| 00:46 | <MikeSmith> | Hixie: thanks for the changes |
| 01:40 | <TabAtkins> | Ah, Whataburger, the most Texan of all fast-food joints. |
| 01:40 | TabAtkins | is filling up on stereotypical food. |
| 06:20 | <JonathanNeal> | Hey all |
| 06:24 | <Hixie> | hey |
| 06:46 | <JonathanNeal> | Hey Hixie, how's it goin' tonight? |
| 06:53 | <MikeSmith> | Hixie: if you can indulge me, there's another minor issue in the markup for the HTML5 spec that I'd hope you can fix upstream -- it's that there's no "type" attribute on the style element that contains the .domintro:before selector |
| 06:53 | <Hixie> | JonathanNeal: good, good, you? |
| 06:54 | <Hixie> | MikeSmith: updating... |
| 06:54 | <othermaciej> | MikeSmith: are you fixing the pubrules checker issues that the webmaster pointed out? |
| 06:54 | <JonathanNeal> | The battle of HTML5 seems all but over within the company, we're good to go. I've been building out the next version of the theme that ships with the product. |
| 06:55 | <othermaciej> | MikeSmith: (sounds like you are, just wanted to confirm) |
| 06:55 | <othermaciej> | JonathanNeal: what is your company's product and/or service? |
| 06:56 | <JonathanNeal> | Liferay, our company site is all HTML5'd out now, but now I've been working to put it in the product. |
| 06:57 | <JonathanNeal> | Along with moving off TABLE-based layouts, and so far it all looks good. Today I built the new default theme that will ship with the next version of the software coming out later this month or sometime about then. |
| 06:57 | <MikeSmith> | othermaciej: yeah, plh fixed a bunch and I'm fixing more now |
| 06:58 | <MikeSmith> | othermaciej: scare quotes implied |
| 06:58 | <JonathanNeal> | I always forget who each of you guys work for, it seems to be a pretty well mixed bunch. |
| 06:59 | <MikeSmith> | othermaciej: in some cases we're not really fixing anything, but instead just trying to get the pubrules checker to stfu |
| 07:00 | <Hixie> | MikeSmith: fixed |
| 07:01 | <othermaciej> | MikeSmith: that's unfortunate, glad things are getting fixed though |
| 07:01 | <MikeSmith> | Hixie: thanks again |
| 07:03 | <MikeSmith> | othermaciej: a number of the drafts are already in place at the 2010/WD-* URLs, so we just need to wait for the webmaster to add symlinks for those |
| 07:04 | <MikeSmith> | the rest I should have in place well before the html wg telcon tonight |
| 07:04 | <othermaciej> | MikeSmith: hawt |
| 07:06 | <JonathanNeal> | othermaciej, what do you do? |
| 07:07 | <othermaciej> | JonathanNeal: I'm on the WebKit team at Apple |
| 07:08 | <JonathanNeal> | Oh neato, I like you guys *except that whole gradient spec* :D |
| 07:08 | <othermaciej> | I blame TabAtkins for the gradient spec |
| 07:09 | <JonathanNeal> | Really? TabAtkins did it? Say it ain't so. He helped move me to HTML5. |
| 07:09 | <JonathanNeal> | othermaciej, fine, can I blame you for your quicktime-ish HTML5 video player? |
| 07:10 | <othermaciej> | JonathanNeal: I think he's currently editing the gradient spec, though I think we were the first to have some form of gradients in CSS |
| 07:11 | <knowtheory> | othermaciej: i saw that this issue http://www.w3.org/html/wg/tracker/issues/10 was closed w/o prejudice |
| 07:11 | <knowtheory> | and i was just curious whether that means no consideration was taken regarding SMIL? |
| 07:11 | <knowtheory> | (i'm just trying to figure out what the status is atm, besides, you know, closed) |
| 07:11 | <othermaciej> | JonathanNeal: what aspect of our video player do you find too "quicktime-ish"? |
| 07:12 | <JonathanNeal> | http://sandbox.thewikies.com/html5-video-framing/ othermaciej |
| 07:12 | <othermaciej> | knowtheory: no one made a concrete proposal to make the <video> element more like SMIL |
| 07:13 | <othermaciej> | knowtheory: if someone were to come forward with a proposal, we could reopen it, though at this point I'd suggest any proposals along those lines start with a bug report |
| 07:13 | <knowtheory> | othermaciej: great, that's what i wanted to know. |
| 07:13 | <knowtheory> | i'm disappointed that Dick Bulterman never got back to you guys |
| 07:14 | <othermaciej> | JonathanNeal: that's not what the latest controls look like |
| 07:14 | <knowtheory> | i work w/ a startup who's doing SMIL based video delivery, and when i found this, i thought it was both interesting, and potentially useful |
| 07:14 | <knowtheory> | especially if there is interoperability between SMIL and (X)HTML5 |
| 07:14 | <JonathanNeal> | I'll either update or delete the page when the next Safari is out then. |
| 07:14 | <othermaciej> | knowtheory: for what it's worth, I think a proposal to incompatibly change the core features of <video> and <audio> would have a huge uphill battle at this point |
| 07:14 | <JonathanNeal> | othermaciej, will there be keyboard controls, do you know? |
| 07:15 | <othermaciej> | JonathanNeal: in Safari 4.04 on SnowLeopard, the controls I get on the live video do not look anything like your screenshot |
| 07:15 | <JonathanNeal> | You don't need to answer that, I'll find out when I find out. |
| 07:15 | <knowtheory> | othermaciej: yeah i appreciate that (which is why i'm particularly disappointed that Bulterman never got back to you guys), i figure i should at least take a look |
| 07:15 | <JonathanNeal> | othermaciej, oh I guess it's a PC thing. |
| 07:16 | <knowtheory> | we're interested in adoption of SMIL, but if that can be accomplished by better coordiation w/ existing standards, and being able to do things like integrate SMIL timing semantics via XML modules, so much the better IMO |
| 07:16 | <othermaciej> | JonathanNeal: we might not have updated Safari on Windows to the new controls yet |
| 07:17 | <JonathanNeal> | You might be interested in seeing http://sandbox.thewikies.com/html5-layout/7cogs-website.html it's filled with all sorts of HTML5 / CSS3 -webkit goodness. |
| 07:17 | <JonathanNeal> | -moz goodness too, even some IE love. |
| 07:18 | <othermaciej> | JonathanNeal: the controls are not keyboardable by default if that is your concern |
| 07:18 | <othermaciej> | but the look and hover behavior are different |
| 07:18 | <knowtheory> | othermaciej: i apologize for being out of the loop on this, i've only really started paying attention to the HTML5 process in earnest recently, i wish i had tuned in sooner :\ |
| 07:19 | <knowtheory> | (apologize for raising this issue at this late stage that is) |
| 07:20 | <othermaciej> | knowtheory: I am just giving you fair warning that you may not have much luck if you propose incompatible changes |
| 07:20 | <othermaciej> | unless you have a really really really compelling rationale |
| 07:20 | <othermaciej> | JonathanNeal: nice looking web page |
| 07:23 | <knowtheory> | othermaciej: Thanks, i appreciate that, and don't really expect anything, but again, squandering the opportunity seems just as foolish, if something amenable could be worked out. |
| 07:23 | <knowtheory> | othermaciej: btw, i see that last call for comments is out, when is that process going to draw to a close? |
| 07:24 | <othermaciej> | knowtheory: the WHATWG copy of the spec went into Last Call, but then converted to unversioned, so I am not sure what the plan is there |
| 07:24 | <othermaciej> | at the W3C, the spec is not yet in Last Call |
| 07:24 | <Hixie> | the stability is just marked section-by-section |
| 07:25 | <Hixie> | going to last call was really a shorthand for saying all the sections were in last call |
| 07:25 | <knowtheory> | Ah okay, thanks. |
| 07:25 | <Hixie> | (you can see the states in the margins of the whatwg spec) |
| 07:25 | <knowtheory> | I saw the Status there under http://www.w3.org/TR/html5/video.html#video |
| 07:26 | <knowtheory> | so i wasn't sure, exactly what the implications of Last Call were here |
| 07:26 | <hsivonen> | "Larry wrote 2 March 2010 member-only email with suggested path forward (for some reason, it's not in the tag⊙wo archives, hence not linkable from this agenda) |
| 07:26 | <hsivonen> | " |
| 07:26 | <Hixie> | knowtheory: oh you don't want to look at the version on the TR/ page, that's almost always out of date |
