| 00:50 | <Hixie> | so does anyone else think that opening registration for TLDs is a spectacularily bad idea? |
| 00:51 | jcranmer | raises his hand |
| 00:51 | <Hixie> | i hope nobody registers a TLD with the same name as any of the machines on my local network |
| 00:51 | <Lachy> | of course it's a terrible idea |
| 00:51 | <jcranmer> | to me, that invites a whole lot of opportunity to fishing |
| 00:51 | <jcranmer> | s/f/ph/ |
| 00:51 | <jcranmer> | logon to www.amazon and enter your credentials! |
| 00:52 | Lachy | goes to register the .paypal TLD... |
| 00:54 | <Lachy> | seriously, though, there are already so many problems with the way the DNS system is managed, it's just one more crazy thing to add to the list. |
| 00:54 | <Hixie> | i don't think i've yet met one person who thinks this is a good idea |
| 00:54 | <Hixie> | yet as far as i can tell it's a done deal |
| 00:54 | <jacobolus> | Hixie: who is planning to open registration for TLD's? |
| 00:54 | <Hixie> | icann |
| 00:54 | <Hixie> | they're not "planning" per se |
| 00:54 | <Hixie> | they've done it, as i understand it |
| 00:54 | <Hixie> | could be wrong |
| 00:55 | <Hixie> | they still have to work out the details, but other than that, they're committed: http://icann.net/en/announcements/announcement-4-26jun08-en.htm |
| 00:57 | <Lachy> | at least there is some sort of approval process for each application |
| 00:57 | <Hixie> | that'll last all of a day, i bet |
| 00:58 | <Hixie> | because when they get a million applications, they'll just approve everything |
| 00:58 | <Lachy> | but I can just imagine domain name squatters registerring valuable TLDs, and then selling them off for huge profits later |
| 01:00 | <jcranmer> | Hixie: the solution is simple |
| 01:00 | <Lachy> | it would be better if they did what Australia did with domain names and forbid them to be resold. |
| 01:01 | <jcranmer> | just define an http: or https: URI to have any a select few TLDs |
| 01:03 | <Hixie> | jcranmer: hrh |
| 01:03 | <Hixie> | heh even |
| 01:04 | <jcranmer> | "A URI's TLD must consist of either a ccTLD or the following TLDs: .com, .net, ..." |
| 01:05 | <Hixie> | ah well |
| 01:05 | <Hixie> | i guess we'll pool some cash and get .whatwg |
| 01:05 | <Hixie> | then we can have http://html5.whatwg/ |
| 01:05 | <Hixie> | and so on |
| 01:07 | <roc> | another bloody gold rush |
| 01:07 | <Philip`> | http://html.5/ |
| 01:07 | <Hixie> | heh |
| 01:07 | <hober> | what.wg |
| 01:07 | <roc> | will they allow IDN in the TLD? |
| 01:07 | <jcranmer> | roc: apparently |
| 01:08 | <jcranmer> | roc: see http://idn.icann.org/#Things_to_test |
| 01:08 | <roc> | God help us all |
| 01:08 | <Philip`> | It's strange that there's an infinite number of possible domain names, and now they're going to allow infinitely more, yet still individual ones are very valuable |
| 01:08 | <roc> | well technically they're not changing the number of available domains |
| 01:10 | <Philip`> | Would this affect http://publicsuffix.org/ much? |
| 01:10 | <jcranmer> | Philip`: you like typing thi5v3ry10ngd0m4inn4m3c0n5i5ting0f4m3d31y0f41ph4num3ric53xi5ting50lelyb3c4u3eth30th3r0n3w45t4k3n? |
| 01:10 | <jcranmer> | .com |
| 01:10 | <roc> | I hope they're really really careful about homographs in TLDs or our whole IDN whitelisting scheme based on TLD is going to sink |
| 01:11 | <jcranmer> | I hope they'll change their mind in a year and revoke all new TLDs |
| 01:12 | <Philip`> | jcranmer: I suppose that's not great, so I'd shorten it to something like http://91.195.35.8 |
