| 00:02 | <jwalden_> | hsivonen: #undef UNICODE /* Stupid Windows CE C++ compiler. */ ? |
| 00:10 | <annevk42> | jwalden_, it's a reserved word or something |
| 00:12 | <jwalden_> | annevk42: you sure? my experience is that most collisions are from idiotic macro names in the Windows SDK |
| 00:13 | <jwalden_> | and an #undef takes care of them, as long as you don't need to use the name in a header that might shared a large number of places |
| 00:14 | <annevk42> | that's what https://bugzilla.mozilla.org/show_bug.cgi?id=497580 suggests |
| 00:14 | annevk42 | follows bugs with "html5" |
| 00:21 | <jwalden_> | it really is quite appropriate that the abbreviation for that platform is WINCE |
| 00:21 | <annevk42> | is WINCE Windows Mobile? |
| 00:32 | <jcranmer> | Windows Compact Edition |
| 00:32 | <jcranmer> | so yes |
| 01:47 | <doublec> | Windows Mobile is based on WINCE but the version numbers are different. Windows Mobile 6 is not based on WINCE 6 for example. |
| 01:47 | <doublec> | it's based on WINCE 5 |
| 08:36 | <jgraham> | Finally an <audio> demo: http://evilbrainjono.net/piano.html Bets on when the first song made like this reaches the charts? |
| 09:10 | jwalden | wishes so many keyboard-using demos didn't have issues when the browser in use has a default action associated with the key in question |
| 09:11 | <jwalden> | find as you type horks things like that pretty hard |
| 09:15 | <hsivonen> | jwalden: yeah, I can probably just #undef UNICODE. silly me. (shows my newbieness with approaching Windows compilation issues) |
| 09:16 | <jwalden> | not being corrupted in that manner is not necessarily bad :-) |
| 09:19 | <Philip`> | Re last night (UTC): Surely a URL that has totally different meanings in different contexts would be violating some fundamental architecture of HTTP? (like the whole "Uniform" bit of "URL") |
| 09:21 | <hsivonen> | Philip`: if it's not an HTTP URL, you wouldn't be violationg HTTP but maybe AWWW |
| 09:22 | <annevk42> | context-dependent identifiers are acceptable |
| 09:22 | <annevk42> | e.g. localhost |
| 09:30 | <jwalden> | heck, all of DNS |
| 11:03 | <zcorpan> | http://forums.whatwg.org/viewtopic.php?t=49#5022 - in a couple of days, another newly registered user will reply with a URL, and i will delete both posts as spam |
| 11:05 | <zcorpan> | http://www.google.com/search?q=%22What+is+the+best+GPS+Navigation+system+I+can+buy+for+my%22 |
| 11:05 | zcorpan | deletes it now instead |
| 11:43 | <annevk42> | whoever invented sync XHR be doomed |
| 11:45 | <jgraham> | annevk42: We're *all* doomed (for some sutiable definition of doomed) |
| 11:55 | <annevk42> | jgraham, that's not helpful |
| 12:02 | <jgraham> | annevk42: I think introducing "imminent doom" as the punishment for inventing bad apis would rapidly deplete the already worryingly empty pool of wwilling specification editors |
| 12:02 | <jgraham> | Although maybe it is a fairer punishment for introducing bad non-standard apis that everyone else than has to adopt |
| 12:02 | <jgraham> | Since it would encourage people to seek peer review |
| 12:09 | <annevk42> | again, not helping! |
| 12:09 | <annevk42> | :) |
| 12:24 | hsivonen | finds reftests failing due to HTML5 now discarding incomplete tag tokens |
| 12:24 | <hsivonen> | apparently not trying to test incomplete tokens, though |
| 12:26 | <jgraham> | hsivonen: How come the reftests have all these weird dependencies on parsing? |
| 12:26 | <jgraham> | (apart from ones that test parsing) |
| 12:26 | <hsivonen> | jgraham: they are written by people |
| 12:27 | <jgraham> | hsivonen: OK, so it is just bugs in the tests rather than something fundamental |
| 12:27 | <jgraham> | That's basically what I was wondering |
| 12:28 | <annevk42> | from some earlier conversation about this it was mostly autogenerated tests from Philip` that had this issue |