| 07:26 | <hsivonen> | does anyone know if the email in question is available elsewhere? |
| 07:26 | <knowtheory> | (again i apologize that i haven't had more time to poke around and find out what the W3C process is yet, i saw that the relevant ticket was closed 2 months ago, and wanted to pop in to find out if it was too late to do anything) |
| 07:27 | <hsivonen> | context: http://lists.w3.org/Archives/Public/www-tag/2010Mar/0011.html |
| 07:27 | <knowtheory> | Hixie: where should i be looking then? |
| 07:27 | <Hixie> | knowtheory: you want either http://whatwg.org/html or http://dev.w3.org/html5/spec/ |
| 07:27 | <JonathanNeal> | othermaciej, thanks |
| 07:27 | <knowtheory> | ah fantastic thanks Hixie |
| 07:27 | <Hixie> | the former has a feedback form in the lower right which might be useful |
| 07:28 | <knowtheory> | Cool, much appreciated |
| 07:34 | <othermaciej> | hsivonen: interesting - haven't seen the email, but then I looked at the February tag archives and found much fascinating content |
| 07:44 | <hsivonen> | othermaciej: http://lists.w3.org/Archives/Member/tag/2010Feb/0065.html and http://lists.w3.org/Archives/Member/tag/2010Feb/0057.html in particular |
| 07:53 | <othermaciej> | hsivonen: I'm wary of linking specific emails lest there be complaints that discussion on secret lists isn't being kept secret enough |
| 07:55 | <hsivonen> | sorry. I thought in was common practice to point to the URLs. |
| 07:56 | <knowtheory> | Incidentally othermaciej & Hixie is the HTML5 group aware of XHTML+SMIL? http://www.w3.org/TR/XHTMLplusSMIL/ |
| 07:57 | <othermaciej> | hsivonen: me too but apparently it gets some people's panties in a bunch |
| 07:57 | <othermaciej> | hsivonen: you shouldn't feel the need to stop doing so |
| 07:57 | <othermaciej> | I'm just explaining why I am not doing so at the moment |
| 07:57 | <hsivonen> | I see |
| 07:58 | <foolip> | knowtheory: SMIL has been raised from time to time, but no ones seems particularly interested and its not clear what problems it would solve or how to integrate it with HTML |
| 07:58 | <othermaciej> | hsivonen: the emails you linked are indeed particularly interesting |
| 07:58 | <knowtheory> | foolip: Funnily enough, those are all questions i'm seeking to answer ;) |
| 07:59 | <foolip> | knowtheory: there was some discussion in the a11y taskforce not too long ago, perhaps you can search the archives |
| 07:59 | <knowtheory> | a11y? I will do some googling, anything you could recommend i look for specifically? |
| 08:00 | <knowtheory> | ah the accessibility group |
| 08:00 | <foolip> | knowtheory: for reference (i.e. if you want to hit me), I'm Philip Jägenstedt in those threads. |
| 08:00 | <knowtheory> | thanks :) much appreciated. |
| 08:00 | <MikeSmith> | knowtheory: http://lists.w3.org/Archives/Public/public-html-a11y/ |
| 08:01 | <knowtheory> | thanks MikeSmith |
| 08:01 | <othermaciej> | knowtheory: my impression is the same as Philip's; people are generally aware of SMIL, but there is not much interest in using it |
| 08:02 | <knowtheory> | Yep, i appreciate that. |
| 08:03 | <hsivonen> | SMIL is a classic case of a W3C WG doing its own thing without enough implementors of incumbent products participating |
| 08:03 | <knowtheory> | Yeah, i've spent the past 3 months researching wtf happened w/ the SMIL standard |
| 08:03 | <knowtheory> | what it can do, what it can't do, how it interacts w/ other standards |
| 08:04 | <danbri> | what can it do? |
| 08:04 | <danbri> | in practice i mean |
| 08:04 | <danbri> | are there .js implementations yet? |
| 08:04 | <knowtheory> | well, SMIL's general purpose is to provide general timing semantics in a declarative fashion |
| 08:04 | <knowtheory> | there are some components of SMIL that have been built by people here and there |
| 08:05 | <danbri> | the only bit i've historically been interested in was timed text, which wasn't even in the older standards, they deferred to whatever realaudio etc happened to consume |
| 08:05 | <danbri> | so i'm curious if http://www.w3.org/TR/2010/CR-ttaf1-dfxp-20100223/ will get any traction |
| 08:05 | <knowtheory> | yeah. the SMIL3 standard has timed text as a module at this point |
| 08:05 | <danbri> | and happy it got renamed to something saner |
| 08:05 | <knowtheory> | i mean most people know that some of SMIL's timing semantics have been integrated into SVG |
| 08:05 | <danbri> | <- http://lists.w3.org/Archives/Public/public-tt/2010Feb/0004.html |
| 08:05 | <knowtheory> | in order to do very basic animation |
| 08:05 | <foolip> | DFXP/TTML is pretty weird too, at least the styling markup |
| 08:06 | <knowtheory> | yeah i'm aware of TTML though not the details of how weird it can get. |
| 08:06 | <othermaciej> | DFXP has the worst acronym I have ever seen for a technology |
| 08:07 | <knowtheory> | the thing w/ SMIL is that it can provide basic timing semantics, which is handy for a variety of tasks, and is particularly useful when coordinating multiple pieces of content |
| 08:07 | <knowtheory> | i can give you examples of what the company i work for has been doing, and is pushing to continue to do |
| 08:07 | <knowtheory> | if people are interested |
| 08:07 | <othermaciej> | my impression of SMIL is that it feels like something designed for the era of multimedia CD-ROMs |
| 08:07 | <knowtheory> | othermaciej: that's sort of true, and sort of not. My biggest complaint about SMIL right now is that the layout tools feel really clunky |
| 08:07 | <othermaciej> | it doesn't seem to fit that well with what people want to do with animation or timed media today |
| 08:07 | <foolip> | The expansion of DFXP is not very helpful either: Distribution Format Exchange Profile |
| 08:08 | <knowtheory> | which is why i'm getting interested in the possibility of integrating SMIL timing semantics into something with a more mature layout model, like XHTML |
| 08:08 | <othermaciej> | I think what people want is things that integrate really well into HTML/CSS, not a new top-level technology |
| 08:08 | <knowtheory> | Well that's the thing, since SMIL's been modularized it's actually possible to pick and choose components to support |
| 08:08 | <knowtheory> | if it's possible to integrate the standards at a base level |
| 08:08 | <othermaciej> | I don't view that as a plus |
| 08:09 | <foolip> | but HTML has no namespaces, so we'd just be copying the element names |
| 08:09 | <othermaciej> | picking and choosing what components to support leads to fragmentation and lack of interoperability |
| 08:09 | <knowtheory> | Yes, i'm going to have to admit, that i am walking into this somewhat unprepared so i dn't know that i have all the answers or a firm grasp of where the points of pain would be for integration. |
| 08:10 | <knowtheory> | othermaciej: yeah, well w/ SMIL it's used on such a wide variety of platforms with such a wide variety of capabilities |
| 08:10 | <hsivonen> | the design of DFXP is really sad. unfortunately, the spec fills a current accessibility void, so there's a risk that the Web gets stuck with it |
| 08:10 | <knowtheory> | they basically had to modularize the standard |
| 08:10 | <knowtheory> | MMS functions through SMIL for instance |
| 08:10 | <othermaciej> | at least some uses of SMIL actually support such tiny subsets that it's almost silly to pretend they are SMIL |
| 08:10 | <othermaciej> | MMS is a good example |
| 08:10 | <hsivonen> | does anyone use MMS? |
| 08:11 | <knowtheory> | In this day? Probably not |
| 08:11 | <othermaciej> | Apple initially considered putting some SMIL support into WebKit to do it but the subset actually supported and used on real phones is so trivial that it was easier to just handle the common special cases |
| 08:12 | <knowtheory> | I mean, at this point there's like 3 issues. First is why is SMIL the way it is (given it's history), second is what does SMIL actually provide to users at large, and 3rd, is it worth supporting and/or does it have a future as a tech |
| 08:12 | <knowtheory> | othermaciej: sure, i think that there is something to be said for being able to declaritively define timed behavior |
| 08:12 | <knowtheory> | the other alternative being JS |
