| 00:00 | <annevk2> | i thought it was when the status bar or something said "done" |
| 00:01 | <annevk2> | http://twitter.com/flexewebs/statuses/1345419976 followed by http://twitter.com/flexewebs/statuses/1345479096 fail |
| 00:05 | Philip` | hasn't even read the spec once :-( |
| 00:21 | <annevk2> | "In case (3), you definitely need also one of the following (and can use these in other cases):" well I disagree right there... |
| 00:22 | <annevk2> | but then threads with Larry involved tend not to be that constructive so I'll let someone else bite :) |
| 00:27 | <annevk2> | oh, it's on www-tag |
| 00:27 | <annevk2> | ok, i'll bite |
| 00:35 | <Niictar> | Anyone alive? |
| 00:36 | <annevk2> | for a few more minutes |
| 00:36 | <Niictar> | Right on |
| 00:36 | <Niictar> | Not that I have a lot to say. But I was curious if this place was going to end up being super quiet all the time like the forum seems to be |
| 00:37 | <Philip`> | Niictar: See the left column of http://krijnhoetmer.nl/irc-logs/ |
| 00:37 | <Niictar> | Kay |
| 00:37 | <Philip`> | Usually there's quite a lot of activity |
| 00:38 | <Philip`> | Not necessarily productive activity, but things do get said :-) |
| 00:38 | <Niictar> | Haha |
| 00:39 | <annevk2> | the mailing list and IRC are the main forms of communication |
| 00:39 | <annevk2> | the forums are there for people who don't like mailing lists but they do sort of take a back seat |
| 00:40 | <Niictar> | Ok. Well, the mailing list has been filled with nothing but arguments about <time>, and yea, the forums are super slow. |
| 00:40 | <Niictar> | And I like IRC, so that's all good for me, then |
| 00:41 | <annevk2> | oh yeah, you need to learn about deleting email following the list :p |
| 00:42 | <Philip`> | Much of the discussion on IRC about the <time> topic has been incredulity at the discussions on the mailing list |
| 00:42 | <Niictar> | Heh, maybe all the space gmail gives me will be put to good use, now :P |
| 00:42 | <Philip`> | so this is a relatively safe place if you're not terribly interested in the details of the <time> discussion |
| 00:45 | annevk2 | -> bed |
| 00:47 | <Niictar> | So, anyway, I was looking up HTML 5 on Twitter today and it was flooded with news about the iPhone supporting the <video> element |
| 00:47 | <Niictar> | But it was just a lot of people saying so without any references |
| 00:47 | <Niictar> | And Google has nothing on it yet |
| 00:47 | <Niictar> | Anyone know anything about this yet? |
| 00:48 | <olliej> | Niictar: i believe iphone 3 is meant to support it |
| 00:48 | <Niictar> | Err, the 3.0 OS for the iPhone to be exact |
| 00:48 | <olliej> | Niictar: desktop safari has supported <video> and <audio> since 3.1 early last year |
| 00:49 | <Niictar> | I can't make the demo on WHATWG work with my version of Safari 4 for Windows, though |
| 00:49 | <Niictar> | And I read Firefox 3.1 supported it, too, but I can't make that one work either |
| 00:50 | <Niictar> | I was thinking it was cause I was using the wrong video format (which I understand still hasn't been finalized?), but I would have thought that the demo Hixie put up on whatwg.org would work |
| 00:50 | <olliej> | I think there were some changes to the <video> spec that we only fixed in webkit trunk in the last week or so |
| 00:51 | <Niictar> | Ah |
| 00:51 | <olliej> | Niictar: what video format are you using |
| 00:51 | <Hixie> | the demo on whatwg.org uses ogg theora |
| 00:51 | <olliej> | Niictar: the changes are primarily playback events, rather than anything fundamental |
| 00:51 | <Niictar> | Erm, well, the whatwg site uses .ogg. And then I was trying .mp4, too |
| 00:51 | <Hixie> | apple only support h.264 last i heard |
| 00:51 | <olliej> | Hixie: broadly speaking we support anything that quicktime can play i believe |
| 00:51 | <olliej> | Hixie: eg. installing the theora codecs will work |
| 00:52 | <olliej> | Hixie: ironic that you're not using a standard video format :D |
| 00:52 | olliej | hides |
| 00:52 | <Niictar> | Ooh, do I need to install codecs on top of making sure I am using a browser that supports <video>? |
| 00:52 | <olliej> | Niictar: not for mp4 on mac |
| 00:52 | <olliej> | err |
| 00:52 | <olliej> | on safari |
| 00:52 | <Hixie> | olliej: i was trying to demo it in firefox, so using h.264 would have been hard. :-) |
| 00:52 | <olliej> | mac or windows |
| 00:53 | <olliej> | Niictar: ah, what platform are you on? mac or windows? |
