></span>

[15:19:23 CDT(-0500)] <span style="color: black"> <travis_84> but there are quite a few projects I want to help on (GSoC accepted or not). The one within my major is the ATutor chat redesign. But I am going big and studying up on something I tried with voice recognition; "Towards non-visual/accessible programming"</span>

[15:21:28 CDT(-0500)] <span style="color: black"> <Bosmon> travis_84 - that's great</span>

[15:21:48 CDT(-0500)] <span style="color: black"> <Bosmon> I was disappointed there had been so little interest in that project so far, which I am sure is one of the most interesting</span>

[15:23:59 CDT(-0500)] <span style="color: black"> <travis_84> it really is</span>

[15:24:47 CDT(-0500)] <span style="color: black"> <Bosmon> There's little enough work being done even in alternative visual methods of programming</span>

[15:24:59 CDT(-0500)] <span style="color: black"> <Bosmon> Most people seemed to get discouraged after the 80s</span>

[15:25:43 CDT(-0500)] <span style="color: black"> <travis_84> especially with touch gestures becoming mainstream, gesture programming with possible voice support sounds possible</span>

[15:25:49 CDT(-0500)] <span style="color: black"> <Bosmon> You may be interested in this PhD thesis that was published in 92... <a href="http://digitalassets.lib.berkeley.edu/techreports/ucb/text/CSD-93-726.pdf" class="external-link" rel="nofollow">http://digitalassets.lib.berkeley.edu/techreports/ucb/text/CSD-93-726.pdf</a></span>

[15:25:56 CDT(-0500)] <span style="color: black"> <Bosmon> This guy is now at Oracle, but he is generally discouraged : P</span>

[15:26:25 CDT(-0500)] <span style="color: black"> <Bosmon> There is a little work at Microsoft in using the tablet form factor for some kinds of coding</span>

[15:26:50 CDT(-0500)] <span style="color: black"> <Bosmon> But it is mostly pretty simplistic stuff which produces lists of rules for actors in games to follow</span>

[15:30:08 CDT(-0500)] <span style="color: black"> <travis_84> hmm, I will add that to my reading lol. I think the first step would be to create a visual language to represent the programming functions and elements?</span>

[15:30:46 CDT(-0500)] <span style="color: black"> <Bosmon> travis_84 - it may well be easier to start with some kind of "visual intermediate", yes</span>

[15:30:52 CDT(-0500)] <span style="color: black"> <Bosmon> Especially if you are a sighted user yourself : P</span>

[15:32:16 CDT(-0500)] <span style="color: black"> <travis_84> True, at least once we have a visual context we have a place to start for the hearing only too</span>

[15:32:29 CDT(-0500)] <span style="color: black"> <Bosmon> Some other people who are interested in the project are Clayton Lewis, a professor here at CU, and T.V. Raman, a blind developer who is a research scientist at Google</span>

[15:33:21 CDT(-0500)] <span style="color: black"> <Bosmon> In talking to Raman, Clayton noticed an interesting idea he had, in treating elements of a non-visual interface as proxies for "queries" which you could say a sighted user runs automatically</span>

[15:33:48 CDT(-0500)] <span style="color: black"> <Bosmon> So he saw that Raman had made a system where he could play games like "Bejeweled" by use of these kinds of "queries"</span>

[15:34:11 CDT(-0500)] <span style="color: black"> <Bosmon> But an important part of the system is that the user could manage and edit this set of queries to suit their own use of the system for a particular task</span>

[15:34:24 CDT(-0500)] <span style="color: black"> <travis_84> wow, Raman sounds like a good person to talk to. thanks</span>

[15:34:41 CDT(-0500)] <span style="color: black"> <Bosmon> Raman is very cool - although very busy : P</span>

[15:35:10 CDT(-0500)] <span style="color: black"> <Bosmon> But he's probably the person in the world who knows most about this kind of system in practice, since he has to use them every day</span>

