| 06:12 | <jacobolus> | if I want to make a cursor that changes as I pan around the canvas, the best way to do that is by making my own collision detection on mousemove and then adding changing the "cursor" css attribute by adding/removing classes? |
| 06:16 | <roc> | probably |
| 08:01 | <sicking> | annevk2: ping |
| 08:01 | <annevk2> | yeah? |
| 08:02 | <sicking> | annevk2: so apparenly the DOM-Core spec isn't compatible with ACID3 :( |
| 08:02 | <annevk2> | well yeah, we knew that |
| 08:02 | <sicking> | annevk2: specifically some of the doctype changes makes you loose a point if you implement them |
| 08:02 | <annevk2> | yes we said that before we made the changes |
| 08:03 | <sicking> | annevk2: have you talked to the people at opera that would implement this how they plan to deal with it |
| 08:03 | <sicking> | annevk2: i poked Hixie and Håkon, but got no answer |
| 08:04 | <annevk2> | I think Ms2ger made the relevant changes to DOCTYPE |
| 08:04 | <sicking> | in the past Hixie hasn't wanted to change ACID3 without an ok from Håkon (don't know the history there) |
| 08:05 | <annevk2> | I have once tried to change Acid3 and I believe Håkon wants some kind of explicit endorsement from all vendors and only wants to do it once or so |
| 08:05 | <sicking> | annevk2: ok, i'll mail the list i guess |
| 08:05 | <annevk2> | It's pretty crazy |
| 08:05 | <annevk2> | I think it is easier to just ignore Acid3 |
| 08:05 | <sicking> | annevk2: hah, yeah, good luck |
| 08:05 | <sicking> | annevk2: especially for the implementations that have 100 points right now |
| 08:06 | <annevk2> | I was planning on blogging on it at some point |
| 08:06 | <annevk2> | Not sure if that helps, but sometimes it does |
| 08:10 | <sicking> | annevk2: i suspect you won't see Opera or webkit implementing DOM4 as things stand. So blogging can only help |
| 08:29 | <zcorpan> | clearly the test needs to change if we all agree to change the spec and want to implement the spec |
| 08:30 | <zcorpan> | and if we agree on what the spec should say, it shouldn't be hard to get endorsement from the relevant vendors, right? |
| 08:33 | <jgraham> | Yeah, this seems like it ought to be a non-problem |
| 08:34 | <jgraham> | I mean it is purely political |
| 08:34 | <jgraham> | Which, I suppose, isn't quite the same thing |
| 08:36 | <jgraham> | I presume the approach would be to make either behaviour allowed. In which case it shouldn't matter how many changes we make to ACID3 because no one will ever lose points |
| 08:36 | <annevk2> | howcome wanted full agreement, only once-per-decade, and the original test had to be available |
| 08:37 | <annevk2> | I don't think that is a good idea personally |
| 08:37 | <jgraham> | What is a good idea? |
| 08:37 | <annevk2> | once-per-decade |
| 08:37 | <jgraham> | Right, that's silly |
| 08:37 | <jgraham> | Håkon might be overestimating how useful ACID3 is or how much people care |
| 08:38 | <annevk2> | I think we should simply comment the tests out, fwiw |
| 08:38 | <jgraham> | That would fit "either behaviour is allowed" |
| 08:39 | <jgraham> | (the thing about the original test being avaliable has already failed, so that seems like a bogus requirement too) |
| 08:51 | <Ms2ger> | Acid3, the curse that keeps on giving? |
| 09:07 | <annevk> | Sort of like Acid2 |
| 09:07 | <annevk> | Fortunately we got comment handling changed in the end |
| 09:09 | <annevk> | And here I thought I would never RT jreschke |
| 09:12 | <hsivonen> | why is this now blocking on howcome? IIRC, SVG font thing blocked on someone else at Opera instead |
| 09:13 | <annevk> | that falls under there being no vendor consensus |
| 09:13 | <jgraham> | hsivonen: Ask Hixie. If it's a conspiracy, no one told me :) |
| 09:15 | <hsivonen> | every main stream tech media browser comparison that I've read in the past 18 months has used Acid3 cluelessly |
| 09:15 | <hsivonen> | also html5test.com cluelessly |
| 09:16 | <hsivonen> | but then I haven't read that many comparisons. the correct number is probably just 2 |
| 09:16 | Ms2ger | is reminded of http://xkcd.com/528/ |
| 09:18 | <zcorpan> | Ms2ger: you mean acid4 should be a picture of hitler? |
| 09:19 | <Ms2ger> | Yep |
| 09:30 | <hsivonen> | I recently read a browser comparison that said Safari uses the WebKit 2 engine but Chrome uses the WebKit engine |
| 09:30 | <hsivonen> | boom. in your face Chrome. Safari is twice as advanced |