| 01:12 | jcranmer | wonders how many sites have a domain name longer than their IP |
| 01:12 | <Philip`> | IPv6 helps solve that problem by swapping the balance |
| 01:18 | <Hixie> | 61%. |
| 01:18 | <Hixie> | getting there. |
| 01:19 | <jcranmer> | 61% and rising or falling? |
| 01:36 | <Hixie> | rising |
| 01:36 | <Hixie> | and the last 30% are trivial (the parser part doesn't have URLs in it) |
| 01:37 | <Lachy> | http://tech.slashdot.org/article.pl?sid=08/06/26/1814205 the comments have some good suggestions |
| 01:39 | <Lachy> | "ICANN has .cheezburger?" :-) |
| 01:41 | <roc> | mmmm |
| 01:41 | <roc> | http://solitaire.exe |
| 01:41 | <roc> | heck |
| 01:41 | <roc> | http://index.html |
| 01:47 | <Lachy> | .mov would be good for films, instead of [xxx]themovie.com commonly used these days |
| 01:47 | <Lachy> | or maybe .film |
| 01:47 | <Hixie> | i'm registering .xn--74h |
| 01:47 | <Lachy> | :-) |
| 01:47 | <Hixie> | .movie or .film is a tld that would actually be useful, i'd support icann adding that |
| 01:48 | <Hixie> | i think that every time i see a movie |
| 01:48 | <Hixie> | speaking of which, wall-e tomorrow!!!!!! weeeee!!!! |
| 01:48 | <Lachy> | heh, me too |
| 01:48 | <Lachy> | oh, cool |
| 01:48 | <Hixie> | gonna have to update the spec! |
| 01:49 | <Hixie> | the <li> example will be out of date :-) |
| 01:49 | <Lachy> | the <li> example is already wrong :-) |
| 01:52 | Lachy | is watching the Wall-e trailer in HD on 24" iMac. Damn, it looks good! |
| 01:52 | <Hixie> | no spoilers! |
| 01:53 | <Hixie> | i've actually managed to avoid seeing the second trailer this time |
| 01:53 | <Hixie> | (pixar always release two trailers, one a year in advance, which uses the same art but has nothing to do with the movie, and one nearer release time, which actually does have some bearing on the movie) |
| 01:55 | <Lachy> | hah! You have to see this one! |
| 01:55 | <Hixie> | i'll see it after :-) |
| 01:55 | <Hixie> | <-- fan boy |
| 01:56 | <Lachy> | http://www.apple.com/trailers/disney/walle/hd/ the one called Vignette is funny |
| 01:58 | <Hixie> | is that the one from when they release ratatouille? |
| 01:58 | <Hixie> | released |
| 01:58 | <Hixie> | i.e. the first one |
| 01:58 | <Lachy> | no |
| 01:58 | <Hixie> | ah |
| 01:58 | <Lachy> | it involves a vaccuum |
| 02:00 | <Hixie> | lalala can't hear you! |
| 02:01 | <Lachy> | don't worry, I wont spoil it for you |
| 02:02 | <Hixie> | :-) |
| 02:06 | <Philip`> | Of course you can't hear him, it's a vacuum |
| 02:06 | <Lachy> | ah, crap. I just realised, since it's a children's film and I'm in Norway, they will have dubbed it in Norwegian. |
| 02:07 | <Lachy> | but it doesn't look like it's being released here yet anyway |
| 02:08 | <Lachy> | I have to get out of this silly backwards country :-( |
| 02:08 | <Hixie> | children's film! |
| 02:09 | <Hixie> | blasphemy! |
| 02:09 | <Lachy> | sorry, family film |
| 02:10 | <Hixie> | the term is "comedy" |
| 02:10 | <Lachy> | suitable for children, but no doubt entertaining for adults. |
| 02:12 | <jcranmer> | you'd think that Hixie is employed by Disney or Pixar or whomever... |
| 02:13 | <jcranmer> | oh, I'm sorry, I wasn't supposed to mention that! |
| 02:13 | <Lachy> | my point was that it's the kind of film they don't just put subtitles on |
| 02:14 | <Hixie> | Lachy: when we saw the incredibles, they had different showings, one dubbed, and one subbed |
| 02:14 | <jcranmer> | Lachy: sometimes, movies are funnier w/o subtitles |
| 02:17 | <Lachy> | ok. I was just told that films suitable for children were dubbed. When it comes out, I'll check |