| 08:13 | <knowtheory> | Depending on what your environment is, and whom your accepting data from, being able to rely on just recieving XML is a good thing |
| 08:13 | <othermaciej> | there's also css animations and css transitions |
| 08:13 | <othermaciej> | although those do not have SMIL's more robust concept of timebases |
| 08:14 | <knowtheory> | yeah. A lot of what we're trying to do specifically involves video, and being able to manipulate it in a variety of ways |
| 08:14 | <knowtheory> | which is why we're interested in what's happening w/ HTML5 |
| 08:15 | <knowtheory> | and SMIL's semantics for doing stuff like that are really good, if you want to concatenate video, or play video along side each other in a single timed environment |
| 08:15 | <knowtheory> | it's pretty solid |
| 08:16 | <knowtheory> | again i need to look at the issue more regarding HTML5s behavior, but i'm not sure how easy something like that would be to do w/ the in browser players that exist now |
| 08:16 | <knowtheory> | there are a bunch of toy usecases that exist already (i've been thinking about this because i intend to talk to people about it at SxSWi) |
| 08:16 | <knowtheory> | for instance, a youtube doubler would be easy to do in SMIL |
| 08:17 | <knowtheory> | and offer better control over timing |
| 08:17 | <knowtheory> | than the existing youtube doubler, which relies on remoting in and out of the flash player w/ JS (although that'll probably change w/ the HTML5 player and direct JS access) |
| 08:18 | <danbri> | 'doubler'? |
| 08:18 | <othermaciej> | knowtheory: concatenation - you'd have to use script |
| 08:18 | <foolip> | knowtheory: it seems the difficult part here is actually getting media to play in sync or concatented without gaps, not the syntax for expressing it |
| 08:18 | <othermaciej> | knowtheory: playing side-by-side - there is no good way to tell the media engine to keep the timebases in sync |
| 08:19 | <knowtheory> | othermaciej: SMIL sets up semantics for dealing with slippage incidentally |
| 08:19 | <hsivonen> | othermaciej: s/the media engine/QuickTime/ ? or more generally? |
| 08:19 | <danbri> | ah http://www.youtubedoubler.com/ :) |
| 08:19 | <foolip> | othermaciej: what is "the media engine"? it seems to be possible in e.g. GStreamer. |
| 08:19 | <knowtheory> | sorry danbri was looking for a particular one to link you to |
| 08:19 | <knowtheory> | but you beat me to the site |
| 08:20 | <knowtheory> | here danbri: http://www.youtubedoubler.com/?video1=http://www.youtube.com/v/HsF_2CNV9v4&start1=3&video2=http://www.youtube.com/v/06CvUjLgK5g&start2=0&h=1 |
| 08:20 | <MikeSmith> | foolip: btw, thanks for the replies about the surveys. I have to say I'm a bit confused about whether the issues you brought up were discussed previously among the media subgroup members |
| 08:21 | <othermaciej> | hsivonen, foolip: I mean there is no good interface at the HTML level |
| 08:21 | <knowtheory> | othermaciej: here's an example i found out today that a friend of mine hacked up of a presentation using HTML5's audio: http://labs.toolness.com/ff-herdict-preso/ |
| 08:21 | <othermaciej> | I know media playback engines *can* keep playback in sync if told do so, there is just no way to tell them |
| 08:21 | <knowtheory> | i told him i'd put that example together a SMIL version of that as well |
| 08:21 | <foolip> | othermaciej: indeed, but that isn't the hard part IMO, the hard part is in the media framework |
| 08:22 | <othermaciej> | knowtheory: syncing with a slideshow type thing is doable - though we have also discussed better ways to do it than the best currently possible |
| 08:23 | <knowtheory> | oh yeah? |
| 08:23 | <knowtheory> | that'd be interesting to hear about |
| 08:23 | <foolip> | MikeSmith: I think it's mostly been me, Silvia and Eric discussing the technical details, so there are plenty of small issues. But I don't think it's meaningful to make it perfect since things will likely be remade after it hits the HTML WG anyway. |
| 08:24 | othermaciej | would like to see it hit the HTML WG ASAP |
| 08:24 | foolip | too |
| 08:25 | <annevk> | subtitle format debates scare me |
| 08:25 | <othermaciej> | knowtheory: there used to be a concept of "cue ranges" |
| 08:26 | <foolip> | annevk: which parts do you find scary? |
| 08:26 | <othermaciej> | annevk: I would lean towards HTML5 being agnostic about supported formats |
| 08:27 | <hsivonen> | othermaciej: that doesn't really address the problem that we need a format the all the relevant vendors agree to implement |
| 08:28 | <hsivonen> | othermaciej: especially if the format is external to the video file |
| 08:28 | <hsivonen> | since in that case there's no value in maintaining the MPEG vs. Xiph duality |
| 08:28 | <othermaciej> | hsivonen: I expect all vendors will support SRT |
| 08:28 | <foolip> | I think no browser vendor has found SRT objectionable, yet |
| 08:28 | <foolip> | it's just that it lacks a spec |
| 08:29 | <hsivonen> | othermaciej: original SRT or RSSesquely extended? |
| 08:29 | <othermaciej> | hsivonen: I expect over time we will also agree on either DXFP or perhaps a subset, but I am less sure of that |
| 08:29 | <othermaciej> | I think trying to encode either or both of those thoughts in the HTML5 spec is likely to spawn flamewars |
| 08:29 | <hsivonen> | RSS <title> all over again scares me |
| 08:29 | <foolip> | say no SRT extensions! |
| 08:29 | hsivonen | agrees with foolip |
| 08:30 | <othermaciej> | I do not really care, but if it lacks a spec, I suppose someone should make one |
| 08:30 | <annevk> | if we can cover all relevant use cases with some SRT extensions that seems favorable over DXFP |
| 08:30 | <foolip> | and I think we shouldn't support the pseudo-HTML typically found in SRT either |
| 08:30 | <annevk> | we'll need some kind of SRT spec to nail all those things down |
| 08:30 | <knowtheory> | SRT? |
| 08:31 | <hsivonen> | annevk: I'd be OK with having a position up vs. position down marker extension in SRT, but the pseudo-HTML is serious badness |
| 08:31 | <foolip> | I expect writing the parsing section for SRT would be fun given the weird stuff that existing in the wild |
| 08:31 | <foolip> | hsivonen: position could be done with CSS couldn't it? |
| 08:32 | <foolip> | hsivonen: I mean completely external to the SRT file |
| 08:32 | <annevk> | foolip, that'd suck |
| 08:32 | <hsivonen> | (I can believe that position up vs. down could be dressed into being a legal requirement, but I have really hard time believing italics is a legal requirement unless someone shows me the relevant chapter and verse in a statute) |
| 08:32 | <annevk> | foolip, you want the ability to move the subtitles around a little to not cover onscreen text |
| 08:32 | <hsivonen> | foolip: I meant positioning a particular string in a particular time window near the top of the video frame when the video frame itself has relevant text at the bottom |
| 08:33 | <MikeSmith> | knowtheory: http://en.wikipedia.org/wiki/SubRip |
| 08:33 | <foolip> | I don't really like the idea of pushing SRT towards the "complex" side of the scale, I'd much rather have one format on each extreme |
| 08:34 | <foolip> | i.e. SRT + something |
| 08:35 | <knowtheory> | thanks MikeSmith :) |
| 08:35 | <foolip> | these legal requirements that are cited from time to time, who do they apply to? TV stations? Set-top box makers? Browser vendors? |
| 08:36 | <knowtheory> | So there's an interest in subtitles/captioning on video? |
| 08:36 | <knowtheory> | foolip: government bodies definitely |
| 08:36 | <foolip> | knowtheory: yes, very much so |
| 08:36 | <knowtheory> | i believe that companies also have an obligation |
| 08:36 | <knowtheory> | but i'm not sure what it is. |
| 08:36 | <knowtheory> | (off the top of my head anyway) |
| 08:36 | <danski> | the laws change from country to country but by and large they do apply to private organisations as well |
| 08:37 | <foolip> | So it could be illegal to not support e.g. positioning subtitles in a software product? That doesn't sound very reasonable. |