| 00:53 | <Niictar> | This one is Safari 4 on Vista. Or anything so far is on Vista. I haven't tried it on Windows 7 or the Macs at work yet |
| 00:53 | <olliej> | Niictar: did you install quicktime with safari? |
| 00:53 | <Niictar> | No |
| 00:53 | <olliej> | Niictar: it's used for video support |
| 00:54 | <olliej> | without it safari isn't going to be playing back anything :D |
| 00:54 | <Niictar> | Ha, I guess that one figures |
| 00:54 | <Niictar> | Ok, well, I will try that later |
| 00:54 | <Niictar> | <video> has a fallback like <canvas> right? |
| 00:54 | <olliej> | Niictar: it would be silly for safari to have its own copy of QT |
| 00:54 | <olliej> | Niictar: yup |
| 00:55 | <Niictar> | Well, I didn't know QT was required for support for video :P |
| 00:55 | <olliej> | Niictar: yeah, it kind of sucks |
| 00:56 | olliej | imagines people would just complain if qt wasn't optional however |
| 00:56 | <olliej> | oh well |
| 00:56 | <olliej> | back to fixing finally |
| 00:56 | <olliej> | i hate finally |
| 00:57 | <Niictar> | So, for the iPhone... could I do something like <video src="video.mov"><embed id="VideoPlayback" src="someFlashFile" style="width:344px;height:285px;" allowFullScreen="true" allowScriptAccess="always" type="application/x-shockwave-flash" /></video> |
| 00:57 | <Niictar> | ? |
| 00:57 | <Niictar> | (in theory) |
| 00:57 | <olliej> | that should do it |
| 00:57 | <olliej> | it should work in safari |
| 00:57 | <olliej> | on the desktop as well |
| 00:57 | <olliej> | ooh |
| 00:57 | <olliej> | actually |
| 00:58 | <olliej> | Niictar: can you see if that falls back correctly in your safari install? |
| 00:58 | <olliej> | Niictar: logically we should fall back if qt isn't available |
| 00:58 | <Niictar> | It doesn't, actually. My page is just blank. ...I think. |
| 00:58 | <olliej> | but i'm not sure |
| 00:58 | <Niictar> | Double checking |
| 00:58 | <olliej> | Niictar: could you file a bug at http://bugs.webkit.org ? |
| 00:59 | <Niictar> | Blank page |
| 00:59 | <Niictar> | Oui |
| 00:59 | <Niictar> | Err, I have to register |
| 00:59 | <Niictar> | Do I have to register? |
| 01:00 | <olliej> | Niictar: yes, because bugzilla emails you when we update the bug |
| 01:00 | <olliej> | Niictar: and we need the bug reporter to verify that we've fixed the bug |
| 01:00 | <olliej> | when we think we've fixed it |
| 01:00 | <Niictar> | Right |
| 01:00 | <Niictar> | Now, a caveat (sp?) |
| 01:00 | <Niictar> | I have some sort of "QT Lite" installed |
| 01:01 | <olliej> | hmmm |
| 01:01 | <olliej> | file anyway |
| 01:01 | <Niictar> | Kay |
| 01:01 | <olliej> | someone will eventually verify the bug exists (or not) and then it will be fixed |
| 01:01 | <olliej> | and the world will be a better place |
| 01:02 | <Niictar> | Fair enough |
| 01:02 | <Niictar> | And now to do all the fun form filling stuff |
| 01:03 | <Niictar> | Priority? |
| 01:03 | <olliej> | Niictar: default will do |
| 01:04 | <Niictar> | Component? |
| 01:04 | <olliej> | webcore |
| 01:04 | <olliej> | ah |
| 01:04 | <Niictar> | Webcore Misc. then |
| 01:04 | <olliej> | sorry |
| 01:04 | <olliej> | yeah |
| 01:04 | <olliej> | that will do |
| 01:04 | <olliej> | actually |
| 01:05 | <olliej> | layout and rendering |
| 01:05 | <olliej> | Niictar: sorry |
| 01:05 | <Niictar> | Makes sense |
| 01:05 | <Niictar> | Do I need to upload the test html file? |
| 01:05 | <Niictar> | ie host it somewhere |
| 01:06 | <olliej> | upload or host if hosting is easier |
| 01:06 | <Niictar> | Ok |
| 01:10 | <Niictar> | It falls back nicely in Firefox |
| 01:11 | <Niictar> | http://html5.ca/video.html if you want to try it now that I have it up |
| 01:11 | <Niictar> | Almost done with the bug report |
| 01:12 | <roc> | Niictar: that doesn't work for me, firefox.ogv is 404? |
| 01:13 | <Niictar> | I pulled it from http://www.whatwg.org/demos/2008-sept/video/video.html oops |
| 01:13 | <roc> | olliej: sorry, but a patent-encumbered pay-to-play "standard" isn't worth much |
| 01:13 | <Niictar> | I don't know of any other particular video file that "should" work |
| 01:13 | <olliej> | roc: vs. a poor quality potentially patent encumbered non-standard? |
| 01:14 | <olliej> | roc: :D |
| 01:14 | <olliej> | roc: i realise a solution is needed |
| 01:14 | <roc> | there's a published spec |
| 01:14 | <roc> | it doesn't have an ISO stamp I guess |