| 02:17 | <Lachy> | jcranmer, all films are better without subtitles. But they don't give me a choice here. |
| 02:18 | <Hixie> | maybe norwegian cinemas have an appreciation of pixar :-) |
| 02:18 | Hixie | saw the incredibles at a norwegian cinema at a pre-release event for free, thanks to eiriks :-D |
| 02:19 | jcranmer | prefers watching Japanese anime in subtitles |
| 02:19 | <jcranmer> | because the dubs can be just WRONG wrt voices |
| 02:19 | <Hixie> | dubbing is wrong in general |
| 02:19 | <jcranmer> | there are a few good dubs, though |
| 02:20 | <Hixie> | still wrong :-) |
| 02:21 | <jcranmer> | what's fun is to watch a dub with subs and see how different they are |
| 02:21 | <Philip`> | At least anime tends to show each scene for ten seconds with either no animation or some simple looping, which gives you plenty of time to read the subtitles without missing the on-screen action :-) |
| 02:22 | <jcranmer> | Philip`: depends on the anime |
| 02:22 | <jcranmer> | if it's something like Dragonball Z, ten seconds is a bit of an understatement |
| 02:23 | <jcranmer> | fighting animes tend to be best viewed at about 2-4x speed on your DVD/VHS player |
| 02:23 | <jcranmer> | or VLC, for that matter |
| 02:23 | <Lachy> | crap like Dragonball Z is best not viewed at all. In fact, a lot of the anime I've seen come out in the past few years has been generally crap |
| 02:24 | <jcranmer> | too true |
| 02:24 | <jcranmer> | any series which goes on past 100 episodes is way too long |
| 02:25 | <jcranmer> | Saiyuki's 50 is even pushing it on length |
| 02:25 | <Hixie> | naruto was pretty good, though i never finished it |
| 02:26 | <jcranmer> | Hixie: you can't have, it's still not finished |
| 02:26 | <jcranmer> | I think they're on ep 200 or something by now |
| 02:26 | <Lachy> | there were some good shows when I was younger though, like Sailor Moon and Samurai Pizza Cats. |
| 02:26 | <Lachy> | never seen naruto before. |
| 02:26 | <Hixie> | i got to about 98 before getting more became more work than i was willing to put into it |
| 02:26 | <jcranmer> | at least the people who translated it got a lot better after some royally bad ones |
| 02:27 | <Hixie> | (naruto is by far the best anime i've seen) |
| 02:27 | <jcranmer> | Ino's "Mind Control Technique" -> "Super secret valentine technique" |
| 02:27 | <jcranmer> | Hixie: I would care to disagree, but it's certainly better than a lot of the pulp on Saturday morning cartoons |
| 02:29 | jcranmer | remembers when Power Rangers was actually good |
| 02:29 | <Lachy> | yeah, the original series of power rangers was awesome. |
| 02:29 | <Lachy> | then it just went downhill quite rapidly |
| 02:30 | <jcranmer> | I think I watched the first two... |
| 02:30 | jcranmer | was only like 5 then |
| 02:32 | <Lachy> | I was about 11-12 then |
| 02:34 | jcranmer | wonders if the logs are rolling right now |
| 02:34 | <jcranmer> | yep |
| 06:43 | <hsivonen> | Lachy: doesn't Norway have Pixar films *also* available with the original sound track? |
| 06:54 | <Dashiva> | Sometimes, but not always |
| 06:55 | <hsivonen> | weird. In Finland, all Disney films are always in theaters in Finnish and English. |
| 06:56 | <hsivonen> | and the official story for explaining dubbing is that someone at Disney made a policy that forbids subtitling their art |
| 06:57 | <hsivonen> | (but now there's a silly mentality to dub animated movies. whoever decided to commission the dubbing of the Simpson's movie clearly had no clue of the product they were importing) |