| 08:38 | <danski> | foolip: i'm trying to catch up with the discussion in progress here :) there are accessibility requirements that vary from country to country, is what i'm saying |
| 08:40 | <knowtheory> | Okay so basically, I should go justify SMIL's existence ;) and then see whether there's any interest (or even a way) to propose some kind of interoperability w/ HTML5 |
| 08:42 | <hsivonen> | knowtheory: even if you manage to justify SMIL's existence, it's not a given that it's the most appropriate syntax for text/html. |
| 08:43 | <knowtheory> | hsivonen: in what contexts? |
| 08:45 | <hsivonen> | knowtheory: it's not a given that mixing SMIL syntax into text/html is a good idea even if SMIL turned out to have some desirable features |
| 08:45 | <knowtheory> | Actually, what is the deal with XHTML5? someone made mention of a flath namespace? |
| 08:46 | <hsivonen> | XHTML5 is the HTML5 language parsed through an XML parser |
| 08:46 | <knowtheory> | Yeah, i get that |
| 08:46 | <knowtheory> | supports nameespaces and all that fun stuff then right? |
| 08:46 | <knowtheory> | so XHTML5 should support imbedding SVG |
| 08:46 | <knowtheory> | or a SMIL document as a result? |
| 08:46 | <hsivonen> | syntactically, yes |
| 08:46 | <knowtheory> | (even if browsers wouldnt' know what to do with the content) |
| 08:46 | <knowtheory> | yes, fine |
| 08:47 | <foolip> | knowtheory: see the very end of http://hsivonen.iki.fi/xhtml2-html5-q-and-a/ |
| 08:47 | <annevk> | why do we care about SMIL? |
| 08:47 | <knowtheory> | Because HTML5 is converging on the media space, which is what SMIL was made for |
| 08:47 | <knowtheory> | and because SMIL does things like provide timing semantics which don't really exist in other standards |
| 08:48 | <danski> | timing semantics. in this specific example, the synching of captions to video, but there are many other applications. |
| 08:48 | <knowtheory> | and because SMIL is already integrated into other standards, like SVG |
| 08:48 | <knowtheory> | I'm not saying that HTML5 has to support it, but i'd like to try to make a case |
| 08:48 | <knowtheory> | that it might be useful/cool if it did |
| 08:49 | <knowtheory> | And i'm not suggesting that HTML5 should contain SMIL :P |
| 08:49 | <knowtheory> | I was just curious whether there was a possibility for interoperability given this issue: http://www.w3.org/html/wg/tracker/issues/10 |
| 08:49 | <zcorpan__> | SVG uses a forked subset of SMIL, not SMIL proper |
| 08:50 | <knowtheory> | Yes, there's a common functional base though |
| 08:59 | <annevk> | hmm, maybe somewhat |
| 09:00 | <knowtheory> | foolip: thanks for the link |
| 10:03 | <gsnedders> | jgraham: http://www.petapixel.com/2010/03/03/canon-l-lens-look-alike-coffee-mug/ |
| 10:25 | <mut> | hey, is there any way to measure the extremities of a drawing on canvas |
| 10:25 | <mut> | like so i can scale it correctly? |
| 10:36 | <Philip`> | mut: No |
| 10:36 | <Philip`> | (other than keeping track of it manually in your scripts) |
| 10:42 | <othermaciej> | was XMLHttpRequest once specified in HTML5? |
| 10:43 | <zcorpan__> | othermaciej: http://www.whatwg.org/specs/web-apps/2005-09-01/#scripted-http |
| 10:43 | <mut> | Philip` ok cheers... :( |
| 10:43 | <othermaciej> | whoah, I didn't know there were historical copies of the spec on whatwg.org |
| 10:44 | <othermaciej> | is there any way to see what previous versions exist? |
| 10:45 | <zcorpan__> | try url-hacking |
| 10:45 | <zcorpan__> | the 2006 version also has xhr |
| 10:45 | <othermaciej> | were Workers ever in HTML5 or were they always separate (can't remember) |
| 10:46 | <othermaciej> | here is my list of specs spun off from HTML5 so far: |
| 10:46 | <annevk5> | othermaciej, http://www.whatwg.org/specs/ has copies |
| 10:46 | <annevk5> | othermaciej, Web Workers was always separate, though is part of Web Applications 1.0 now |
| 10:46 | <annevk5> | s/copies/pointers/ |
| 10:46 | <othermaciej> | - 2dcontext, microdata, server-sent events, web messaging, web sockets api, web sockets protocol, web sql data base, web storage, xmlhttprequest |
| 10:46 | <othermaciej> | now that I typed that I realize I am missing the sniffing spec |
| 10:47 | <zcorpan__> | workers was initially a separate source doc but was later merged with the main source (but has never been part of the generated html5 spec) |
| 10:49 | <othermaciej> | can anyone else think of specs that got split out besides what I listed? |
| 10:50 | <othermaciej> | I guess there is also Web Addresses, though that is in a messed up state |
| 10:50 | <othermaciej> | also are any of what I listed wrong? |
| 10:51 | <zcorpan__> | window was split but merged back |
| 10:52 | <othermaciej> | I don't think it was ever actually removed from the spec, though I did try to write a separate spec for it at one point |
| 10:52 | <Hixie> | selectors api was originally an issue marker in the html5 spec, though there was never any normative prose written for it in the html5 spec |
| 10:52 | <Hixie> | the alternative stylesheet stuff in cssom |
| 10:52 | <Hixie> | the url stuff |
| 10:52 | <Hixie> | the origin stuff |
| 10:55 | <Hixie> | <device> |
| 10:59 | <annevk5> | the internet |
| 10:59 | <annevk5> | oh wait |
| 11:07 | <MikeSmith> | anybody know any Hadoop committers, or anybody involved with Hadoop that might be interested in doing a presentation about it? |
| 11:08 | <MikeSmith> | annevk5: btw, thank you for having the html5-diff doc pubrules-clean (unlike me, who got a serious beat-down yesterday for not having the h:tml doc in the proper state) |
| 11:12 | <annevk5> | Hixie, in http://www.whatwg.org/specs/web-apps/current-work/complete.html#protocol-overview the ABNF does not look nice |
| 11:14 | <annevk5> | hmm, maybe I don't have the correct fonts or something |
| 11:16 | <Hixie> | "doesn't look nice"? |
| 11:17 | <Hixie> | what doesn't look nice about it? |
| 11:20 | <annevk5> | for some reason no monospace font is selected on my machine |
| 11:20 | <MikeSmith> | the comment for the name-char production wraps in an unsightly way |
| 11:21 | <MikeSmith> | that comment would be less prone to unsightliness if it were to be split into two lines |
| 11:21 | <MikeSmith> | the any-char production risks potential unsightliness as well |
| 11:22 | <MikeSmith> | *the comment for the any-char production |
| 11:24 | <annevk5> | encoding as UTF-8 seems wrong btw for resource name |
| 11:25 | <annevk5> | it would have to be percent-escaped i'd think |
| 11:25 | <annevk5> | also, it seems the connection steps stop mentioning UTF-8 encoded from step 7 on |
| 11:26 | <annevk5> | oh, nm |
| 11:27 | <Hixie> | MikeSmith: i recommend getting a bigger monitor |
| 11:27 | <Hixie> | that way it won't wrap ata ll |
| 11:28 | MikeSmith | sets aside his iPhone and goes to check on his main work machine, a TTY |
| 11:28 | <MikeSmith> | no luck there |
| 11:28 | <Hixie> | i do not buy that you were looking at that page on your iphone |
| 11:28 | <Hixie> | it's a 5MB HTML page of DOOM |
| 11:28 | <MikeSmith> | <snort> |
| 11:29 | <Hixie> | your iphone would probably catch fire if it didn't hang first |
| 11:29 | <MikeSmith> | I don't even have an iPhone, actually |
| 11:29 | <Hixie> | aha! |
| 11:29 | <MikeSmith> | I would welcome the donation of one from a generous benefactor |
| 11:29 | <gsnedders> | Liar! |
| 11:29 | MikeSmith | puts out his pants fire |
| 11:30 | <MikeSmith> | but it does honestly look like crap on my TTY |
| 11:30 | <MikeSmith> | and my Apple IIe |
| 11:30 | MikeSmith | goes back to playing Wizardry |
| 11:47 | <mut> | :/ just realised that the save() and restore() functions on canvas are stacked |
| 11:47 | <mut> | that could have saved me some time about 3 or 4 hours ago |
| 11:47 | <Hixie> | how did you think they worked before? |
| 11:47 | <Hixie> | one-off? |
| 11:47 | <mut> | yea |
| 11:48 | <mut> | dont laugh :P |
| 11:48 | <Hixie> | anything in particular make you think that? if the spec is misleading i can try to fix it |
| 11:48 | <mut> | nah im noob |
| 11:48 | <Hixie> | everyone's a noob when they first read the spec, anything i can do to help noobs is helpful to many :-) |
| 11:48 | <mut> | seems realy obvious now tbh |
| 11:48 | <Hixie> | k |