| 01:14 | <roc> | the quality's not bad and getting better |
| 01:14 | <Hixie> | i don't think theora is a "non-standard" |
| 01:14 | <roc> | we've done our legal analysis |
| 01:14 | <olliej> | roc: it also doesn't have hardware acceleration |
| 01:15 | <Niictar> | Any other sources I could grab for that video element? |
| 01:15 | <roc> | HW acceleration on mobile is a problem yeah |
| 01:15 | <Hixie> | but the video people i speak to do think its quality leaves something to be desired |
| 01:15 | <olliej> | roc: the reality is that neither h264 nor theora is a real "solution" |
| 01:16 | <olliej> | roc: which is sucky |
| 01:16 | <roc> | well |
| 01:16 | <roc> | Theora solves a lot of use cases |
| 01:16 | <olliej> | so does h264 |
| 01:16 | <roc> | H.264 doesn't solve anything for those of us who care about open source implementations |
| 01:17 | <olliej> | theora doesn't solve anything for those who want hw acceleration, etc |
| 01:17 | <Hixie> | i agree that 264's licensing makes it a non-option unless that is addressed |
| 01:17 | <roc> | that's only an issue on mobile really |
| 01:17 | jcranmer | looks at in re Bilski |
| 01:17 | <olliej> | roc: neither works for everyone |
| 01:18 | <roc> | and it's only an issue on mobile because the device vendors currently choose to bake in H.264 and nothing else |
| 01:18 | <roc> | give us some programmability and things can change |
| 01:19 | <Philip`> | Hmm, phones which can automatically reprogram their DSPs for new video codecs... |
| 01:20 | <sayrer> | The h264 vs ogg debate is so lame |
| 01:20 | <sayrer> | it's not a "neither works for everyone" situation |
| 01:20 | <sayrer> | that is an amoral argument |
| 01:22 | <Niictar> | Well, for now, I would love to be able to have video display on my site on the iPhone and on the desktop |
| 01:22 | <Niictar> | Regardless of the debate |
| 01:22 | <Niictar> | But I don't know anything about it right now |
| 01:23 | <Niictar> | I don't even know what extention h.264 uses |
| 01:23 | <sayrer> | don't worry, we're busily coding up a quicktime wrapper API |
| 01:24 | <roc> | not any more |
| 01:25 | <Niictar> | Well, time for food. Thanks for the help |
| 01:25 | Niictar | is away |
| 01:25 | Hixie | gets into a twitter conversation with someone who says that <input type=date> will be buggy and therefore we should use XForms |
| 01:25 | <Hixie> | ...because apparently XForms <input ref="x"> <x xsi:type="xsd:date"> will not be buggy...? |
| 01:27 | Niictar | is back to say he has upload a .mp4 file and his Safari 4 installation is giving him a green box. Maybe error correction for bad video format should be implimented? |
| 01:27 | <Niictar> | And actually now I am away again |
| 02:24 | <olliej> | sicking: what do you mean the spec shouldn't mandate view source behaviour? :D |
| 04:52 | <Niictar> | Quiet |
| 05:30 | <Niictar> | Oh man |
| 05:31 | <Niictar> | I just realized one thing |
| 05:31 | <Niictar> | firefox.ogv was referenced as a relative link, hence the 404. Dumb. |
| 05:35 | <Niictar> | But it still doesn't work |
| 05:35 | <Niictar> | Well, at least it's being consistent =( |
| 06:35 | <Hixie> | 100%! |
| 06:41 | <MikeSmith> | Hixie: 100% completed on the author-view edits? |
| 06:43 | <Hixie> | yeah |
| 06:43 | <Hixie> | i'm sure it's shock-full of typos and errors and i look forward to everyone's feedback over the next 5 years :-) |
| 06:43 | <Hixie> | i wonder if i can get away with marking this an "editorial" edit |
| 06:44 | <MikeSmith> | congrats |
| 06:44 | <MikeSmith> | お疲れさま (otsukaresama) |
| 06:44 | <MikeSmith> | Hixie: you define "editorial" as changes that don't effect implementations? |
| 06:44 | <Hixie> | checked in |
| 06:44 | <MikeSmith> | excellent |
| 06:49 | <Hixie> | http://www.w3.org/mid/49C09567.5000207⊙dc |
| 06:50 | <Hixie> | we've been saying for months that this is what will happen when rdf gets popular and is why rdf won't work in the wild |
| 06:55 | <MikeSmith> | Hixie: I find it hard to comment on that without being seen as partisan. |
| 06:56 | <MikeSmith> | but I'm glad that Manu remains involved in the discussions |
| 08:43 | <bell007> | hi |
| 08:45 | <bell007> | hi , p.parse(content).toxml()[19:-14] , why it remove src="http://www.test.com/image.png"; ??? |
| 08:46 | <bell007> | import html5lib from html5lib import sanitizer p = html5lib.HTMLParser(tokenizer=sanitizer.HTMLSanitizer) |
| 08:48 | <Philip`> | bell007: Which version of html5lib are you using? |