| 12:37 | <hsivonen> | anyway, lacking ">" in the string set as the value of innerHTML is something to watch out for in case it matters for Web compat |
| 12:40 | <rubys> | hsivonen: http://intertwingly.net/blog/2009/06/12/Validator-Nu-on-GCJ |
| 12:41 | <hsivonen> | rubys: cool! |
| 12:43 | <jgraham> | rubys: Did you run the testsuite? |
| 12:44 | <rubys> | against HTML2XML? That's all I have built so far. |
| 12:44 | <jgraham> | Against the htmlparser |
| 12:44 | jgraham | guesses no then |
| 12:45 | <jgraham> | (Also I make not be making much sense since I don't really know how hsivonen's code is organised) |
| 12:45 | rubys | mutters people are *so* impatient. :-) |
| 12:46 | hsivonen | deals with *new* reftests that assume <col> becomes a child of <table> |
| 14:07 | hsivonen | finds a test case failing because <div> breaks out of <p><span> in HTML5 |
| 14:52 | gsnedders | buys headphones |
| 14:52 | gsnedders | can't believe he really just spent that much on headphones :P |
| 14:56 | <Lachy> | gsnedders, what type of headphones did you buy and how much? |
| 14:56 | <gsnedders> | Lachy: Audio-Technica ATH-M50 |
| 14:57 | <gsnedders> | just under £120 |
| 14:59 | <Lachy> | why did you need such expensive headphones? |
| 15:00 | <gsnedders> | Lachy: audiophilia? :P |
| 15:00 | <gsnedders> | Lachy: Also needing something with very little sound leakage, and comfy |
| 15:01 | <Lachy> | Do you have exceptional hearing ability, and are you able to hear the difference in quality? |
| 15:01 | <gsnedders> | Lachy: yes |
| 15:02 | <gsnedders> | Lachy: Also, a lot of cheaper headphones (esp. those under £60, the real reason for the high price is getting one which meets all three requirements) have either weak bass of treble which almost anyone will notice |
| 15:03 | <Lachy> | can you also tell the difference between a lossless recording and a high quality, though compressed, 256kbps AAC or 320kbps MP3? |
| 15:03 | <gsnedders> | No, my hearing isn't that good. |
| 15:03 | <annevk42> | whoa, I thought my EUR 30 headphones were expensive |
| 15:04 | <Lachy> | annevk42, that's about the price range of my headphones, I think. They're good enough for me |
| 15:05 | gsnedders | would get annoyed at the sound quality of more or less all 30 EUR headphones :P |
| 15:11 | <Lachy> | I want to get a good quality headset to replace my USB headset, which instead works with the 3.5mm optical output on my Mac. But I haven't found one yet. |
| 15:23 | jgraham | has lousy hearing but quite nice 80EURish in ear headphones |
| 15:23 | <jgraham> | Which I bought for the noise reduction rather than the playback quality |
| 15:24 | <jgraham> | Although the build quality is insufficent |
| 15:25 | <Lachy> | anyone know of a good quality, stereo, wireless headset/mic, suitable for general use, like skype calls or youtube videos? |
| 15:39 | <gsnedders> | jgraham: in ear headphones I tend to find uncomfortable, and also normally a lot of sound leakage |
| 15:39 | <gsnedders> | (And also tend to be too small to get any really good quality) |
| 15:43 | <jgraham> | gsnedders: WFM |
| 15:43 | <ezyang> | I have PX 200 headphones: they're small, but surprisingly good quality for their size. |
| 15:44 | <jgraham> | I mean they're not super comfortable for many hour periods but for being on a noisy office or, especially, on a plane they make much more sense than anything else |
| 15:44 | <gsnedders> | ezyang: But they're Sennheiser headphones, so that's not overly surprising :P |
| 15:44 | <ezyang> | I <3 my headphones |
| 15:45 | jgraham | doesn't understand why gsnedders thinks they wwould have a lot of sound leakage |
| 15:46 | <gsnedders> | jgraham: They tend not to have much to stop the sound from getting out |
| 15:46 | <ezyang> | If it sounds good, does it really matter? |
| 15:46 | <jgraham> | gsnedders: The ones I'm talking about are the ones that are basically an earplug with a headphone in the middle |