| 06:58 | <Dashiva> | Animation is for kids |
| 06:58 | <Dashiva> | Get with the times :P |
| 06:58 | <hsivonen> | Harry Potter isn't, though. |
| 06:58 | <hsivonen> | or it's for literate kids |
| 07:01 | <Dashiva> | Are they still burning those books? |
| 07:02 | <hsivonen> | I don't know. I meant the Harry Potter movies aren't dubbed. |
| 08:12 | <hsivonen> | parsetree.validator.nu now uses GET for the textarea |
| 08:25 | <Lachy> | hsivonen, I may have been misinformed about how it works here. I will find out next time a pixar film is shown here |
| 08:56 | <hsivonen> | takkaria: so far my endeavors to write test cases support the previously assumption about the secondary insertion mode |
| 09:02 | <Hixie> | URLification complete! |
| 09:03 | <Hixie> | almost |
| 09:04 | <philipj> | could someone point me to some background discussion about HTMLMediaElement addCueRange/removeCueRanges methods? they seem rather like the odd man out to me, but some use case etc might be nice before complaining on the mailing lists |
| 09:04 | <Hixie> | the main use cases are subtitles and keeping a separate slideshow in sync |
| 09:04 | <Hixie> | while the user seeks arbitrarily |
| 09:04 | <Hixie> | or plays the content backwards |
| 09:05 | <philipj> | sure, that's the intention, but has anyone shown how it can be done with this interface? |
| 09:05 | <philipj> | I'm sure it's possible, but seems awkward |
| 09:06 | <Hixie> | oh |
| 09:06 | <Hixie> | ? |
| 09:06 | <philipj> | has any browser implemented it yet? |
| 09:06 | <philipj> | otherwise I'd suggest some changes |
| 09:06 | <Hixie> | i don't know what the implementation status is |
| 09:06 | <Hixie> | but what changes did you have in mind? |
| 09:06 | <philipj> | the pauseOnExit boolean seems out of place. calling pause() in the exitCallback would do that |
| 09:07 | <philipj> | is this a special case that deserves special treatment? |
| 09:07 | <Hixie> | you can't guarantee that the callback will be called quick enough to pause on the right frame |
| 09:07 | <philipj> | sounds like an implementation issue to me |
| 09:07 | <Hixie> | not really |
| 09:08 | <Hixie> | if you want the content paused on a particular frame (e.g. at the end of a fade-to-black), you don't want the cpu load on the machine to affect that |
| 09:08 | <Hixie> | there's no sane way an implementation can run js code fast enough to guarantee that it'll work right |
| 09:09 | <Hixie> | after all, some other code might be running at the precise millisecond you need to pause at |
| 09:09 | <Hixie> | in many UAs, that code might even be in another tab |
| 09:10 | <philipj> | you're right |
| 09:10 | <philipj> | is such precision necessary though. why not just seek to the frame while paused? |
| 09:11 | <Hixie> | the use case there would running an interstitial ad from a separate <video> element |
| 09:11 | <Hixie> | you don't want the user to see the next frame or hear the next bit of audio, then hear clipping, then see the ad |
| 09:11 | <Hixie> | that's just ugly |
| 09:12 | <Hixie> | same with, say, an interactive movie |
| 09:12 | <philipj> | ok, fair enough |
| 09:12 | <Hixie> | where you play a scene, then wait for input |
| 09:12 | <philipj> | second, why is there a VoidCallback interface, why not just a function like for addEventListener |
| 09:13 | <philipj> | is there something that's made simpler by encapsulating it in an object? |
| 09:13 | <annevk> | addEventListener takes an EventListener |
| 09:14 | <annevk> | (in the ECMAScript binding this boils down to a function) |