| 08:54 | <bell007> | Philip:Google Code SVN Version. I found it remove src="http://www.test.com/image.png"; in toxml() step. |
| 09:11 | <jgraham> | bell007: You have a checkout from today? I think there was a regression like that that got foxed at the weekend |
| 09:11 | <jgraham> | *fixed |
| 09:37 | hsivonen | is really, really annoyed at bugs in Keynote's video export support |
| 09:38 | <jgraham> | When reading "saying somethinf is unspecified [...] is more a warning to programmers" leaves me feeling thoroughly depressed, does that mean I've spent too long drrinking the WHATWG koolaid? |
| 09:39 | <hsivonen> | does anyone happen to know if Keynote ’09 fixes any video export bugs compared to Keynote ’08 |
| 09:41 | <Philip`> | jgraham: I've seen people outside the web standards community who seem to agree that that idea is depressing |
| 09:43 | <Philip`> | e.g. when there are people who think relying on undefined behaviour in C++ is fine because it works in their compiler, and the undefinedness is just a warning to be careful, there are other people who think that's stupid and it's going to bite those people in the end and they really shouldn't disregard the standards |
| 09:44 | <jgraham> | Philip`: So one of the most depressing things is the assumption that people will be aware of what is in the standard in order to be careful around it |
| 09:44 | <Philip`> | so it should be treated as a (uncheckable) error rather than a warning to programmers |
| 09:45 | <Philip`> | though this is probably a quite different interpretation of your quote than how it would be interpreted in the context of web standards |
| 09:46 | <jgraham> | In this case I think the error condition is checkable but the argument is being made that it should be implementation-defined and programmers should have to be careful to save the bother of actually checking it |
| 09:49 | <Philip`> | Perhaps the idea is that people should test their code in two browsers, and if it differs then it's because they're relying on undefined behaviour and that should indicate to them that they need to fix it |
| 09:51 | <jgraham> | I think the idea is that no person should write code like this in the first place. Which seems like an optimistic assumption |
| 09:52 | <gsnedders> | So, my plan at getting up early failed |
| 09:52 | <Philip`> | 10am is pretty early |
| 09:53 | gsnedders | wonders why http://twitter.com/gsnedders and http://search.twitter.com/search?q=from%3Agsnedders are different |
| 10:00 | <mpt> | gsnedders, the search is weird. Yesterday I found that it produced different results based on the order of the search terms. |
| 10:05 | <Philip`> | The search is weirder than that - it even varies depending on the time of day |
| 10:48 | <hsivonen> | I'm amused to see that Philip` has found a bug in Genx |
| 10:51 | <gsnedders> | hsivonen: Is that not inevitable? |
| 10:57 | <Philip`> | It's just a continuation of my quest to prove that it is humanly impossible to write code that generates well-formed XML |
| 10:58 | <annevk2> | It also proves Tim is a bozo |
| 10:58 | <Philip`> | *Everyone* is a bozo :-) |
| 10:58 | <gsnedders> | Philip`: You bozo. |
| 10:58 | <Philip`> | At least, everyone who attempts to write code that generates XML is a bozo |
| 10:59 | <Philip`> | and at some point you have to start wondering whether the problem is with XML rather than with the entire human race |
| 10:59 | <Philip`> | (the most likely problem being that there are a lot of rules, and thus a lot of chances to get things wrong) |
| 11:00 | <Philip`> | (In a more permissive markup language you could still have bugs like forgetting to escape "<" in text, but you wouldn't have to worry about also escaping "]]>" in text) |
| 11:02 | <Philip`> | ((By "more permissive" I just mean one in which fewer strings are errors, not necessarily one in which there is no draconian error handling)) |
| 11:03 | <Lachy> | Philip`, one day, I will prove you wrong by writing a tool that always outputs well formed XML |
| 11:04 | <Philip`> | Well, it's easy to write a tool that always outputs well-formed XML |
| 11:04 | <Lachy> | I don't mean something simple that just does: print "<foo>Hello World!</foo>"; |
| 11:05 | <Philip`> | It needs to accept arbitrary user input, and preferably use that input to determine what tags and namespaces to output (i.e. not just text and attribute values) |
| 11:05 | <Lachy> | of course |