| 15:47 | <gsnedders> | jgraham: I know |
| 15:47 | <gsnedders> | ezyang: If you're in an office, you don't want sound leakage. |
| 15:47 | <ezyang> | I have an officemate who plays music on speakers, so YMMV |
| 15:47 | <jgraham> | gsnedders: They have a ear plug thing to stop sound getting out |
| 15:48 | jgraham | hasn't actually measured this so could be wrong |
| 15:48 | <jgraham> | Also I tend to be able to play music at lower volumes because it is quieter |
| 15:48 | gsnedders | tends to listen to music fairly quietly too |
| 15:48 | ezyang | cranks it up |
| 15:53 | <Lachy> | I can't listen to music on headphones and work at the same time. It's too distracting |
| 15:53 | <Lachy> | though, I have less difficulty if the music is on relatively low in the background |
| 16:05 | <Lachy> | Apparently Apple are changing the default gamma in Snow Leopard to match Windows. http://www.tuaw.com/2009/06/12/make-your-displays-gamma-in-leopard-match-snow-leopard/ |
| 16:05 | <Lachy> | so at least in theory, colours should start to look the same on both platforms |
| 16:07 | <jgraham> | Lachy: We had this conversation a few days ago. For images it still depends if the image has a built-in colour profile and if your viewing software supports it or not |
| 16:09 | <Lachy> | jgraham, I know |
| 16:31 | <jgraham> | A hint to people writing testsuites: the more complex you make your harness, the more complex it is to figure out what is actually going on |
| 16:51 | <anne-m> | mpt, if I want Ubuntu to get the wireless driver issues to be sorted out; where would the best place be to donate money? |
| 16:52 | mpt | blinks |
| 16:52 | <anne-m> | mpt, due to some upgrade my mother's computer suddenly switched to the non-proprietary drivers that made DNS resolving suck bigtime |
| 16:52 | <mpt> | from the Random Questions Department :-) |
| 16:53 | <mpt> | hm, one moment, I'll ask around |
| 16:53 | <anne-m> | mpt, took me a long time (20min or so) to figure out it was a driver issue and how I could switch back to the proprietary driver |
| 16:53 | <anne-m> | mpt, you're basically the only person I know that is closely involved :) |
| 16:55 | <mpt> | anne-m, do you happen to know which driver it is? |
| 16:56 | <mpt> | (Different people work on different drivers) |
| 16:56 | <anne-m> | the proprietary one is madwifi and the non-proprietary one (that I had to disable) is ath5k |
| 16:58 | <mpt> | madwifi and ath5k are both open-source drivers produced by the MadWifi project <http://madwifi-project.org/>; |
| 16:58 | <mpt> | "ath5k is a relatively new and emerging driver and ... is intended to replace MadWifi in the long run" |
| 16:59 | <gauthierm> | What events should be fired when dragging the scrubber on a media element? FIrefox alphas fire play/pause events as well as seeking and seeked. Safari fires only seeking/seeked. |
| 16:59 | <mpt> | anne-m, so the answer to your question is <http://madwifi-project.org/wiki/Donations>; |
| 17:03 | <anne-m> | thanks mpt (though weird that "Hardware Drivers" says madwifi is proprietary) |
| 17:03 | virtuelv | applauds the "where can I donate" attitude instead of "xxx sucks" |
| 18:49 | <thin> | how many lines of code to draw a triangle, stroke it black and fill it grey? |
| 18:49 | <thin> | with canvas |
| 18:50 | <thin> | approximately |
| 18:50 | <annevk42> | 1 |
| 18:51 | <thin> | will it someday be possible to generate pixel data with canvas/javascript, drop that into the dom as one of those data nodes, then use that image data as a background image in my css? |
| 18:52 | <gauthierm> | probably already possible, but you'd have to write an encoder to encode the pixel data. |
| 18:55 | <thin> | for certain small images do you think doing what I described might be a faster option than using an actual image file? |
| 18:59 | <thin> | Are there any open source, web-friendly image formats trying to compete with jpg/gif/png like ogg is with video/audio? |