| 09:33 | <woef> | Is there a place where I can read up on the history of the article element. As in, "why was it introduced in the first place"? |
| 09:35 | <abarth> | hsivonen: i tried to add WebKit3, but my patch was rejected |
| 09:43 | <jgraham> | Dear lazy irc: My virtualbox won't do the fit resolutionto window / mouse integration thing, even though I seemed to have installed the guest additions. Ideas? |
| 09:59 | <annevk> | oh fuck |
| 09:59 | <annevk> | Acid3 prevents Range simplification too |
| 09:59 | <divya> | jgraham: w.r.t resolution you can change it in the VM itself |
| 10:00 | <annevk> | in particular exception handling |
| 10:27 | <annevk> | I wrote http://annevankesteren.nl/2011/09/acid3-problem |
| 10:27 | <annevk> | Somewhat hurriedly, improvements welcome |
| 10:28 | <annevk> | Will try to write a WHATWG Weekly as well somewhere today |
| 10:42 | <zcorpan> | annevk: at least i argued with Hixie when he wrote the test to make a number of things optional to make room for spec changes, like Attr support |
| 10:43 | <zcorpan> | but clearly we need more room now |
| 10:50 | <zcorpan> | AryehGregor: maybe you can use SDF or the parser test format for serializing DOM |
| 11:32 | <smaug____> | anyone with IE9? |
| 12:06 | <krijnh> | People should really stop ironing and making tea and coffee at the same time |
| 12:09 | <divya> | is that a rhetorical statement krijn |
| 12:09 | <divya> | oops |
| 12:10 | <krijn> | It makes fuses go booboo |
| 12:10 | <krijn> | Totes booboo even |
| 13:50 | <annevk> | The Security WG of the W3C is named WAS WG |
| 14:17 | <wilhelm> | TabAtkins: ping |
| 14:19 | <annevk> | Created http://wiki.whatwg.org/wiki/Acid3 for suggestions on subtests that need changing |
| 14:19 | <annevk> | Feedback and improvements welcome |
| 14:20 | <zcorpan> | AryehGregor: the createLink command doesn't seem to say to create a new "a" element if the selection has no links |
| 14:23 | <zcorpan> | AryehGregor: also, if your selection *contains* a link or several links, i think we just change the href of all the links without messing with them further |
| 14:24 | <zcorpan> | AryehGregor: so for [foo<a href=x>x</a>bar<a href=y>y</a>baz], when you do execCommand('createlink','','z'); you should get [foo<a href=z>x</a>bar<a href=z>y</a>baz] |
| 14:58 | <annevk> | Ms2ger, parsing and serialization does not extend Document |
| 14:58 | <annevk> | Ms2ger, it should |
| 15:14 | <annevk> | http://blog.whatwg.org/weekly-parsing-apis |
| 16:41 | <dglazkov> | good morning, Whatwg! |
| 16:55 | <annevk> | hmm |
| 16:55 | <annevk> | it seems somewhat intentional that DOM Parsing does not support document.innerHTML |
| 17:05 | <annevk> | I don't even have commit access to DOM Parsing |
| 18:11 | <Hixie> | sicking: sorry, i was on vacation, still dealing with backlog of mail |
| 18:11 | <Hixie> | i expect i can convince howcome that it makes sense to release an update for acid3 that includes a bunch of changes to take into account intents |
| 18:12 | <Hixie> | i'd rather like to avoid making regular updates so it would be good to have a comprehensive list of what needs changing |
| 18:12 | <sicking> | Hixie: that would be awesome |
| 18:13 | <sicking> | Hixie: i'll see what i can do, but i won't have time to read all of acid3 |
| 18:15 | <Hixie> | i think anne said he had a wiki page |
| 18:23 | <sicking> | Hixie: cool |
| 18:23 | <zewt> | http://www.google.com/support/forum/p/gmail/thread?tid=10f874f83050af98&hl=en ugh |
| 18:29 | <Philip`> | Perhaps they could add a feature where the recipient of an email can vote +1/-1 on its text style |
| 18:29 | <Philip`> | If too many people vote -1, the sender's text style gets reverted to the default and they are forbidden from changing it back to exactly that style again |
| 18:29 | <zewt> | voting "it's the reader who decides what his email looks like, not the sender" |
| 19:15 | <TabAtkins> | wilhelm: pong |
| 19:53 | <hsivonen> | so first it was "boo. HTML5 is too large. remove stuff!" now it's "boo stuff is getting split out of HTML5 the wrong way" |
| 19:53 | <TabAtkins> | Don't look for consistency in Shelley's bugs. |
| 19:53 | <TabAtkins> | That way lies madness. |
| 20:37 | <zcorpan> | hmm. i thought i unsubscribed from public-html-announce. why am i still getting emails? |
| 20:37 | <zcorpan> | MikeSmith? |
| 20:46 | <jgraham> | TabAtkins: s/for consistency in/at/ |
| 20:46 | <TabAtkins> | I'll accept that patch. |
| 21:02 | <zcorpan> | http://www.w3.org/Bugs/Public/show_bug.cgi?id=14086 - seems like we should add async parsing APIs |