[15:36:13 CDT(-0500)] <span style="color: black"> <Bosmon> And after this, you could perhaps bind them to hot-keys so you could play the game at a reasonable kind of rate after you got used to it.... "Find the first jewel in the column to the right of the cursor scanning downwards that is blue", for example, could be one of these kinds of queries</span>

[15:36:26 CDT(-0500)] <span style="color: black"> <Bosmon> But you might not be able to find out what kinds of queries you wanted until you had played the game for a bit</span>

[15:37:14 CDT(-0500)] <span style="color: black"> <Bosmon> So, this kind of thing could eventually be the basis of a system for producing "code"... but so far, any work you do on this will be a world first <ac:emoticon ac:name="smile" /></span>

[15:38:48 CDT(-0500)] <span style="color: black"> <travis_84> all this said, i do feel extremely under qualified... I am going for it anyway, I have been self taught and learn fast. This is definitely something I want to be apart of either way</span>

[15:40:09 CDT(-0500)] <span style="color: black"> <Bosmon> travis_84 - you are already a world expert on the subject <ac:emoticon ac:name="smile" /></span>

[15:40:28 CDT(-0500)] <span style="color: black"> <Bosmon> It will be great to have you involved</span>

[15:40:48 CDT(-0500)] <span style="color: black"> <Bosmon> So, this project of Raman's using "last generation technologies" might have some ideas which can be plundered... <a href="http://emacspeak.sourceforge.net/emacspeak.html" class="external-link" rel="nofollow">http://emacspeak.sourceforge.net/emacspeak.html</a></span>

[15:42:10 CDT(-0500)] <span style="color: black"> <Bosmon> But of course emacs is based completely on the idea of streams of text characters as the ultimate abstraction for code... which ultimately isn't going to be good enough</span>

[15:42:46 CDT(-0500)] <span style="color: black"> <Bosmon> Still, it is in some ways the "Rolls Royce" of editors, and some people feel that it's even a possibly sane approach to try porting it to the web : P</span>

[15:42:49 CDT(-0500)] <span style="color: black"> <Bosmon> Such as sgithens here....</span>

[15:42:51 CDT(-0500)] <span style="color: black"> <travis_84> true</span>

[15:43:08 CDT(-0500)] <span style="color: black"> <Bosmon> Still, the world of even traditional web-based IDEs is extremely wild and woolly still</span>

[15:43:33 CDT(-0500)] <span style="color: black"> <Bosmon> There are several solutions emerging at once, but they are all pretty shaky even at best</span>

[15:43:44 CDT(-0500)] <span style="color: black"> <Bosmon> There is Bespin, which became "ACE"....</span>

[15:43:47 CDT(-0500)] <span style="color: black"> <Bosmon> There is "CodeMirror"</span>

[15:43:55 CDT(-0500)] <span style="color: black"> <Bosmon> And there is IBM/Eclipse's "Orion"</span>

[15:44:24 CDT(-0500)] <span style="color: black"> <Bosmon> All of them are text-based, and all of them have serious issues even just as text-based IDEs, let alone environments that would enable more interesting kinds of structured editing</span>

[15:45:14 CDT(-0500)] <span style="color: black"> <Bosmon> I worked on a project with CodeMirror recently, and discovered it doesn't even have API support for helping the user to write syntactically correct programs!!</span>

[15:47:05 CDT(-0500)] <span style="color: black"> <travis_84> IDE's like Netbean, and Dr.java from Princeton have been quite frustrating as well since it ignores keyboard focus for windows magnifier for those not so blind</span>

[15:47:42 CDT(-0500)] <span style="color: black"> <Bosmon> Almost all of these "baked-in" projects have no concern for accessibility at all</span>

[15:47:54 CDT(-0500)] <span style="color: black"> <Bosmon> So the web represents a really huge opportunity to improve the situation enormously</span>

[15:48:23 CDT(-0500)] <span style="color: black"> <Bosmon> Especially if we can create a kind of "open architecture" where end users can contribute and share their own accessibility tweaks and "queries"</span>