| 18:59 | <annevk42> | I think WebKit has an extension to CSS for that |
| 18:59 | <annevk42> | thin, JPEG/PNG/GIF are not open enough for you? |
| 19:00 | <thin> | wasn't there some hub-bub about gif files a few years back? |
| 19:00 | <annevk42> | But you can already do toDataURL() on your <canvas> and then plug that into your CSS through the CSSOM |
| 19:00 | <annevk42> | thin, I think that patent expired by now |
| 19:00 | <annevk42> | thin, or was invalidated |
| 19:00 | annevk42 | forgot |
| 19:01 | <thin> | are you anne van kesteren? |
| 19:03 | <annevk42> | thin, yes |
| 19:03 | <thin> | cool, I like your blog |
| 19:03 | <annevk42> | :) |
| 19:03 | <gauthierm> | thin: png was created as a response to the gif patent. |
| 19:36 | <annevk42> | donated |
| 19:42 | jgraham | wonders how donations to small software development efforts works |
| 19:42 | <jgraham> | Like how does the money affect the development |
| 19:43 | <jgraham> | I can understand "here is $x to do y" |
| 19:44 | <jgraham> | But "Here is $x to make it better" doesn't have any obvious impact unless sum($x_i) is enough for the developers to give up (part of) their outside job |
| 19:45 | <gauthierm> | Is anyone here working on the HTMLMediaElement spec? |
| 19:45 | <annevk42> | jgraham, I'd imagine part of the money is used for buying the hardware |
| 19:46 | <annevk42> | jgraham, but I trust them to do the right thing |
| 19:46 | <jgraham> | Oh I guess for a hardware driver project that does make sense |
| 19:46 | <annevk42> | gauthierm, as in editing it? I guess Hixie |
| 19:47 | <Hixie> | gauthierm: i'm editing that part of html5 right now |
| 19:47 | <jgraham> | annevk42: I'm not suggesting you did anything silly. I'm just curious what the dynamics are when one donates to a (pure software) project |
| 19:47 | <gauthierm> | Hixie: Cool. I'm having some trouble with media events when the scrubber is clicked and dragged. |
| 19:48 | <gauthierm> | Some browsers fire a 'pause' when you click and hold the scrubber and others do not. |
| 19:48 | <annevk42> | jgraham, I hope it actually works; I've had issues using my MasterCard online |
| 19:48 | <gauthierm> | Hixie: Is one implementation correct, or is that not covered by the spec? |
| 19:50 | <Hixie> | gauthierm: that's basically a user interface issue, so it depends on whether the browser implementor wants the video to continue playing when you let go of the scrubber or not |
| 19:50 | <Hixie> | gauthierm: or whether they want it to play as you're dragging |
| 19:51 | <annevk42> | Hixie, so the way this "Geopriv" works is that the user sets a bunch of variables and the author then either honors them or does his own thing? |
| 19:51 | <gauthierm> | ok. The extra events are causing a problem for me in Firefox in particular. Perhaps I should bug them to fire extra events when the interface is interacted with. |
| 19:51 | <gauthierm> | afaik, both Firefox and Safari's interface is the same, but the events fired are different. |
| 19:52 | <Hixie> | annevk42: as far as i can tell, except that the settings get set in the browser's ui instead of the page's ui, so the user thinks it's the browser who's going to control it (and who is to blame, of course, when it inevitably gets ignored) |
| 19:52 | <Hixie> | gauthierm: which events? |
| 19:52 | <annevk42> | Hixie, wow :/ |
| 19:52 | annevk42 | sends email |
| 19:52 | <gauthierm> | Hixie: Firefox fires play/pause when you drag the scrubber. Safari does not. |
| 19:53 | <Hixie> | annevk42: note that the proposal isn't really geopriv, it's something else. Geopriv itself is an XML format for transmitting geo data with privacy controls, which i think makes a lot of sense (and has nothing to do with UI) |
| 19:53 | <Hixie> | gauthierm: does safari keep playing the video if you just hold the scrubber without moving it? |