| 09:14 | <Hixie> | as anne says, VoidCallback is a function |
| 09:14 | <Hixie> | i don't like the way it's written today |
| 09:14 | <Hixie> | i really just want to have it take a Function or something |
| 09:14 | <Hixie> | but this is an open issue between html5 and webidl |
| 09:15 | <Hixie> | we haven't quite worked out how to do it yet |
| 09:15 | <philipj> | ok, what is supposedly the problem? |
| 09:15 | <Hixie> | it's unclear how to indicate that in JS you should take a function, but in a language without function pointers, you should take an object with a single handleEvent method |
| 09:16 | <Hixie> | same problem addEventListener's definition has |
| 09:17 | <philipj> | I see |
| 09:17 | <philipj> | so how would you expect VoidHandler to map to JS? |
| 09:18 | <Hixie> | (addEventListener is defined to take an interface-implementing object, not a function -- it's made into a function using prose, just like VoidCallback) |
| 09:18 | <Hixie> | (we use VoidCallback and not EventListener in the media case because there's no argument) |
| 09:18 | <Hixie> | what i really want is for IDL to support a native function type |
| 09:18 | <Hixie> | that expands into an interface with a matching handleEvent method in the languages where that is necessary |
| 09:18 | <Hixie> | but webidl doesn't yet do that |
| 09:18 | <Hixie> | zero-argument function |
| 09:18 | <philipj> | ok, what prose is this that I should be reading? |
| 09:19 | <Hixie> | i think the spec says that somewhere, hold on |
| 09:20 | <Hixie> | webidl, section 4.5, last paragraph |
| 09:20 | <Hixie> | http://dev.w3.org/cvsweb/~checkout~/2006/webapi/Binding4DOM/Overview.html?content-type=text/html;%20charset=utf-8 |
| 09:20 | <Hixie> | "As a special case, a Function object F is considered to implement an interface if that interface is declared with exactly one operation." |
| 09:21 | <Hixie> | i really want the spec to further say that in JS, an object _doesn't_ satisfy that interface |
| 09:21 | <philipj> | ok, so I suppose we want to add a similar statement to the html5 spec |
| 09:21 | <Hixie> | (heycam and i need to sit down and hammer this out at some point) |
| 09:21 | <Hixie> | html5 imports webidl |
| 09:21 | <philipj> | oh |
| 09:22 | <philipj> | then maybe a note, it is not obvious at all when reading the spec if you're unfamiliar with webidl |
| 09:22 | <Hixie> | webidl is one of those amazing specs that makes my life like a bazillion times easier |
| 09:22 | <Hixie> | yeah, i agree |
| 09:22 | <Hixie> | i haven't really done anything with it yet since i want to just change the whole way we do callbacks |
| 09:22 | <Hixie> | i'll make a note |
| 09:22 | <annevk> | can't we just say "in Function foo" for now? |
| 09:23 | <annevk> | that's what Web IDL should probably say anyway |
| 09:23 | <Hixie> | annevk: what's Function's signature? |
| 09:23 | <annevk> | signature? |
| 09:23 | <philipj> | for implementation purposes, may I assume that VoidCallback is a pointer takes a function with no arguments, and not handle an object with a handleEvent (should be handleCallback?) method? |
| 09:23 | <Hixie> | anne: arguments, return value |
| 09:24 | <Hixie> | philipj: yeah, do the same as for EventListener / addEventListener, except with no arguments instead of one argument |
| 09:24 | <annevk> | Hixie, I thought the idea was no arguments, if it's more complex, never mind |
| 09:24 | <Hixie> | annevk: some are no arguments (like VoidCallback), others take arguments. And in some languages, "Function" makes no sense. |