| 11:05 | <Philip`> | Also it's not allowed to simply pass its serialiser output back into an XML parser and print an error message if it wasn't well-formed |
| 11:06 | <Philip`> | (It should either modify the data into something well-formed, or generate error messages at the moment when the invalid document fragment is constructed) |
| 11:06 | <jgraham> | Lachy: If you mean "a tool that can produce any output that conforms to the XML1.0+XMLNS grammar and never produces output that does not conform to said grammar" then it is worth noting that many people before you have trie and failed |
| 11:06 | <Lachy> | no, I will make it so that, e.g., it outputs non-Unicode byte sequences as U+FFFD, and other stuff like that |
| 11:07 | <Philip`> | Also, if I can't find any bugs then I'll just pretend I haven't even bothered looking at it yet, to avoid disproving my argument |
| 11:07 | <Lachy> | jgraham, I know. That's why I'm accepting the challenge |
| 11:08 | <jgraham> | "Good Luck With That" |
| 11:08 | <annevk2> | "Waste of Time" |
| 11:08 | <jgraham> | Yeah, it seems like there are more interesting things to work on |
| 11:09 | <jgraham> | Like making a markup language with less chances of going wrong in the first place |
| 11:09 | <Philip`> | It would only be worthwhile if you could convince a significant number of people to use your library |
| 11:09 | <Lachy> | sure, I don't plan to work on it now. But one day, when I have some time and motivation |
| 11:10 | <jgraham> | Oh I have a lot of projects like that. I plan to get high speed network access in my coffin to finish them all |
| 11:10 | <Philip`> | "I can write a better XML serialiser library than all the XML experts who have tried and failed before me, and I will prove it at some undetermined point in the future which is likely to never occur!" ;-) |
| 11:16 | <hsivonen> | annevk2: re: www-tag, back on the telecon where the versioning ACTION was minted, I said reopening the discussion probably wouldn't be productive... |
| 11:16 | Philip` | notes that the Genx bug is particularly weird, since there's an explicit piece of code added to allow the first character of an NCName to be ':', and that code has no other possible purpose |
| 11:18 | <annevk2> | hsivonen, ah, maybe I shouldn't have said anything at all then |
| 11:19 | <hsivonen> | annevk2: that's not what I meant. I meant I agree with your assesment about the earlier thread. |
| 11:48 | <hsivonen> | The presentation recording feature of Keynote is evil. It would be a great idea if it worked. Now it lures a person to commit to the idea of recording but then its so buggy that fixing the results requires unreasonable effort. |
| 11:48 | gsnedders | wonders if hsivonen has filed a bug |
| 11:49 | <hsivonen> | gsnedders: last time I checked, Apple's bug reporter wouldn't let me log in even after I emailed Apple about it |
| 11:49 | <gsnedders> | Odd. |
| 11:50 | <hsivonen> | gsnedders: also, I have ’08, so presumably I should first buy ’09 and see if the bug persists there |
| 11:51 | <Philip`> | You should use the bug reporter to report the bug that prevents you logging in |
| 11:51 | <Philip`> | That sounds like a good business model - release a product that doesn't work, and then have people buy the next version of the product just to see if you've fixed it yet |
| 11:52 | <hsivonen> | I'm quite annoyed. I'll go some bugs of my own making. |
| 12:00 | <ap> | hsivonen: if everything else fails, there's http://developer.apple.com/bugreporter/bugrptform.html |
| 12:01 | <hsivonen> | ap: thanks |
| 12:01 | <gsnedders> | ap: Mac OS 9 is an option for platform on that form! |
| 12:02 | <ap> | gsnedders: this may or may not be a mistake, I don't know |
| 12:03 | <ap> | gsnedders: e.g. a current product such as AirPort Base Station may be supported with OS 9 clients |
| 12:04 | gsnedders | shrugs |
| 12:23 | <annevk3> | http://blog.amplesdk.com/2009/03/18/getting-ready-for-tomorrow-ample-sdk-to-use-html5-doctype/ but they keep the xmlns cruft. Weird. |
| 13:46 | Lachy | wishes people would trim CC lists in replies more frequently |
| 14:51 | <annevk3> | argh |
| 14:51 | <annevk3> | proofreading.is.needed |
| 14:53 | <jgraham> | annevk3: Well for a start I suggest using spaces rather than dots between your words |
| 14:54 | Philip` | read that sentence and didn't notice the dots, since they were camouflaged with the dust on his monitor |