| 19:53 | <gauthierm> | Hixie: checking now |
| 19:53 | <annevk42> | Hixie, mkay, somewhat confusing title then |
| 19:53 | <Hixie> | annevk42: yeah |
| 19:54 | <gauthierm> | Hixie: Safari does pause playback when you drag teh scrubber. |
| 19:56 | <Hixie> | gauthierm: then that's a safari bug, it should be pausing the video in the api too |
| 19:58 | <gauthierm> | ok. It would be nice from a JavaScript developer's perspective if there was a way to know whether the video was paused through clicking the pause button or paused through dragging the scrubber. |
| 19:59 | <gauthierm> | The way it's implemented in Mozilla at least, the two are impossible to distinguish. |
| 20:03 | <gauthierm> | Hixie: If I'm filing a bug with Webkit, is that behaviour specified in section 4.8.10.11? |
| 20:06 | <Hixie> | gauthierm: yeah "Where possible (specifically, for starting, stopping, pausing, and unpausing playback, for muting or changing the volume of the audio, and for seeking), user interface features exposed by the user agent must be implemented in terms of the DOM API described above, so that, e.g., all the same events fire." |
| 20:14 | Philip` | got "here is $x to do y" for his font subsetting tool, which was nice and unexpected |
| 20:15 | <Philip`> | hsivonen / annevk42: Re URLs: Oh, good point(s) |
| 20:19 | <annevk42> | URLs? My memory is failing me... :/ |
| 20:20 | <Philip`> | annevk42: http://krijnhoetmer.nl/irc-logs/whatwg/20090612#l-305 |
| 20:21 | <annevk42> | aah |
| 20:35 | <annevk42> | Hixie, the second of "similar note in audio codecs section" should have s/audio/video |
| 20:40 | <Hixie> | d'oh |
| 20:41 | <Hixie> | btw |
| 20:41 | <Hixie> | i hate software patents |
| 20:41 | <Hixie> | 16bit wave? really? that's the best we can do? |
| 20:41 | <Hixie> | and nothing at all for video? |
| 20:41 | <Hixie> | what bs |
| 20:42 | <sicking> | flac is patented? |
| 20:43 | sicking | is surprised that microsoft doesn't have patents on wave |
| 20:43 | <Hixie> | apple won't implement codecs that have a submarine patent risk that they haven't already taken on |
| 20:43 | <sicking> | *that* is bs |
| 20:44 | <sicking> | and a lie |
| 20:44 | <sicking> | *everything* has submarine patent risk |
| 20:44 | <annevk42> | lots of big companies have taken the risk with vorbis, including Microsoft IIRC |
| 20:44 | <Hixie> | mozilla won't implement codecs that require patent licenses unless they can cover their redistributors, which they can't |
| 20:44 | <Hixie> | opera won't implement codecs that require paying money |
| 20:44 | <Hixie> | sicking: "that they haven't already taken on" is a key part of that sentence |
| 20:45 | <Hixie> | and i'm not arguing about who's right or wrong |
| 20:45 | <Hixie> | because frankly it doesn't matter |
| 20:45 | <sicking> | Hixie, no, i mean implementing cross-site XHR has submarine patent risk |
| 20:45 | <Hixie> | they could not want to implement codecs that didn't start with an "A" and we'd still be in the same boat |
| 20:45 | <sicking> | Hixie, implementing @font-face has submarine patent risk |
| 20:45 | <Hixie> | sicking: argue with them, not me |
| 20:46 | <sicking> | Hixie, implementing anything new has submarine patent risk |
| 20:46 | <sicking> | Hixie, i know |
| 20:46 | <Hixie> | you'll note my employer is implementing everything :-) |
| 20:46 | <Hixie> | ogg theora, vorbis, 264, you name it |
| 20:46 | <sicking> | well, I can't say that I'm happy your employer is implementing things with known patents |
| 20:46 | <sicking> | but that's a separate conversaion |
| 20:46 | <sicking> | well, actually it's related |
| 20:47 | <Hixie> | yeah, well |
| 20:47 | <Hixie> | if the unpatented formats didn't suck... |
| 20:47 | <Hixie> | but anyway |
| 20:47 | <Hixie> | i don't want to argue about it |
| 20:47 | <Rik|work> | Hixie: i really think that's no big deal for the spec, we'll see something similar to what happened with jpg, png and gif |