| 11:48 | <mut> | just didn't realise |
| 11:49 | <gsnedders> | Hixie: Make the spec shorter |
| 11:49 | <mut> | just spotted it in a tutorial and facepalmed |
| 11:49 | <Hixie> | gsnedders: press command+minus a few times |
| 11:50 | <mut> | i love how canvas is similar to cnc gcode :) saved my learning curve |
| 11:50 | gsnedders | doesn't like how he knows nothing about 2D graphics APIs |
| 11:56 | <jgraham> | Hixie: That's not really shorter by any useful metric |
| 11:57 | <Hixie> | jgraham: he didn't specify |
| 11:57 | <jgraham> | I reccommend doing s/[aeiou]// |
| 11:58 | <jgraham> | It is srprsng hw sy it is t rd stff wth n vwls |
| 11:58 | <gsnedders> | You included "i" there. |
| 11:58 | <jgraham> | Er, yes |
| 11:58 | <jgraham> | I failk |
| 11:58 | <gsnedders> | Indeed. Like normal. |
| 11:59 | <jgraham> | I isn't a real vowel anyway |
| 11:59 | <gsnedders> | No, you're jgraham. |
| 11:59 | <gsnedders> | Not much like a consunent either |
| 11:59 | <gsnedders> | And I still can't spell. |
| 11:59 | <jgraham> | Was that supposed to say constanant? |
| 11:59 | <gsnedders> | Yes |
| 12:00 | <jgraham> | Well at least the fail is spread around :) |
| 12:01 | <gsnedders> | jgraham made me take off my headphones :( |
| 12:01 | gsnedders | doesn't like that man in the corner |
| 12:02 | <jgraham> | Oh you were probably only listening to Christian noisecore or some such anyway |
| 12:02 | <jgraham> | I'm sure your ears thanked me |
| 12:02 | <gsnedders> | Within Temptation, actually |
| 12:02 | <gsnedders> | Though I guess symphoic metal is something you'll hate anyway |
| 12:02 | <gsnedders> | I mean, includes "metal" in the name of the genre |
| 12:03 | <jgraham> | Much like Midge Ure, that means nothing to me |
| 12:03 | <workmad3> | gsnedders: do you mean 'symphonic metal' by any chance? |
| 12:03 | <mut> | heh canvas is ace, couldnt think of many applications to use it at first, but applications for it keep popping into my head |
| 12:03 | <gsnedders> | workmad3: Have I not already proven I cannot spell? |
| 12:03 | workmad3 | prefers operatic metal e.g. 'nightwish' personally |
| 12:03 | <workmad3> | gsnedders: very true :) |
| 12:04 | <gsnedders> | workmad3: jgraham prefers the Eels and Radiohead. |
| 12:04 | <gsnedders> | But I guess he could tell you that. |
| 12:04 | <jgraham> | Not exclusively |
| 12:05 | <gsnedders> | Well, I don't like sym[autocomplete this yourself] |
| 12:06 | <gsnedders> | metal |
| 12:06 | <gsnedders> | exclusively |
| 12:06 | <gsnedders> | I pressed return a bit too soon then |
| 12:06 | <gsnedders> | I mean, even the intersection of our music tastes manages to be a non-empty set. |
| 12:06 | <Hixie> | the new stuff in the websocket protocol added 40 lines of code to my simple server |
| 12:07 | <workmad3> | I don't like operatic metal exclusively... I'll listen to most rock or metal and enjoy it to some extent :) |
| 12:07 | <Philip`> | gsnedders: Symbiotic metal? |
| 12:07 | <workmad3> | I don't like RnB, hiphop, rap, or pop though |
| 12:08 | Philip` | prefers plastic to metal |
| 12:08 | <workmad3> | easier to stamp out the same thing time after time? :P |
| 12:09 | jgraham | notes that gsnedders listed two bands and compared then to a genre |
| 12:09 | <gsnedders> | jgraham: Okay, okay… in 48 days I'll start being sensible. |
| 12:09 | <gsnedders> | Whatever "ikag" means. |
| 12:11 | <jgraham> | Ijak |
| 12:11 | <jgraham> | Or rather !Ijak |
| 12:11 | <gsnedders> | Whatever that word we were discussing earlier on the whiteboard means. |
| 12:13 | <Hixie> | woo, my server returns the same handshake as the spec's example |
| 12:13 | <Hixie> | sweet |
| 12:13 | <gsnedders> | Hixie: Congrats, you managed to write Perl without killing yourself. |
| 12:17 | <Hixie> | ok time for bed |
| 12:18 | <MikeSmith> | workmad3: how about folk metal? |
| 12:18 | <MikeSmith> | I'm a hardcore fan of that folkmetal stuff |
| 12:19 | <MikeSmith> | this kind of thing: |
| 12:19 | <MikeSmith> | http://upload.wikimedia.org/wikipedia/commons/0/08/Skyforger_16127_Cernunnos_Fest_11.jpg |
| 12:21 | <daedb> | MikeSmith: You like Finntroll? |
| 12:21 | <jgraham> | That's a real genre? What does it consist of? Metal riffs played on the banjo? |
| 12:21 | MikeSmith | scrambles to look up Fintroll in Wikipedia |
| 12:22 | <jgraham> | Hmm the photo suggests normal metal + bagpipes for more aural pain |
| 12:22 | <MikeSmith> | jgraham: don't mock the gods of folktalic metal |
| 12:22 | <gsnedders> | Those bagpipes don't even have enough pipes to be very loud |
| 12:22 | <daedb> | Finntroll is damn near the ultimate folk metal band... combine metal with humppa and have finns singing in swedish, it's awesome :) |
| 12:23 | <MikeSmith> | daedb: I am mostly a fan of the physique of the Fintrolls |
| 12:23 | <MikeSmith> | http://upload.wikimedia.org/wikipedia/commons/9/96/Finntroll%2C_Metaltown_2008.JPG |
| 12:23 | MikeSmith | wonders what those guys eat, of if they do in fact eat |
| 12:23 | <daedb> | MikeSmith: I was at the concert where that photo was taken :) |
| 12:23 | <MikeSmith> | daedb: I am pretty sure those dudes are tweakers |
| 12:24 | <MikeSmith> | daedb: I don't actually know nothing about them.. I suppose I might seriously like them if I actually listened |
| 12:25 | <MikeSmith> | or especially if I went to a live show |
| 12:25 | <daedb> | I actually expected them to be more fun live than they were, but it wasn't bad. |
| 12:26 | <MikeSmith> | I've found there's a lot of music I don't really care about much if I listen to a recording, but can like it at lot of I see it live |
| 12:26 | <mut> | metalheads |
| 12:27 | <MikeSmith> | I saw this band a while back and there music is a joke but they actually they actually tore up seriously when they played live - |
| 12:27 | <MikeSmith> | http://www.pinkypiglets.com/ |
| 12:28 | <MikeSmith> | jgraham: I'm pretty sure http://upload.wikimedia.org/wikipedia/commons/0/08/Skyforger_16127_Cernunnos_Fest_11.jpg is actually a scene from the movie Spinal Tap |
| 12:28 | <MikeSmith> | and if it's not, it should be |
| 12:31 | <jgraham> | Sight. Metal. Can't live with it |
| 12:31 | <jgraham> | *Sigh |
| 12:31 | <jgraham> | Sigh |
| 12:31 | <gsnedders> | At least you've _listened_ to some, unlike too many people who slag it off endlessly. |
| 12:33 | <MikeSmith> | jgraham: if you think you've seen the worst of it, try some Christian metal |
| 12:33 | <MikeSmith> | that'll change your mind |
| 12:33 | <gsnedders> | Hey, there is some good Christian metal |
| 12:33 | gsnedders | hides |
| 12:34 | <jgraham> | I don't mean to slag it off endlessly or anything. It's just that it seems to be what *all* the kids here (and by kids, I don't just mean gsnedders, I eman anyone under the age of 40) are into |
| 12:34 | <MikeSmith> | "Christian metal" is kind of like "Satanic bake sale" |
| 12:34 | <gsnedders> | jgraham: Welcome to Sweden :P |
| 12:34 | <daedb> | Especially 80's christian metal, like Stryper :p |
| 12:35 | <MikeSmith> | not to use the name of Lord Satan in vain, by the way |
| 12:35 | <jgraham> | Nothing like some cookies of deatho or sponge cake of hell to brighten a rainy sunday morning |
| 12:35 | <MikeSmith> | daedb: yes, the Stryper guys did indeed have some serious armadillos on their trowsers |
| 12:36 | <MikeSmith> | Stryper did spandex better than anybody |
| 12:37 | <jgraham> | In other news, the lack of takers for the Ukelele Orchestra is most disappointing |
| 13:26 | <Lachy> | jgraham, how many people said they'd go? |
| 13:26 | <jgraham> | Lachy: None |
| 13:27 | <jgraham> | Unless gsnedders changes his mind I guess |
| 13:27 | <gsnedders> | I said I might. |
| 13:27 | <jgraham> | s/changes/makes up/ |
| 13:32 | <Dashiva> | Amazing how big signatures some people on public-html have |
| 13:33 | <Dashiva> | And how the readability of quote-heavy threads seems inversely proportional to how much the thread deals with a11y |
| 13:33 | <Lachy> | jgraham, you just need to promote it better. |
| 13:35 | <jgraham> | Dashiva: My understanding is that is because the requirements for making email understandable to blind users are the exact opposite of the requirements to make the conversations understandable to everyone else |