| 09:25 | <annevk> | well, WebIDL would say that in some languages Function would be implemented as FunctionCallbackObject or some such |
| 09:25 | <Hixie> | right |
| 09:25 | <Hixie> | but we need syntax to define the signature |
| 09:26 | <Hixie> | typedef URLEventToBoolCallback = Function (in DOMString url, in Event event): boolean; or whatever |
| 09:29 | <philipj> | finally then, do we expect script authors to create a unique function for each entry and exit point of each range? it's either that or inspecting the currentTime isn't it? |
| 09:30 | <Hixie> | yeah, typically they'd create lambdas |
| 09:30 | <Hixie> | or use currying to generate the callbacks |
| 09:31 | <Hixie> | that's why removeCueRange() doesn't take a function pointer but just takes a class name |
| 09:32 | <philipj> | I suppose the same exitCallback could be used for all ranges, at least in the subtitles case (clear text) |
| 09:33 | <heycam> | Hixie, why do you specifically want to disallow an object with a handleEvent property (or such)? |
| 09:34 | <Hixie> | heycam: because it's unnecessary and unnecessary things tend to be poorly tested and thus buggy |
| 09:35 | <Hixie> | and especially in this case, handling arbitrary js objects is going to be very buggy |
| 09:36 | <heycam> | extended attribute? :) |
| 09:37 | heycam | bbl, visitors |
| 09:38 | jgraham | wonders how many months it will be before Pixar deign to release the exact same film in the UK |
| 09:39 | <Hixie> | ok bed time |
| 09:39 | <Hixie> | nn |
| 09:44 | jgraham | discovers the delay is only 3 weeks this time which is better than 4 months for Ratatouille |
| 11:12 | <Windstoss> | Maybe you folks can gave me a hint. It is said, HTML 5 integrates a new version of DOM Level 2 HTML. The HTMLElement interface defines event handler DOM attributes, AFAIK not covered in DOM2 HTML. On the other hand, addEventListener() specified in DOM2 Event does not seem to be specified. Whats the relation of HTML 5 to DOM2 Event? |
| 11:13 | <annevk> | HTML5 replaces DOM2HTML, not DOM2Events |
| 11:13 | <annevk> | DOM2Events is just something HTML5 works together with |
| 11:15 | <Windstoss> | annevk: And where do the HTMLElement event handler come from? Was it actually part of DOM2HTML? |
| 11:15 | <Windstoss> | annevk: Well, I guess there is a reference to DOM 2 Events I haven't spotted yet ;) |
| 11:15 | <annevk> | It might reference DOM 3 Events |
| 11:16 | <annevk> | the event handler attributes were never defined before |
| 11:16 | <annevk> | obviously they should've been and now are |
| 11:16 | <Windstoss> | ah, so it is part of what some call DOM 0? |
| 11:17 | <annevk> | yeah, likely |
| 11:17 | <annevk> | though not anymore, it's now in HTML 5 :p |
| 11:20 | <Windstoss> | Thanks… nice interview, btw (crueltobekind.org) |
| 11:25 | <annevk> | ty |
| 11:25 | <hsivonen> | annevk: do you have presentation materials available online? |
| 11:29 | <annevk> | what do you mean? |
| 11:29 | <annevk> | as in, opera.com and wiki.whatwg.org/wiki/HTML5_Presentations ? |
| 11:30 | <hsivonen> | annevk: is there a set of slides, a sound recording or a video of you reboot presentation on the Web? |
| 11:30 | <annevk> | aah, I haven't given it just yet |
| 11:30 | <hsivonen> | oh. reboot is long |
| 11:30 | <annevk> | but they're making high quality video recordings and hopefully they put it sooner online than last year |
| 11:30 | <annevk> | it's just two days |
| 11:31 | <hsivonen> | I had thought the interview was done on location |
| 11:32 | <annevk> | I see, that was done over skype last weekend |