| 14:56 | jgraham | postulates that people who complain that HTML5 is taking too long have never wanted to use the STIX fonts |
| 14:59 | <jgraham> | Hmm it took 3 messages for the "web addresses... thread to turn into a bikeshed about the name. I wonder what the record is |
| 15:00 | <Philip`> | jgraham: I guess they haven't been waiting for the final volume of TAOCP either |
| 15:06 | Philip` | wonders what tricks there are for making a LaTeX document shorter without cutting out any of the content |
| 15:11 | <jgraham> | Philip`: Larger margins |
| 15:11 | <jgraham> | Smaller fonts |
| 15:11 | <jgraham> | Smaller figures |
| 15:12 | <iugrina> | Philip`: baselinestretch |
| 15:19 | <Philip`> | I don't want to change margins or font size, since there's specific formatting requirements and a provided .cls file for it |
| 15:20 | <Philip`> | I'm already using Times which saves half a page compared to Computer Modern |
| 15:23 | Philip` | decides to delay that boring formatting cleanup stuff until later |
| 15:55 | <jgraham> | Web browsers suck |
| 15:56 | <annevk3> | now there's an axiom I haven't seen before |
| 15:57 | <jgraham> | It's not really an axiom, I have evidence. |
| 15:59 | <Philip`> | Axioms aren't things without evidence, they're just things without proof |
| 16:00 | <Philip`> | I could take a piece of paper, draw two points on it, then draw a straight line through those points, and it would be evidence supporting Euclid's first axiom |
| 16:01 | <jgraham> | Philip`: That seems subtly wrong. Axioms are principles on which you use logic to build conclusions. But my statement was a conclusion, not an axiom that went into forming that conclusion |
| 16:02 | <jgraham> | So I guess I was wrong to say "it's not an axiom, I have evidence" |
| 16:03 | <jgraham> | It should be more like "it's not an axiom it's the combination of an axiom that can be used to define suckiness plus deductions" or something |
| 16:03 | <Philip`> | I guess they're more precisely *irreducible* principles on which you use logic to build conclusions |
| 16:05 | <Philip`> | I suppose you could hypothesise a universe in which web browsers axiomatically do not suck, but the universe will probably not be consistent |
| 16:29 | Philip` | successfully shortens his document by about minus six lines |
| 16:33 | <takkaria> | Philip`: simply consistent or absolutely consistent? :) |
| 16:36 | <Philip`> | takkaria: Yes |
| 16:36 | Philip` | doesn't know the difference, and should probably be working on stuff other than finding out the difference :-p |
| 16:38 | Philip` | hatches a cunning plan |
| 16:38 | <Philip`> | If I remove some citations from my document, it saves a few characters on some lines plus it saves whole lines in the References section at the end |
| 16:39 | Philip` | can't see any downsides |
| 16:44 | <jgraham> | Plus your work looks more original |
| 16:44 | <jgraham> | On the downside you annoy someone who you might want to work for in the future |
| 16:46 | <Philip`> | Good points :-) |
| 16:47 | <Philip`> | (Fortunately these are only actually links to web sites for Cisco and Quagga, and people can type those terms into Google if they really want to find them) |
| 19:51 | <Hixie> | ok |
| 19:51 | <Hixie> | database section is now out of html5 |
| 19:51 | <Hixie> | along with websockets and eventsource |
| 19:52 | <Philip`> | Are they in anywhere else yet? |
| 19:52 | <Hixie> | http://dev.w3.org/html5/webstorage/ |
| 19:52 | <Hixie> | http://dev.w3.org/html5/websockets/ |
| 19:52 | <Hixie> | http://dev.w3.org/html5/eventsource/ |
| 19:53 | <hsivonen> | Hixie: are there still whatwg-lisensed versions? |
| 19:53 | <Hixie> | no |
| 19:53 | <Hixie> | at least, not post-processed versions |
| 19:53 | <Hixie> | the source is still part of the html5 spec's "source" document |
| 19:54 | <annevk3> | yay: http://www.w3.org/TR/cors/ |
| 19:54 | <gsnedders> | Oh dear… |
| 19:54 | <gsnedders> | That just seems horrific |
| 19:54 | <gsnedders> | The source document doesn't even correlate to the output now? Yuk. |
| 19:54 | <Hixie> | the source document is run through a bunch of scripts to generate the output documents |
| 19:55 | <Hixie> | from that one file i now generate one whatwg spec, four w3c specs, and an rfc |
| 19:55 | <hsivonen> | Hixie: does this mean that Web Apps rechartered? |
| 19:56 | <Hixie> | no idea |
| 19:56 | <gsnedders> | http://diveintomark.org/archives/2009/03/18/if-it-fails-for-some |