| 13:35 | <jgraham> | This seems to be a deficiency in the email clients avaliable to the blind |
| 13:37 | <jgraham> | Lachy: I have used all my reserves of [whatever the antonym of shyness is] already. |
| 13:38 | <Dashiva> | I wonder how a11y people in general feel about making content inaccessible to non-disabled users for the benefit of disabled users |
| 13:39 | <MikeSmith> | jgraham: what is Ukulele Orchestra? |
| 13:39 | <Lachy> | jgraham, assertiveness is an antonym of shyness |
| 13:40 | <jgraham> | MikeSmith: http://www.youtube.com/user/UkuleleOrchestra |
| 13:40 | <jgraham> | Lachy: Hmm. Maybe |
| 13:42 | <Lachy> | also, the requirements of making e-mail readable by blind people should be the same as for everyone else. But AIUI, the problem is that mail clients don't deal with quotes properly, which seems odd because converting a plain text e-mail to equivalent HTML markup (using <blockquote>, <ul>, etc. is trival |
| 13:43 | <Lachy> | and after the conversion, reading the mail is the same as reading an HTML page. |
| 13:43 | <jgraham> | Lachy: Yes the quoting thing is just what I meant |
| 13:43 | <jgraham> | Non blind people need it to follow the thread. Blind people allegedly get confused |
| 13:46 | <MikeSmith> | jgraham: going to a live performance? |
| 13:46 | <jgraham> | MikeSmith: That's the plan |
| 13:46 | <jgraham> | Although I haven't actually bought any tickets yet |
| 13:47 | <MikeSmith> | where are you guys these days? still Liknoping? |
| 13:47 | <MikeSmith> | *Linkoping |
| 13:47 | <jgraham> | Yes |
| 13:47 | <MikeSmith> | hmm, "Liknoping" sounds better |
| 13:47 | <MikeSmith> | I think I will call it that from now now |
| 13:47 | <MikeSmith> | *now on |
| 13:49 | <annevk> | I just call it Sweden |
| 13:51 | <jgraham> | I think that's quite unfair to the rest of Sweden |
| 13:52 | <MikeSmith> | good god almighty, I did a search just now and I find that there is no heavy metal band yet named Mjöllnir |
| 13:53 | <MikeSmith> | what the hell are these heavy-metal guys thinking? |
| 13:53 | <MikeSmith> | i mean, "Finntroll" vs. "Mjöllnir"? |
| 13:53 | <MikeSmith> | no brainer |
| 13:53 | <boblet> | maybe they’re all too busy playing Halo 3? |
| 13:54 | <MikeSmith> | those guys should might just as well called themselves MoominTroll instead |
| 13:54 | <boblet> | (Master Chief wears Mjolnir-brand underoos) |
| 13:54 | <MikeSmith> | dudes, we need to start a heavy metal band named Mjöllnir, right away |
| 13:55 | <davidhund__> | would this be a good time (and place) to ask the Swedish Heavy Metal Gods a question re: alt attribute on images in a <figure>? |
| 13:56 | <davidhund__> | I was just wondering, from a usability p.o.v., would it be correct to state that an image in a <figure> should always have alternative text (in the 'alt' attribute)? |
| 13:56 | <boblet> | if someone could programmatically turn emotions in email into noise, you wouldn’t even need to write any music; just point it at public-html and press rec |
| 14:00 | <Philip`> | davidhund__: The spec explicitly allows alt to be omitted if you're in a <figure> with a <figcaption> |
| 14:00 | <davidhund__> | But, this means that the figcaption could describe content that's 'not there', right? |
| 14:01 | <Philip`> | and indicates it's the appropriate thing to do when you don't have a real textual alternative to the image (e.g. you don't know what's in the image) but you do have some other text that's related to the image (e.g. a title) |
| 14:02 | <Philip`> | An <img> with no alt isn't 'not there' |
| 14:03 | <Philip`> | <img alt=""> is not there, but just <img> is unknown and dependent on context |
| 14:03 | <Philip`> | (as I understand it) |
| 14:03 | <davidhund__> | hrm |
| 14:05 | <Philip`> | I think the idea is that if you can replace the image with equivalent text, then that text goes in alt; otherwise you shouldn't specify alt at all (and shouldn't put in text that's descriptive but not equivalent - that should go in a <figcaption> instead) |
| 14:05 | <davidhund__> | How would <figure><img ... [some chart] alt=""><figcaption>The latest trends in Foo</figcaption></figure> be helpful to people with screenreaders? |
| 14:07 | <Philip`> | Empty alt="" should only be used for decorative images or similar things where the image can be omitted entirely; you shouldn't use it in that example |
| 14:08 | <davidhund__> | exactly. |
| 14:08 | <davidhund__> | Hence my question: "Would it be correct to say that alt text is required on images in a <figure>" |
| 14:11 | <Philip`> | No, because there are other examples where alt should be omitted (when it's impossible to do anything that would be helpful to people with screenreaders, and putting bogus alt text would be worse than omitting it and having associated non-alt text in a figcaption etc) |
| 14:12 | <Philip`> | like the spec's example of a photo sharing site saying "<figure> <img src="1100670787_6a7c664aef.jpg"> <figcaption>Bubbles traveled everywhere with us.</figcaption> </figure>" |
| 14:13 | <Philip`> | (where the site knows nothing about the image, other than a description of it (which is not equivalent to the image itself)) |
| 14:15 | <davidhund__> | sure, just wondering about the usefulness of the figcaption in that case then… |
| 14:15 | <davidhund__> | thanks Philip' |
| 14:15 | <MikeSmith> | Lachy: I think you should be the part of Loki in our assemblage of rock gods known as Mjöllnir |
| 14:15 | <MikeSmith> | I shall be Thor, of course |
| 14:16 | <MikeSmith> | gsnedders: you will be Baldr, the emo-God |
| 14:16 | <Philip`> | davidhund__: The usefulness is that UAs can automatically determine the text that's associated with the image, and announce it to the user, so the user has better than no idea about what the image is |
| 14:17 | <daedb> | MikeSmith: http://www.vikingarock.se/banden/mjolner.php |
| 14:18 | <daedb> | Different spelling, but there is at least one Mjöllnir-band |
| 14:18 | <MikeSmith> | oh shit |
| 14:19 | <MikeSmith> | well, our first act as a band will have to be to find those guys and kick the crap out of them and steal their name |
| 14:19 | <Dashiva> | I thought your "band" was StandardsSuck |
| 14:19 | <Philip`> | MikeSmith: You should just apply for a trademark on their name, and then get your lawyers to force them to change |
| 14:20 | <MikeSmith> | Dashiva: Marcos is not of Northern European stock, so he can't be in the band |
| 14:20 | <boblet> | MikeSmith: who drives the @w3c twitter account? |
| 14:20 | <MikeSmith> | or Nordic-Germanic |
| 14:20 | <Dashiva> | MikeSmith: He could be the manager |
| 14:20 | <MikeSmith> | boblet: a bot |
| 14:20 | <daedb> | MikeSmith: Also, http://www.metal-archives.com/band.php?id=22679 (different band, same name, seems more serious) |
| 14:20 | <Dashiva> | http://www.bradcolbow.com/archive.php/?p=204 |
| 14:21 | <boblet> | give the bot my thanks :) |
| 14:21 | <Philip`> | davidhund__: (By the way, I don't fully understand the spec's current alt requirements, and think they're crazy (because they're too hard to fully understand or even read), so don't trust anything I say :-) ) |
| 14:21 | <MikeSmith> | Dashiva: we will let Marcos carry our guitar cases and select the fans who have the privilege to come backstage and bask in our glory |
| 14:22 | <MikeSmith> | boblet: the bot says, "give W3C more props" |
| 14:22 | <davidhund__> | Philip' yeah alt-attr seems so simple but I find it quite hard to 'get'. Since a figure is supposed to be a "unit of self-contained content" it is referenced from somewhere, right? So now the subject referenced ('Bubbles') is not described at all. The figcaption is not describing the subject, simply giving more context (in your example) |
| 14:22 | <gsnedders> | w00t! ꦕꦫꦏꦤꦤ꧀ |
| 14:23 | <boblet> | heh |
| 14:26 | <Lachy> | MikeSmith, I have no idea waht Mjöllnir is |
| 14:27 | <Lachy> | and what makes Thor and Loki rock gods? Thor is the god of thunder, I can't remember what Loki is |
| 14:27 | <daedb> | Mjöllnir is Thor's hammer. |
| 14:28 | <Philip`> | I remember some old shareware game where Loki was a snake guy who irritatingly could only be hit from behind |