| 20:47 | <Hixie> | i just want to whine that this makes my life hard from the html5 perspective |
| 20:48 | <sicking> | the thing is, the "we won't implement anything with submarine patent risk" argument is BS. What it really is is "the risk/reward ratio isn't good enough right now" |
| 20:48 | <Hixie> | Rik|work: maybe, though the codec space is _much_, _much_ more patent-encumbered. |
| 20:48 | <Hixie> | sicking: yeah i was simplifying their position for effect, you're right that it's really about risk/reward. |
| 20:48 | <sicking> | we can't do anything to lower the risk part, so we'll have to up the reward part |
| 20:48 | <Rik|work> | gif was patent-encumbered and yet everybody used it |
| 20:48 | <sicking> | which means that we have to get people to use ogg, and only ogg |
| 20:48 | <annevk42> | the GIF patent only covered encoding |
| 20:48 | <annevk42> | not decoding |
| 20:49 | <Hixie> | Rik|work: (a) it wasn't thought to be originally, (b) it was only encoding that was covered, so browsers were safe, and (c) yeah, that's why people are paranoid now |
| 20:49 | <Hixie> | anyway |
| 20:49 | <Hixie> | i should go get lunch |
| 20:49 | <Hixie> | bbl |
| 20:50 | <sicking> | Rik|work, what hixie said. GIF isn't a proof doing patented stuff is fine. It's a lesson costly learned |
| 20:55 | <gauthierm> | Mozilla's doing some good work to make theora suck less. |
| 21:03 | <sayrer> | sicking: the key part is "not arguing about right and wrong" |
| 21:05 | <sayrer> | oh wouldn't it be nice to take an amoral stance regarding <video> |
| 21:06 | <sayrer> | Mozilla could drop Windows Media in there |
| 21:06 | <sayrer> | and we'd have a lot of the Web taken care of |
| 21:09 | <annevk42> | sayrer, isn't that sort of what enabling GStreamer does? |
| 21:10 | <sayrer> | yes, but if we did (have? not sure if we went for it) I would and do openly disagree that it is the right thing |
| 21:11 | <sayrer> | I wouldn't make up some absurd rationalization |
| 21:12 | <annevk42> | not yet apparently |
| 21:12 | <sayrer> | yeah, I don't like it, but I also haven't spent much time working on the mobile product or the video element |
| 21:13 | <annevk42> | (I actually thought it was in, but https://bugzilla.mozilla.org/show_bug.cgi?id=422540 suggests it isn't) |
| 21:13 | <sayrer> | mozilla also doesn't publish mass quantities of h264 video |
| 21:13 | <sayrer> | so there is that |
| 21:14 | <sayrer> | anyway, the whole "some people have paid MPEG-LA, some people haven't, so it's a tossup" argument is just such bullshit |
| 21:15 | <sayrer> | try applying that logic to any number of historical dilemmas |
| 22:08 | <darbour> | I'm having a problem with htmllib5 sanitizer if anyone can help |
| 22:09 | <darbour> | when I give "<p class="some-p">LALALLALA</p>" the sanitizer returns "<p class="some-p">blarg\n" |
| 22:09 | <darbour> | * replace LA* with blarg |
| 22:11 | <jgraham> | darbour: In theory I can help but a) I'm about to go to bed and b) that seems rather odd |
| 22:12 | <jgraham> | Can you give an actually running example? |
| 22:13 | <sayrer> | I think he means that it eliminated his closing tag |
| 22:13 | <darbour> | exactly |
| 22:14 | <darbour> | sorry I was a little vague there |
| 22:14 | <sayrer> | it did that because it is not strictly necessary |
| 22:14 | <sayrer> | also, the sanitizer is pedantic |
| 22:14 | <sayrer> | there you go |
| 22:14 | <sayrer> | :) |
| 22:14 | <jgraham> | Oh. It sounds like it's the serializer setup you have or something |
| 22:15 | <jgraham> | Running code would still help :) |
| 22:15 | <sayrer> | I wonder if the html5lib sanitizer replaces font elements |
| 22:15 | <darbour> | for <p class="first-p">1<br/>2</p> it returns <p class="first-p">1<br>2 |
| 22:15 | <darbour> | where would i begin to look to stop this behavior? |