| 19:57 | <hsivonen> | Hixie: great to see that the split finally happened |
| 19:58 | annevk3 | has mixed feelings about the split |
| 19:59 | <annevk3> | I see two potential issues with having them separate: 1) license 2) maintenance |
| 19:59 | <Hixie> | if anyone comes around asking where the specs they were implementing have gone, please point them to the dev.w3.org pages :-) |
| 20:00 | <Hixie> | maintenance shouldn't be an issue since they're all autogenerated from one source document |
| 20:00 | <annevk3> | but they no longer need to be LC quality by the end of this year |
| 20:01 | <hsivonen> | gsnedders: I wonder what the twitter response to Troy's post will be like |
| 20:01 | <gsnedders> | hsivonen: Troy's post? |
| 20:01 | <Hixie> | annevk3: true, true |
| 20:01 | gsnedders | expects he's missing something |
| 20:02 | <hsivonen> | gsnedders: see http://blog.whatwg.org/omit-alt#comment-7771 |
| 20:02 | <gsnedders> | Ah |
| 20:02 | <Hixie> | annevk3: i guess you'll just have to make sure i get to them anyway :-) |
| 20:02 | <annevk3> | Hixie, can I book a few weeks in September? :p |
| 20:03 | <gsnedders> | :P |
| 20:04 | <Hixie> | annevk3: i recommend not waiting until then :-) |
| 20:19 | <annevk3> | http://annevankesteren.nl/2009/02/rel-canonical#comment-6736 SEO spam or sort of useful? |
| 20:19 | <annevk3> | I suppose I could just delete his comment for calling a link relation a "tag" |
| 20:20 | <Philip`> | It seems worth keeping the comment if and only if it provides some useful information or insight to the discussion |
| 20:21 | <Philip`> | which is probably an easier judgement than trying to work out whether a comment was intended as spam or not |
| 20:21 | <gsnedders> | Hixie: Are you so for micromanaging a bibliography you couldn't even use some programmatic thing as a stop-gap solution? |
| 20:22 | <annevk3> | gone |
| 20:28 | <jwalden> | Hixie: fyi, html5 has a few refs to the database API still, search for "database" to find them |
| 20:29 | <annevk3> | the one on Window probably being the most problematic |
| 20:43 | <annevk3> | same goes for localStorage and sessionStorage fwiw |
| 20:47 | <annevk3> | http://twitter.com/mibbit_svn/statuses/1350511417 |
| 20:47 | <annevk3> | https://twitter.com/Hixie/status/1350291221 |
| 21:00 | <roc> | where are Web Sockets, Event Source and Web Storage going to live? |
| 21:02 | <annevk3> | WebApps WG, likely |
| 21:03 | <annevk3> | after shepazu and MikeSmith do some charter-fu |
| 21:13 | gsnedders | stretches, and goes back to reading Lolita (oh bad immoral me) |
| 21:24 | <Hixie> | gsnedders: dunno |
| 21:24 | <Hixie> | jwalden: yeah, i've marked some of those cases with "XXX"s |
| 21:25 | <Hixie> | roc: if you mean wg-wise, webapps is where i was told they should be |
| 21:30 | Hixie | peers into his mail archives to see what he should reply to |
| 21:30 | <Hixie> | good god |
| 21:30 | <Hixie> | 142 e-mails on <time> |
| 21:35 | jgraham | mocks Hixie for agreeing to reply to all that email |
| 21:38 | <Hixie> | i do not think agreeing to that was a mistake |
| 21:38 | <Hixie> | i also do not think that not agreeing to reply to all the public-html mail was a mistake |
| 21:39 | <Hixie> | the key difference being that the latter list has had long periods of unmitigated community breakdown |
| 21:39 | <Hixie> | which leads to e-mails that are not worth the electrons they are written on |
| 21:40 | <gsnedders> | Hixie has obviously become bored with stuff at a byte level |
| 21:43 | <jgraham> | Hixie: I was joking :) |
| 21:44 | <Hixie> | jgraham: :-) |
| 21:44 | <jgraham> | Although I am glad that I do not have to reply to all that email |
| 21:46 | <gsnedders> | "You got 21 out of 25 correct. (That's 84%.)" — not bad for a quiz on a book that I haven't read in a year, and a quiz that asks quite obscure questions |
| 21:48 | <annevk3> | http://lists.w3.org/Archives/Public/public-css-testsuite/2009Mar/0045.html I guess I should keep an open mind... |
| 22:04 | <roc> | Why do people think it makes any sense to have thousands of test cases with no automation? |
| 22:05 | <roc> | do they actually have people sitting in cubicles running through these tests all day? |
| 22:05 | <gsnedders> | Yes |
| 22:05 | <Hixie> | i'd imagine they do the same as opera, and have a system that takes screenshots |
| 22:05 | <roc> | I refuse to believe you |
| 22:05 | <gsnedders> | I refuse to believe myself. |