| 14:29 | <MikeSmith> | jgraham: Lachy's comments make me think that you may be more worthy of donning the mantle of Loki |
| 14:29 | <Philip`> | but fortunately your hammer was like a boomerang so you could do that alright |
| 14:29 | <Philip`> | (I guess you were Thor in the game?) |
| 14:29 | <Philip`> | Also it had skunks |
| 14:29 | <MikeSmith> | jgraham: but I had been thinking you should be Tannhäuser |
| 14:29 | <MikeSmith> | which I realize Tannhäuser is not a god |
| 14:29 | <MikeSmith> | but he is better than a god |
| 14:30 | <MikeSmith> | he is the inspiration for Wagner's greatest music |
| 14:30 | daedb | wants to be Baldr |
| 14:30 | <MikeSmith> | and Wagner is the original inspiration behind all heavy metal |
| 14:30 | <gsnedders> | daedb: MikeSmith already assigned me to him |
| 14:31 | <daedb> | bah |
| 14:31 | <MikeSmith> | as heavy metal Gods, we bow before two ÜberLords, Lord Wagner and Lord Satan |
| 14:32 | <Philip`> | http://www.mobygames.com/game/dos/god-of-thunder/screenshots/gameShotId,4870/ - ooh, that was it |
| 14:32 | <Philip`> | Thor was so cute |
| 14:34 | <MikeSmith> | that is not Thor |
| 14:34 | <MikeSmith> | there can be not cuteness in Thor |
| 14:35 | <MikeSmith> | only anger and righteous vengefulness |
| 14:35 | <MikeSmith> | the greatest incarnation of Thor of all time was this guy: |
| 14:35 | <MikeSmith> | http://en.wikipedia.org/wiki/Beta_Ray_Bill |
| 14:36 | <Dashiva> | Wait, there are HTML action figures? http://www.bradcolbow.com/archive.php/?p=106 |
| 14:36 | <MikeSmith> | this is a guy that was such a baddass he fought the existing Thor and beat him and then became Thor in his place |
| 14:38 | <MikeSmith> | Dashiva: those action figures are missing their orthopedic chairs and laptops and coffee cups |
| 14:40 | <Dashiva> | That sounds more like inaction figures |
| 14:41 | <boblet> | lol |
| 14:43 | <MikeSmith> | Philip`: the spec's current discussion of alt text is baroque to the point that I think the only thing more it needs now is accompanying footnotes or commentary in the margins, like a Thomspson Study Bible |
| 14:44 | <MikeSmith> | I think a graduate course could be constructed around that section |
| 14:44 | <MikeSmith> | taught by Jacques Derrida |
| 14:45 | <MikeSmith> | and/or Jacques De Molay |
| 14:47 | <jgraham> | MikeSmith: In unrelated news, thanks for being roughly the only person on linkedin who isn't entirely po-faced |
| 14:47 | <MikeSmith> | jgraham: I don't understand. my linkedin profile is deadly serious |
| 14:48 | <davidhund__> | but MikeSmith I don't think Derrida would appreciate the spec metanarrative to start… :-P |
| 14:48 | <Dashiva> | I have problems determining whether linkedin is actually professionally useful or not |
| 14:49 | <MikeSmith> | Dashiva: one thing, it's useful for hooking up with sexy women |
| 14:50 | <MikeSmith> | in my experience at least |
| 14:50 | <MikeSmith> | Is it supposed to good for professional stuff of some kind? |
| 14:51 | MikeSmith | is wondering if he somehow has missed understood the "Linked" implications of the "LinkedIn" name |
| 14:51 | <Philip`> | I've only ever used it for stalking people |
| 14:52 | <MikeSmith> | davidhund__: overall, there is too much clarity in the HTML5 spec to interest Derrida |
| 14:52 | <MikeSmith> | he likes ambiguity and confusion |
| 14:53 | <knowtheory> | everything is open to interpretation |
| 14:53 | <boblet> | MikeSmith: Beta Ray Bill seems like an inspiration for the Halo series Brutes |
| 14:53 | <knowtheory> | ;) |
| 14:53 | <davidhund__> | he would be opposed to the many truth claims for sure :-) It's just a powerplay by a small group… :-P |
| 14:53 | <MikeSmith> | davidhund__: that's why I think he would be much more at home in the section that attempts to give guidance on alt text |
| 14:53 | <Dashiva> | MikeSmith: I've got a partial understanding that it's supposed to be useful for making connections, getting job opportunities, etc |
| 14:54 | <MikeSmith> | Dashiva: that doesn't sound like much fun |
| 14:54 | <davidhund__> | Mikesmith agreed re: alt |
| 14:55 | <jgraham> | Dashiva: I would like some evidence to back up that idea. I tend to suspect it is just clever hype to make people buy in |
| 14:56 | <MikeSmith> | boblet: I think Beta Ray Bill was an inspiration for many |
| 14:56 | <MikeSmith> | Walt Simonson is one of the greats of all time |
| 14:57 | <Dashiva> | jgraham: Yes, that's how I feel too |
| 14:57 | <Philip`> | jgraham: Maybe it's just hype, but maybe it's real, and surely you don't want to take the risk of missing out on exciting career opportunities |
| 14:57 | <Dashiva> | If it isn't, what does linkedin offer over, say, facebook? |
| 14:58 | jgraham | has a perfectly fine job at the moment |
| 14:58 | <Dashiva> | jgraham: Sure, but by the time things change it's a bit late to start preparing :P |
| 15:00 | <Philip`> | Maybe tomorrow Microsoft will buy Opera so they can replace Trident and then what'll you do? |
| 15:00 | <jgraham> | Laugh |
| 15:01 | <gsnedders> | jgraham: "fine"? You call a job in a country where it's still -5°C in March _fine_? |
| 15:01 | <jgraham> | I wan't using "fine" as a synonym for "sunny" |
| 15:04 | <beowulf> | MikeSmith: o rly? o.O |
| 15:04 | <Dashiva> | gsnedders: It's a small price to pay for a liveable july |
| 15:04 | beowulf | has another look at LinkedIn |
| 15:05 | <gsnedders> | Dashiva: Well, July in (Eastern) Scotland is liveable… just depressingly gray. |
| 15:06 | <beowulf> | MikeSmith++ # man about town |
| 15:24 | <zcorpan__> | "Thus I don't think having normative dependencies on non-free specifications could be allowed." |
| 15:25 | zcorpan__ | notes that iso 8879 isn't freely available |
| 15:27 | <Philip`> | zcorpan: That seems like a good example of why it's a terrible idea |
| 15:28 | <Philip`> | Someone should complain about the [BEZIER] reference, because that appears to almost entirely not exist |
| 15:29 | <Philip`> | It looks like it probably was published somewhere but I can't find anything except second-hand references to it |
| 15:29 | <Philip`> | (and it's a normative dependency) |
| 15:30 | <zcorpan> | Philip`: go ahead |
| 15:30 | <Philip`> | But the someone would have to come up with a better reference to use instead |
| 15:30 | <zcorpan> | no, someone could just complain |
| 15:30 | <Philip`> | and I couldn't find any that didn't require payment so I can't tell if they actually define what's needed |
| 15:31 | <Philip`> | Someone could, but it wouldn't be very useful |
| 15:31 | <jgraham> | References to things that require payment are almost useless |
| 15:32 | <Philip`> | Almost but not quite |
| 15:32 | <jgraham> | Right, but to the extent that technically less correct, but free, references, should be preferred where possible |
| 15:33 | <jgraham> | and references to non-free documents only included if there is literally no other possibility |
| 15:34 | <Philip`> | The problem with the BEZIER one is it's not non-free, it's entirely unfindable |
| 15:58 | <annevk> | didn't the SVG WG write up the BEZIER math? |
| 15:58 | annevk | wonders how hard it would be for someone to write down the relevant math so we can just include it directly |
| 16:00 | <Philip`> | The maths is trivial, it's just like http://upload.wikimedia.org/math/5/e/3/5e32b674a98a9f70e492851f9ad92b61.png |
| 16:00 | <Philip`> | but Hixie seemingly wanted something as close to the 'original' source as possible |
| 16:03 | <annevk> | seems easier to spec the math and save everyone some time |
| 16:04 | <Philip`> | In practice, it's easier to not include a reference or definition at all (I don't see one in SVG) because everyone knows what it means and/or can look it up on Wikipedia |
| 16:05 | <Lachy> | Philip`, what is that formula you linked to? |
| 16:06 | <Philip`> | Lachy: Cubic Bézier curves |
| 16:09 | <Philip`> | (...from http://en.wikipedia.org/wiki/Bézier_curve#Examination_of_cases) |
| 17:34 | <MikeSmith> | zcorpan, you around? |
| 18:28 | <annevk> | Microsoft just filed a bug on <canvas>: http://www.w3.org/Bugs/Public/show_bug.cgi?id=9189 |