[15:50:30 CDT(-0500)] <span style="color: black"> <Bosmon> Here's a sample of Raman's writing style : P</span>

[15:50:33 CDT(-0500)] <span style="color: black"> <Bosmon> "On this theme, when once challenged by a proponent of a crash-prone but well-marketed mousetrap with the assertion "Emacs is a system from the 70's", the creator of Emacspeak evinced surprise at the unusual candor manifest in the assertion that it would take popular idiot-proven interfaces until the year 2070 to catch up to where the Emacspeak audio desktop is today."</span>

[15:54:03 CDT(-0500)] <span style="color: black"> <travis_84> hah</span>

[15:54:08 CDT(-0500)] <span style="color: black"> <travis_84> nice</span>

[15:55:35 CDT(-0500)] <span style="color: black"> <travis_84> the whole thing could be open like how games allow their games to be modded to fit the users preferences. code representations could be open to the user and how they feel it should be represented</span>

[15:55:40 CDT(-0500)] <span style="color: black"> <Bosmon> His "auditory icons" idea is pretty interesting</span>

[15:55:53 CDT(-0500)] <span style="color: black"> <Bosmon> Yes, I think that's a really great analogy</span>

[15:56:04 CDT(-0500)] <span style="color: black"> <Bosmon> If you've played WoW, for example, you've seen that this "aftermarket" is completely huge</span>

[15:56:17 CDT(-0500)] <span style="color: black"> <Bosmon> There's a giant bank of keystrokes, icons, and scripts that every expert player comes to rely on</span>

[15:56:55 CDT(-0500)] <span style="color: black"> <travis_84> I have and I did that</span>

[15:57:06 CDT(-0500)] <span style="color: black"> <Bosmon> To the extent that if you HAVEN'T customised the game in this way, you can't actually expect to become a top player</span>

[15:57:09 CDT(-0500)] <span style="color: black"> <Bosmon> ha, cool</span>

[15:57:18 CDT(-0500)] <span style="color: black"> <Bosmon> Thankfully I gave up once I got to level 60 <ac:emoticon ac:name="smile" /></span>

[15:57:19 CDT(-0500)] <span style="color: black"> <travis_84> made macros to make keystrokes into single icons</span>

[15:57:40 CDT(-0500)] <span style="color: black"> <Bosmon> But I realised that I was going nowhere in the game unless I was prepared to devote really serious amounts of time to customising it</span>

[15:58:08 CDT(-0500)] <span style="color: black"> <travis_84> ya same</span>

[15:58:34 CDT(-0500)] <span style="color: black"> <Bosmon> But actually, now I'm playing "Skyrim" I am really missing the fact that I used to be able to have quick access to stuff</span>

[15:58:40 CDT(-0500)] <span style="color: black"> <Bosmon> Although this was essential in a game that wasn't "stoptime"</span>

[15:58:57 CDT(-0500)] <span style="color: black"> <travis_84> and well I think i could contribute to a real world 'guild" right?</span>

[15:59:07 CDT(-0500)] <span style="color: black"> <Bosmon> But still, Skyrim is clearly designed for the console market, where there ARE no key bindings</span>

[15:59:37 CDT(-0500)] <span style="color: black"> <Bosmon> Which is showing the kinds of ways that the move towards consoles and tablets is at least temporarily making things hugely worse for unsighted users</span>

[15:59:48 CDT(-0500)] <span style="color: black"> <Bosmon> I think you really could : P</span>

[16:00:24 CDT(-0500)] <span style="color: black"> <Bosmon> Another thing Raman is interested in is ways of making touchscreens "audibly responsive"</span>

[16:00:44 CDT(-0500)] <span style="color: black"> <Bosmon> So that as you brush your finger over them, you get little "bumps and ticks" indicating when you have moved into an area which is active in a certain way</span>

[16:01:06 CDT(-0500)] <span style="color: black"> <travis_84> agreed, with kinect for pc. coding is not limited to that either</span>