| 22:16 | <jgraham> | darbour: What do you mean "returns" usually the sanitizer returns an in-memory tree, not a string. How are you creating the string? |
| 22:17 | <annevk2> | hmm, dropping the closing tag seems problematic |
| 22:17 | <jgraham> | annevk2: I think that somewhere somethin is being serialized with omit-optional-tags=True |
| 22:18 | <jgraham> | Er but not quite called that |
| 22:18 | <darbour> | should I look at the html5lb serializer code? |
| 22:20 | <jgraham> | darbour: In the HTMLSerializer there is an attribute omit_optional_tags which is True by default. I think if you set this to False on your serializer instance your problems will go away |
| 22:21 | <darbour> | awesome, I'll try that. Thanks for the help! |
| 22:47 | <annevk2> | I tend to think http://intertwingly.net/blog/2007/05/08/Dont-Break-The-Web#c1178698369 is still a pretty good answer to part of the conformance question |
| 23:26 | <annevk2> | sicking, could be that I missed something, but I believe it's just Mozilla that pays hsivonen for the validator work |
| 23:26 | <sicking> | annevk2, i thought w3c was planning on using some of his work too. Not sure if that has been decided for the HTML5 parser yet |
| 23:27 | <Hixie> | validator.w3.org uses some of the validator.nu code |
| 23:27 | <Hixie> | though in my experience hsivonen's instance is more reliable |
| 23:31 | <annevk2> | http://groups.google.com/group/native-client-announce/browse_thread/thread/e69fe64e8decbe16 -- interesting, albeit scary |
| 23:33 | <annevk2> | Hixie, a DOMString can contain a zero byte too, that'd be problematic as well for Web Sockets |
| 23:33 | <Hixie> | why? |
| 23:33 | <annevk2> | isn't that the end marker? |
| 23:34 | <Hixie> | no, oxFF is |
| 23:34 | <Hixie> | 0xFF even |
| 23:34 | <olliej> | annevk2: Hixie: is there not a length property instead |
| 23:34 | <Hixie> | for binary data yes |
| 23:34 | <olliej> | all data should be considered binary |
| 23:34 | <Hixie> | not for text though |
| 23:34 | <olliej> | 0xff can occur in valid unicode |
| 23:34 | <Hixie> | not in utf-8 |
| 23:35 | <olliej> | what about utf16 |
| 23:35 | <sayrer> | annevk2: none of hsivonen's points seem to apply to @summary or <font> very well |
| 23:35 | <Hixie> | websocket uses utf-8 |
| 23:35 | <olliej> | urgh |
| 23:35 | <olliej> | so we have to do a utf16-utf8 conversino :-/ |
| 23:35 | <annevk2> | sayrer, yup |
| 23:36 | <Hixie> | olliej: i'd rather you have to do it than have every single web author out there have to do it |
| 23:36 | <annevk2> | sayrer, Hixie made a few additional points that seem to cover those |
| 23:36 | <sayrer> | annevk2: not really |
| 23:36 | <sayrer> | but that's ok, I don't have to argue it |
| 23:37 | <annevk2> | me neither :) |
| 23:37 | <sayrer> | I'm just going to fork it :) |
| 23:37 | <sayrer> | now that I have some time |
| 23:38 | <annevk2> | good idea |
| 23:38 | <annevk2> | hopefully the extra competition will turn into something useful |
| 23:39 | <sayrer> | maybe |
| 23:39 | <annevk2> | Hixie, ah ok |
| 23:39 | <sayrer> | I'm just going to cut, and insert text that lets anyone extend it. hopefully the extra competition turns into something useful. |
| 23:39 | <Hixie> | annevk2: hm? |
| 23:40 | <annevk2> | Hixie, my misreading of Web Sockets |
| 23:40 | <Hixie> | ah |
| 23:42 | <annevk2> | IETF drafts are distinctly unreadable :/ |
| 23:42 | <sayrer> | yeah, they need decorations about how they don't work |
| 23:43 | <annevk2> | they could use some hypertext, for starters :) |
| 23:45 | <sayrer> | and some newspeak |
| 23:46 | <sayrer> | they already of the amoral stance towards patents |
| 23:46 | <sayrer> | half way there |
| 23:46 | <sayrer> | of the, have the |
| 23:46 | <sayrer> | you get it |
| 23:46 | <annevk2> | not really |
| 23:47 | <sayrer> | ok |
| 23:51 | <Hixie> | hm, annotations in rfcs describing which parts aren't actually implemented would be really useful |