| 21:13 | <zcorpan> | we could add something like void parseAsync(DOMString str, DOMString contentType, Function successCallback, optional Function errorCallback); |
| 21:13 | <zcorpan> | to DOMParser |
| 21:13 | <zcorpan> | and maybe void serializeAsync(Node root, Function successCallback, optional Function errorCallback); to XMLSerializer if we want to have async serializing |
| 21:14 | <zcorpan> | or we should have completely new APIs for the async cases where you can choose whether you want to parse and serialize as xml or html |
| 21:33 | <Philip`> | How about asynchronous JSON parsing? |
| 21:33 | Philip` | wonders if more people would want that than XML/HTML |
| 21:35 | <jgraham> | You can do that with a worker ofc |
| 21:35 | <jgraham> | Although |
| 21:35 | <jgraham> | Since the result might get serialized as json when you postmessage it it isn't obviously a win |
| 21:36 | <jgraham> | Unless the browser optimises in a clever way |
| 21:36 | <jgraham> | (depends what the use case is, I suppose) |
| 21:38 | <Philip`> | Asynchronous doesn't have to mean threaded - when the browser's receiving the serialised message, it could spend 10ms deserialising into the main thread and then do some other work and then deserialise a bit more, and only report there was a message after it's finished slowly deserialising the content |
| 21:38 | <Philip`> | (Receiving the serialised message from the worker, that is) |
| 21:38 | <jgraham> | true |
| 21:39 | <Philip`> | so that shouldn't have to be a blocking operation |
| 21:39 | <Philip`> | and doesn't sound like it needs a particularly clever optimisation |
| 21:40 | <jgraham> | But if you go JSON -> datastructure -> JSON -> datastructure on something where parsing/serializing is slow enough to matter, doing three times the work doesn't seem that appealing |
| 21:42 | <zcorpan> | yeah async json also seems reasonable |
| 21:42 | <smaug____> | Hixie: any chance you could remove assertEquals(doctype.ownerDocument, null, "doctype's ownerDocument was wrong after creation"); before modifying acid3 some more? |
| 21:43 | <smaug____> | (that doctype thing just happens to block some other work I'm doing) |
| 21:51 | <zcorpan> | hmm, an async parser API (whether html, xml or json) could also be streaming |
| 21:52 | <zcorpan> | i guess serializer can also be streaming |
| 21:54 | <Hixie> | smaug____: yeah, gonna do that probably early next week |
| 21:57 | <Hixie> | is http://www.w3.org/Bugs/Public/show_bug.cgi?id=13355 the first escalation for a bug resolved by an "editor's assistant"? |
| 22:02 | <smaug____> | Hixie: ok, thanks! |
| 22:35 | zcorpan | sees Hixie has tried almost all games on G+ |
| 22:35 | <Hixie> | i have? |
| 22:35 | <zcorpan> | Hixie: any recommendations? |
| 22:36 | <Hixie> | i recommend portal 2 |
| 22:36 | <Hixie> | or if you have someone to play with, Worms (also available on Steam) |
| 22:36 | <Hixie> | Bad Company 2 is also pretty good though I haven't played for a while and I hear a sequel is gonna be out soon |
| 22:38 | <zcorpan> | they aren't on G+ :) |
| 22:38 | <Hixie> | i stand by my recommendations :-P |
| 22:39 | <zcorpan> | ok :) |
| 22:39 | <wilhelm> | Hixie: BF3 looks fantastic. |
| 22:40 | <Hixie> | so long as you can blow houses to pieces with explosives, i'm game |
| 22:46 | <wilhelm> | Much more so than in the previous games. |
| 22:46 | <Hixie> | jgraham: even with making the spec smaller, i'm still getting 504s pretty regularly now |
| 22:46 | <Hixie> | jgraham: i'd ray about 50% of the time at least |
| 22:46 | <Hixie> | wilhelm: really! |
| 22:46 | <Hixie> | wilhelm: eeenteresting |
| 22:51 | <wilhelm> | Hixie: http://www.youtube.com/watch?v=uQmAdukloJQ <- C4 |
| 22:52 | <Hixie> | wilhelm: that doesn't really show what happens to the building, but it does look cool |
| 22:57 | <Hixie> | jgraham: uh, ÿ used to turn into ÿ but now turns into &#x00FF; |
| 22:59 | <Hixie> | how does that even happen |
| 23:00 | <zcorpan> | lxml b0rkedness? |
| 23:00 | <Hixie> | why would it change? |
| 23:01 | <zcorpan> | dunno. tcp packet boundary happened to be in the middle maybe? |
| 23:01 | <Hixie> | that would be precious |
| 23:01 | <timeless> | heh |
| 23:02 | <Hixie> | wow i think you might be right |
| 23:02 | <Hixie> | that's fantastic |
| 23:02 | <Hixie> | i added a comment just before it and it fixed it |
| 23:03 | <zcorpan> | heh |
| 23:04 | <zcorpan> | hmm, past bedtime for me |
| 23:05 | zcorpan | waves |
| 23:06 | <Hixie> | later |