[16:01:28 CDT(-0500)] <span style="color: black"> <travis_84> think of a motion based coder</span>

[16:01:39 CDT(-0500)] <span style="color: black"> <Bosmon> the possibilities, of Tiberium, are limitless! : P</span>

[16:02:10 CDT(-0500)] <span style="color: black"> <Bosmon> But so far, all things being equal, losing access to keyboards has been a huge loss for most unsighted users</span>

[16:02:46 CDT(-0500)] <span style="color: black"> <Bosmon> And that input modality is a lifeline to all kinds of other users as well.... those that depend on non-analog "switches and sensors" of various kinds, like puffer switches</span>

[16:04:00 CDT(-0500)] <span style="color: black"> <travis_84> eye tracking based communicators</span>

[16:04:33 CDT(-0500)] <span style="color: black"> <Bosmon> Yes, could be great</span>

[16:04:41 CDT(-0500)] <span style="color: black"> <Bosmon> Can Kinect do eye tracking?</span>

[16:04:49 CDT(-0500)] <span style="color: black"> <travis_84> sorry excuse me for a sec brb</span>

[16:05:31 CDT(-0500)] <span style="color: black"> <Bosmon> Looks like it can, a bit : P</span>

[16:05:32 CDT(-0500)] <span style="color: black"> <Bosmon> <a href="http://www.kinecthacks.com/eye-detection-feature-of-kinect/" class="external-link" rel="nofollow">http://www.kinecthacks.com/eye-detection-feature-of-kinect/</a></span>

[16:20:09 CDT(-0500)] <span style="color: black"> <travis_84> back</span>

[16:22:43 CDT(-0500)] <span style="color: black"> <travis_84> So Bosmon I guess a question I have is what is the expectation of the project?</span>

[16:38:08 CDT(-0500)] <span style="color: black"> <Bosmon> Hi travis_84</span>

[16:38:28 CDT(-0500)] <span style="color: black"> <Bosmon> Well, it's just a GSoC project (for now!)</span>

[16:38:38 CDT(-0500)] <span style="color: black"> <Bosmon> The expectation is that something reasonable will happen : P</span>

[16:38:48 CDT(-0500)] <span style="color: black"> <Bosmon> Representing any progress at all on the underlying aims of the project...</span>

[16:39:00 CDT(-0500)] <span style="color: black"> <Bosmon> And hopefully something which might represent a useful basis for future work</span>

[16:39:37 CDT(-0500)] <span style="color: black"> <Bosmon> For example, one of the successful GSoC projects of last year, the videoPlayer, had its codebase adopted as the basis of the core Fluid videoPlayer component which we are working on today</span>

[16:40:16 CDT(-0500)] <span style="color: black"> <Bosmon> Our aim in Fluid is always to build communities, rather than to necessarily achieve particular technical outcomes</span>

[16:40:45 CDT(-0500)] <span style="color: black"> <Bosmon> Although achieving technical outcomes is always helpful : p</span>

[16:41:10 CDT(-0500)] <span style="color: black"> <travis_84> true</span>

[16:41:59 CDT(-0500)] <span style="color: black"> <Bosmon> Given there is essentially zero work in this field so far, almost anything you achieved would represent a significant outcome <ac:emoticon ac:name="smile" /></span>

[16:42:50 CDT(-0500)] <span style="color: black"> <Bosmon> An important area to look at, that we didn't explicitly mention so far, is the world of "dataflow programming" represented by environments like Apple's Quartz Composer, or "Pure Data"</span>

[16:43:18 CDT(-0500)] <span style="color: black"> <Bosmon> Although these are incredibly immature, they represent the best efforts to date to move programming into a "structural" rather than a "textual" idiom</span>

[16:43:57 CDT(-0500)] <span style="color: black"> <Bosmon> But a huge amount of this is up to you</span>