| 22:06 | <gsnedders> | Trusting someone who is reading a novel where a central concern is the unreliability of the narration would be very naïve |
| 22:08 | <roc> | we solved this problem with reftests |
| 22:08 | <roc> | I suppose someone can create a reference page for each of these tests |
| 22:08 | <annevk3> | should be quite simple for most of these tests |
| 22:11 | <Philip`> | You couldn't possibly automate all the tests - it would cut out thousands of man-hours of work and destroy the job security of many QA engineers, which is unacceptable in the current economic climate |
| 22:15 | <roc> | don't worry, we can employ them writing more automated test |
| 22:16 | <Hixie> | roc: not most of them |
| 22:21 | <virtuelv> | Hixie: thanks for splitting out Storage |
| 22:21 | <Hixie> | np |
| 22:21 | <Hixie> | is there anything else that can be split out easily that i missed? |
| 22:29 | <annevk3> | Hixie, cross-document messaging I suppose |
| 22:29 | <Hixie> | that relies on Window too much |
| 22:29 | <Hixie> | MessageChannel maybe |
| 22:30 | <Hixie> | but that's tiny |
| 22:31 | <Philip`> | You could split out all the HTML from the spec |
| 22:38 | <Hixie> | hsivonen: yt? |
| 22:39 | <virtuelv> | Hixie: Window object? |
| 22:40 | <virtuelv> | (Ok, I realise that might be going far) |
| 22:41 | <Hixie> | "easily" was a key word |
| 22:41 | <annevk3> | I think it would be nice if Window / page loading / etc. was separate though |
| 22:42 | <virtuelv> | Hixie: how bound is text selection to HTML? |
| 22:43 | <Hixie> | it's bound to editing, which is bound to contentEditable, which is bound to HTML |
| 22:45 | <virtuelv> | but no, other than Window, I don't think there is much left to rip, and still call the spec "HTML" |
| 22:45 | <Hixie> | k |
| 22:46 | <virtuelv> | (And I agree with anne on the «nice» |
| 23:08 | Niictar | is still trying to wrap his mind around the <video> element |
| 23:09 | <Niictar> | I can see there are problems when an author tries to impliment his/her own fallback content |
| 23:10 | <Niictar> | But even with that, I still don't see why browsers shouldn't indicate they are having a problem rendering a video |
| 23:25 | <olliej> | There are some cool canvas demos at http://www.chromeexperiments.com -- despite warnings about only working in chrome they work fine in any recent webkit based browser (eg. Safari, Epiphany, ...) |
| 23:34 | <Niictar> | The first two work fine in Firefox so far =( |
| 23:35 | <annevk3> | http://lists.w3.org/Archives/Public/public-forms/2009Mar/att-0051/2009-03-18.html#topic4 |
| 23:46 | <Niictar> | olliej, do you follow the mail list? |
| 23:47 | <olliej> | Niictar: what mailing list? |
| 23:48 | <Niictar> | Sorry, the one for feedback on the HTML 5 specs. whatwg⊙wo to be exact |
| 23:48 | <olliej> | yeah |
| 23:48 | <olliej> | but i'm not one of the webkit video people |
| 23:48 | <olliej> | i do the canvas implementation that chrome takes credit for |
| 23:50 | <Niictar> | Ah, ok |
| 23:50 | <Philip`> | How much of their canvas implementation is specific to Skia? |
| 23:51 | <olliej> | Philip`: err, nothing? |
| 23:52 | <olliej> | Philip`: krit and I did a lot of work to make it our canvas implementation just work in terms of the standard webcore graphiccontext object |
| 23:52 | <olliej> | s/it // |
| 23:52 | <Niictar> | So your comment about "logically we should fall back if qt isn't available" is unfortunately not coming from an area of expertise about the <video> implementation |
| 23:53 | <olliej> | Niictar: no it is -- i believe in the absence of QT we should be acting like we don't support video or audio |
| 23:53 | <olliej> | so the video/audio tags sould behave as they would in any other browser than doesn't support them |
| 23:54 | <olliej> | eg. be treated as ordinary tags with no special properties |
| 23:54 | Philip` | gets confused until realising the discussion is about QT, not Qt |
| 23:55 | <Niictar> | Should I bother to continue my arguments for something like that on the mailing list, though? |
| 23:55 | Niictar | is unsure of the etiquet(sp?) expected |
| 23:56 | <Niictar> | etiquette* |
| 23:58 | <olliej> | Niictar: i think breaking video when QT isn't present is a webkit bug |
| 23:58 | <olliej> | Niictar: absence of fallback rules in <Video> is something i'm not at all sure about |
![http://www.test.com/image.png"](http://www.test.com/image.png"){kind=link}