| 18:32 | Philip` | told them to :-) |
| 18:36 | <AryehGregor> | Does anyone know if they'll be implementing it in IE9, or what? |
| 18:36 | <Philip`> | I assume they know, but nothing has been announced |
| 18:37 | <Philip`> | Given the detail that they're reviewing the spec in, I assume it's because they're either implementing it or planning to implement it at some point in the future |
| 18:37 | <AryehGregor> | I figured that much, but I wonder if it's for IE9 or 10. |
| 18:37 | <AryehGregor> | Actually, have they even announced when IE9 will be released? |
| 18:37 | <AryehGregor> | Even approximately? |
| 18:38 | Philip` | has no idea |
| 19:01 | <LoneStar99> | was told to come here for html5 canvas inquires is that correct? |
| 19:02 | <AryehGregor> | LoneStar99, yes, this is one place you can ask about that. |
| 19:08 | <LoneStar99> | i have a drawing program that works, fine but since it is a mobile, app it broke. noticed line 7 of the following makes the code fail http://pastebin.com/fvRpPFeu |
| 19:11 | <LoneStar99> | need to figure out how to not use "context.beginPath();" but lines are still anti-aliased |
| 19:11 | <Philip`> | LoneStar99: It's not clear what that code is meant to be doing |
| 19:12 | <Philip`> | It sounds like you're calling it each time the user clicks somewhere? |
| 19:12 | <Philip`> | If you don't call beginPath, then every time you call lineTo you're making the current path get longer and longer |
| 19:13 | <Philip`> | and each time you call stroke() you're re-drawing every line segment that's ever been added to that path |
| 19:13 | <zcorpan> | MikeSmith: am now |
| 19:13 | <LoneStar99> | yes it is based o this code: http://devfiles.myopera.com/articles/649/example1.js |
| 19:13 | <Philip`> | which might be what makes it look non-antialiased (because it's overdrawing the lines) |
| 19:13 | <Philip`> | What you probably want is to do beginPath, and then moveTo with the *previous* mouse coordinates, and then lineTo with the current mouse coordinates |
| 19:14 | <Philip`> | so that each time you're just drawing a single line between the previous and current mouse positions |
| 19:15 | <Philip`> | Hmm, that example seems to do it wrong (it never resets so it'll stroke longer and longer paths) |
| 19:16 | <LoneStar99> | the weird thing is that drawing works, without "context.beginPath();" but once uncommented, it fails to display |
| 19:16 | <Philip`> | If you just call beginPath and then lineTo, it's not drawing a line from somewhere to somewhere |
| 19:16 | <Philip`> | because you've only given it a single point |
| 19:16 | <Philip`> | which is why you first need to moveTo the previous point |
| 19:18 | <LoneStar99> | how would I moveTo previous point? |
| 19:19 | <LoneStar99> | was chatting with another developer, and he said he does not even use "context.beginPath();" to draw lines... how would I be able to do that? |
| 19:20 | <Philip`> | You need to add some global variables and store ev._x and ev._y in them each time you draw, and then use those values in a moveTo the next time you draw (after beginPath and before lineTo) |
| 19:21 | <Philip`> | If you don't use beginPath then you'll never reset to an empty path, so you'll have the problem of repeatedly drawing over the early parts of the path |
| 19:21 | <Philip`> | (which is an easy mistake because it's hard to notice, but it eventually makes things go really slow and look ugly) |
| 19:21 | <LoneStar99> | simply need to anti-alias anything drawn... what is the best way? |
| 19:23 | <LoneStar99> | this is the sample code working, but without anti-aliasing http://devfiles.myopera.com/articles/649/example1.html |
| 19:25 | <Philip`> | It is antialiasing, but it's got this overdraw problem because it never resets the path |
| 19:25 | <Philip`> | so if you move mouse a hundred times then the first part of the line will have been stroked a hundred times |
| 19:25 | <Philip`> | and the antialiased grey gets drawn a hundred times and turns black |
| 19:26 | <Philip`> | which is why it looks aliased |
| 19:29 | <LoneStar99> | kinda confused... |
| 19:31 | <Philip`> | If you say: ctx.strokeStyle = 'red'; ctx.moveTo(0, 0); ctx.lineTo(100, 0); ctx.stroke(); ctx.strokeStyle = 'green'; ctx.lineTo(100, 100); ctx.stroke(); |
| 19:31 | <Philip`> | then it will stroke the first line segment in red, and then it will stroke *both* line segments in green |
| 19:32 | <Philip`> | because they're both still part of the current path |
| 19:32 | <Philip`> | and that's what that example is doing (except it's a lot more than two line segments, and it's stroking them all in black) |
| 19:35 | <LoneStar99> | i need lines to look like this: http://caimansys.com/painter/index.html |
| 19:35 | <LoneStar99> | not generic looking... |
| 19:36 | Philip` | goes away for a while |
| 20:19 | <Hixie> | annevk: if an implement thinks the maths would have helped them, send me the maths and i'll add it. |
| 20:50 | Philip` | thinks anybody realistically implemented Bézier curves is going to need a lot more information than the equation |
| 20:50 | <Philip`> | *implementing |
| 20:50 | <Dashiva> | Nah, edge cases never happen in the real world :P |
| 20:50 | <Philip`> | It'd only be useful if there was ambiguity about what "cubic Bézier curve" means |
| 20:50 | <Philip`> | but there isn't |
| 20:50 | <Philip`> | because there's only one definition and everybody uses it |
| 20:53 | <Hixie> | i agree, hence "if an implementor thinks" :-) |
| 21:07 | <othermaciej> | yay, drafts are published |
| 21:07 | othermaciej | is glad that is over with |
| 21:10 | <annevk> | oh wait, I guess I confused bezier with compositing |
| 21:11 | <Philip`> | Oh |
| 21:11 | <Philip`> | Not quite the same thing :-p |
| 21:12 | <Philip`> | The original paper on compositing is online for free so that's nice |
| 21:13 | <Hixie> | othermaciej: how long did the process take from start to finish? |
| 21:13 | <othermaciej> | Hixie: what do you count as the start? |
| 21:13 | <annevk> | yeah I know it's not the same, didn't realize bezier curves had a similar issue |
| 21:13 | <Hixie> | othermaciej: when you decided we should try to publish, i guess |
| 21:14 | <othermaciej> | start of Call for Consensus to publish: Feb 10 |
| 21:14 | <othermaciej> | recorded WG decision to publish: Mar 24 |
| 21:15 | <othermaciej> | er |
| 21:15 | <othermaciej> | Feb 24 |
| 21:15 | <othermaciej> | so about a month, or 8 days, depending on how you care to count |
| 21:15 | <Hixie> | not too bad i guess |
| 21:15 | <othermaciej> | this signals that if we are serious about the heartbeat requirement, the next publication round should start in 2 months |
| 21:15 | <Hixie> | yeah |
| 21:17 | <Dashiva> | As discussed in #html-wg earlier, the whole process might get easier if we published more than heartbeat required |
| 21:25 | <LoneStar99> | Philip': my code is posted in the last thread of this link http://uoid.me/canvas2 |
| 21:42 | <othermaciej> | Dashiva: I could get behind publishing monthly |
| 21:43 | <Philip`> | LoneStar99: I think you need something like http://pastie.org/854584 |
| 21:47 | <LoneStar99> | Philip': something random the original code was written by a dude that just left the chat... |
| 21:47 | <LoneStar99> | sorry the channel "ROBOd " |
| 21:52 | <colinsullivan> | Can anyone help me with an HTML5 audio implementation? |
| 21:53 | <LoneStar99> | Philip': that works perfect! |
| 21:53 | <LoneStar99> | thank you so muc |
| 21:53 | <LoneStar99> | h |
| 22:13 | <Hixie> | well, we've not put JavaScript on the TR/ page |
| 22:14 | <Hixie> | now, even |
| 22:23 | <LoneStar99> | Philip': i noticed it works, but when I draw anything, it automatically connect to another line... |
| 22:28 | <othermaciej> | Hixie: how come "hide UA text" gets a colored border but "highlight UA text" does not? |
| 22:55 | <Hixie> | othermaciej: because it's obvious that text is highlighted, but not at all obvious that text is hidden |
| 22:56 | <Hixie> | othermaciej: and i needed a way to remind myself that text was hidden when i was in UA mode so i wouldn't spend hours looking for text that was hidden |
| 23:12 | <othermaciej> | Hixie: heh, I see |