[16:44:28 CDT(-0500)] <span style="color: black"> <Bosmon> You made decided that moving too far from textual programming would be too much of a leap right now, and you might want to look initially at more web-based equivalents of Raman's "emacspeak"</span>

[16:44:33 CDT(-0500)] <span style="color: black"> <Bosmon> You may decide</span>

[16:45:18 CDT(-0500)] <span style="color: black"> <Bosmon> Or you may like to concentrate on the part of the system that manages the "social inventory", the "marketplace" or what you want to call it for "accessibility queries and solutions"</span>

[16:46:25 CDT(-0500)] <span style="color: black"> <Bosmon> Here we are with the vague floor plan of what looks like a cathedral of about 120 years work : P</span>

[16:46:38 CDT(-0500)] <span style="color: black"> <Bosmon> We can start laying the first stones where we like....</span>

[16:47:13 CDT(-0500)] <span style="color: black"> <Bosmon> Or, more usefully, where we first expect to be able to encourage engagement from others by producing partial solutions which let people get at least some early value</span>

[16:47:35 CDT(-0500)] <span style="color: black"> <travis_84> I mean I was thinking say I am able to make a basic "web app" that uses javascript to make or code an html web-page using gesture... is that a significant addition? it is a jump off point at least</span>

[16:47:49 CDT(-0500)] <span style="color: black"> <Bosmon> travis_84 - sounds like a great starting point</span>

[16:48:11 CDT(-0500)] <span style="color: black"> <Bosmon> Even authoring markup is a significantly unsolved problem, even for sighted users with the best technology we have <ac:emoticon ac:name="smile" /></span>

[16:53:10 CDT(-0500)] <span style="color: black"> <travis_84> alright I have a lot of reading to do and need to try and figure out what I am going to say in the proposal, thanks for your insights and for the added motivation and help</span>

[16:53:40 CDT(-0500)] <span style="color: black"> <Bosmon> it's great - looking forward to reading it!</span>

[16:53:47 CDT(-0500)] <span style="color: black"> <Bosmon> Do keep in touch, here or on the mailing list</span>

[16:54:23 CDT(-0500)] <span style="color: black"> <NickMayne> Hey Bosmon</span>

[16:54:33 CDT(-0500)] <span style="color: black"> <NickMayne> Just giving your latest checkin's a test <ac:emoticon ac:name="smile" /></span>

[16:54:41 CDT(-0500)] <span style="color: black"> <NickMayne> Thanks for fixing so quickly btw.</span>

[16:54:41 CDT(-0500)] <span style="color: black"> <Bosmon> The wikipedia page is already a good jumping off point for resources: <a href="http://en.wikipedia.org/wiki/Dataflow_programming" class="external-link" rel="nofollow">http://en.wikipedia.org/wiki/Dataflow_programming</a></span>

[16:54:44 CDT(-0500)] <span style="color: black"> <Bosmon> Hey there NickMayne</span>

[16:54:59 CDT(-0500)] <span style="color: black"> <Bosmon> Hope it works out for you</span>

[16:55:11 CDT(-0500)] <span style="color: black"> <NickMayne> Will let you know in about 1 minute</span>

[16:55:13 CDT(-0500)] <span style="color: black"> <Bosmon> I found there were a number of timing issues in the 2-editor case, which are hopefully resolved now</span>

[16:55:34 CDT(-0500)] <span style="color: black"> <NickMayne> I gave a demo of the Fluid Inline editing stuff.</span>

[16:55:42 CDT(-0500)] <span style="color: black"> <Bosmon> That's awesome</span>

[16:55:43 CDT(-0500)] <span style="color: black"> <NickMayne> At the Orcahrd CMS Steering committee</span>

[16:55:53 CDT(-0500)] <span style="color: black"> <Bosmon> Btw, I think it is pretty hilarious that you are in Richmond <ac:emoticon ac:name="smile" /></span>

[16:55:57 CDT(-0500)] <span style="color: black"> <Bosmon> I grew up about 5 miles away from there</span>

  • No labels