| 11:33 | <annevk> | she e-mailed all presenters beforehand |
| 12:00 | hsivonen | wonders who wrote "RDFa" in the HTML WG charter |
| 12:09 | <Lachy> | hsivonen, luckily it's only written in there as an example of something that could be included, not something we necessarily have to include. |
| 12:11 | <Lachy> | does anyone here know anything about the Internationlisation Tag Set, which is also mentioned in our charter? |
| 12:20 | hsivonen | finds bug in foreign content impl |
| 12:20 | <Lachy> | oh, wow, it looks like an incredibly complicated spec |
| 12:33 | hsivonen | deems it a spec bug instead of impl. bug |
| 14:28 | Philip` | quite likes the "http://whatwg.org/html5#src1"; link syntax |
| 14:30 | <Philip`> | It's a whole three characters longer than the equivalent TinyURL, though :-( |
| 14:34 | <Lachy> | hey, does anyone know where the contribution guildelines are written up, which explain things like documenting use cases, problems and finding supporting evidence? I remember Hixie wrote up something like that somewhere, but can't find it |
| 14:39 | <hsivonen> | Lachy: http://wiki.whatwg.org/wiki/FAQ#Is_there_a_process_for_adding_new_features_to_the_spec.3F |
| 14:39 | <Lachy> | thanks |
| 14:40 | <hsivonen> | Lachy: the previous question covers removing stuff that Hixie had put in presumably without having point #2 covered in advance :-) |
| 15:59 | <BenMillard> | I've been collecting website addresses mentioned in offline media for a while, so I've published them: http://projectcerbera.com/web/study/url/offline |
| 16:00 | <BenMillard> | no analysis or catagorisation, just the URLs grouped by date with the source they came from |
| 16:00 | <hsivonen> | BenMillard: did you omit http:// and www or did the ads do that consistently? |
| 16:01 | <BenMillard> | hsivonen, I omitted it when it was present |
| 16:01 | <BenMillard> | it may have been interesting to keep it when present, though |
| 16:02 | <BenMillard> | there may be small transcription errors, too |
| 16:03 | <BenMillard> | some adverts flash the website name for a really short amount of time, so a few addresses I glimpsed but couldn't write while others are incomplete (usually missing the TLD) |
| 16:09 | <annevk> | ah, elharo spotted the URI thread |
| 16:09 | <Philip`> | BenMillard: The world seems to contain quite a lot of advertising :-/ |
| 16:10 | <BenMillard> | yeah, I was surprised to stumble upon neary 40 URLs in the first day! |
| 16:11 | <BenMillard> | this is a small but neat complement to DMOZ, which is like an online URL collection :) |
| 16:12 | <BenMillard> | entirely UK-centric, though |
| 16:14 | <BenMillard> | I'm thinking of collecting every URL from my local high street to see how big a business has to be to have a website |
| 16:15 | <annevk> | Does RB understand what <th headers> does? It seems that he thinks you'd use it to point to <td> elements? |
| 16:15 | <annevk> | Maybe I shouldn't care |
| 16:15 | annevk | deletes e-mail |
| 16:19 | <itpastorn> | Off topic but you guys are usually friendly. I am looking for JavaScript (ES, DOM) material that are written from a computer science perspective (besides Doug C:s stuff) |
| 16:19 | <BenMillard> | annevk, he does seem to be making that error |
| 16:23 | <annevk> | itpastorn, dunno; I don't really know good ES/DOM resources at all |
| 16:23 | annevk | learned it by try and error and IDL fragments in the DOM specifications... |
| 16:25 | <itpastorn> | IDL? |
| 16:26 | <annevk> | see http://www.w3.org/TR/DOM-Level-2-Events/events.html#Events-UIEvent for instance |
| 16:26 | <annevk> | stands for "interface definition language" |
| 16:29 | <itpastorn> | Ah, those grey boxes.... Not International Darts League ;-) |
| 16:29 | <annevk> | heh |
| 16:31 | <annevk> | I sort of like the idea of open-ended TLDs |
| 16:33 | <annevk> | Though it seems that the way it works is that there's a still a single platform handing them out and it will earn a lot of money on this |
| 16:46 | <BenMillard> | annevk, at the November 2007 F2F I complimented your blog |
| 16:46 | <BenMillard> | I've been working on bringing my own into the 21st century, summarised here: http://www.accessifyforum.com/viewtopic.php?t=10762 |
| 16:58 | <Philip`> | Is it "an 0xDD byte" or "a 0xDD byte"? |
| 17:06 | <hober> | an |
| 17:06 | <hober> | if you read the zero |
| 17:07 | <Philip`> | An implicit part of my question was "do you read the zero?", so it's no good making the answer conditional on that :-p |
| 17:08 | <Philip`> | I'd read it as "an oh ex dee dee byte" but I don't know if I'm just crazy |
| 17:11 | <jcranmer> | "an octet with value 0xDD" |
| 17:11 | <jcranmer> | oh you can go Intel and say "DDh" |
| 17:11 | <MikeSmith> | Philip`: an awks DD byte |
| 17:11 | <BenMillard> | Philip`, I'd say what you read |
| 17:39 | <gsnedders> | Philip`: an |
| 19:14 | <Philip`> | I wish Google search results would link to a fragment identifier in the page near the search terms |
| 19:15 | <jacobolus> | Philip`: you don't have incremental search in your browser? |
| 19:15 | <Philip`> | Does http://www.w3.org/TR/html401/interact/forms.html#adef-accept-charset make any sense? I can't work out what it's talking about |
| 19:16 | <Philip`> | jacobolus: Yes, but that means I have to type the search terms into the search box, then click the result, then press the incremental search key and type the search terms in a second time |
| 19:16 | <jacobolus> | Philip`: I dunno… I just ⌘E, ⌘G them on a Mac. :) |
| 19:17 | <Philip`> | I don't even know how to pronounce those keys |
| 19:17 | <jacobolus> | command-E, command-G |
| 19:18 | <Philip`> | I usually alt-D tab {type} enter click slash {type again} |
| 20:49 | <Hixie> | hsivonen: nah removing features applies to all features, including the ones introduced carefully |
| 21:34 | <Hixie> | should <base href> be affected by xml:base? |
| 21:34 | <Hixie> | i guess not, since xml:base is affected by <base href> |
| 21:35 | <Hixie> | right now the loop is broken by saying that xml:base isn't affected by base href when resolving base href |
| 21:35 | <Hixie> | but i think it should just not affect base href |
| 21:37 | <Dashiva> | Wasn't that what the conclusion was last time too? |
| 21:37 | <Hixie> | probably :-) |
| 21:53 | <Hixie> | apparently <base> doesn't work in document.write()n documents in IE |
| 21:53 | <Hixie> | lovely |
| 22:19 | <Hixie> | <a href>, <a ping>, <q cite>, <blockquote cite>, <ins cite>, and <del cite> appear to be the only attributes affected by base url changes |
| 22:19 | <Hixie> | and they're only affected for UI purposes and :link/:visited |
| 22:57 | <Hixie> | where does RFC3986 say how to resolve a relative URI reference relative to a base URI that uses the mailto:, data:, javascript:, or similar schemes? |
| 23:07 | <Hixie> | crap, backslashes |
| 23:07 | <Hixie> | wonder we should do with _those_ |
| 23:10 | <gsnedders> | Hixie: It doesn't define anything special |
| 23:10 | <gsnedders> | Hixie: WebKit just replaces all \ with / |
| 23:54 | <Hixie> | gsnedders: not if you type them into the location bar |