| 03:33 | <Yuhong> | FYI, on DBCS character encodings, here is a good test: http://ha.ckers.org/weird/variable-width-encoding.cgi |
| 05:23 | <WeirdAl> | annevk: ping, spec conflict with implementations on XHR.open(), responseType != "" |
| 05:23 | <WeirdAl> | not just Mozilla, allegedly Chrome & Safari too |
| 05:23 | <WeirdAl> | https://bugzilla.mozilla.org/show_bug.cgi?id=680816#c18 |
| 05:24 | <WeirdAl> | I'm wondering if the spec should be changed. |
| 14:08 | <annevk> | weirdal: please file bugs (not sure what the problem is with the spec in the second case though) |
| 16:38 | <annevk> | can you say "A tell of tales"? |
| 16:38 | <annevk> | I rather like the sound of it, but I'm not sure if it makes any sense |
| 16:41 | <gsnedders> | annevk: no |
| 16:41 | <annevk> | aah :( |
| 16:48 | <zewt> | maybe in poker? heh |
| 17:18 | <annevk> | http://www.luxetveritas.nl/blog/?p=1291 local politics o_O |
| 17:19 | <annevk> | whenever I look at other social systems, I'm pretty pleased with the WHATWG/W3C environment I'm in |
| 17:57 | <smaug____> | interesting, Finland has that much higher trade union membership% (if I trust the translation) |
| 18:05 | <annevk> | smaug____: will fix the async thingie now |
| 18:06 | <annevk> | smaug____: yeah, quite high |
| 18:06 | <smaug____> | great, thanks |
| 18:08 | <smaug____> | annevk: where should I file fullscreen spec bugs ? |
| 18:09 | <annevk> | dunno |
| 18:09 | <annevk> | email whatwg⊙wo at this point I think |
| 18:12 | <annevk> | smaug____: should abort() unset the synchronous flag |
| 18:12 | <annevk> | ? |
| 18:15 | <zewt> | should it just be cleared by open()? (eg. "if async is false, set the synchronous flag, otherwise clear it") |
| 18:16 | <smaug____> | annevk: I think so |
| 18:16 | <annevk> | zewt: no |
| 18:17 | <annevk> | smaug____: okay, so just like abort() always sets the error flag, always unset the synchronous flag |
| 18:18 | <annevk> | zewt: in particular, now you change things during UNSENT that depend on that flag, it needs to be cleared when things are UNSENT/DONE |
| 18:18 | <zewt> | mmm |
| 18:24 | <annevk> | food \o/ |
| 18:45 | <annevk> | heh http://news.bbcimg.co.uk/media/images/57217000/jpg/_57217771_la-grande-divergenceweb2.jpg |
| 20:50 | <annevk> | great, the charset related attributes of Document are not interoperable |
| 20:51 | <annevk> | hmm |
| 20:51 | <annevk> | document.defaultCharset is only supported in Chrome |
| 20:51 | <annevk> | oh and WebKit, but casing is different |
| 20:51 | <annevk> | seems to me that feature can be nuked |
| 20:54 | <annevk> | Gecko also does not support document.charset |
| 20:54 | <annevk> | characterSet |
| 20:54 | <annevk> | only ^^ |
| 20:54 | <annevk> | so maybe Mozilla should encourage others to drop theirs... |
| 20:57 | <kennyluck> | Yeah, I noticed that too. Very annoying. Can we at least pick one and make the rest obsolete? I mean, in the sense that the obsoleted ones go into the Obsolete chapter. (which doesn't mean browsers don't need to implement them) |
| 20:58 | <annevk> | in DOM4 the obsolete chapter means they get killed from browsers |
| 20:58 | <annevk> | but it's not called obsolete |
| 20:58 | <Ms2ger> | It's called Historical |
| 20:58 | <kennyluck> | annevk, that certainly doesn't apply to HTML LS. |
| 20:58 | <annevk> | getting rid of document.charset would definitely be good, because it allows changing the encoding of a document |
| 20:59 | <annevk> | (which affects URLs and form submission I think) |
| 20:59 | <annevk> | return values are also not very consistent |
| 20:59 | <kennyluck> | yuck! |
| 21:00 | <annevk> | e.g. "ascii" yields "windows-1252" in some, "US-ASCII" in others, and "us-ascii" in others again |
| 21:00 | <annevk> | browsers are so much fun |
| 21:01 | <Ms2ger> | https://bugzilla.mozilla.org/show_bug.cgi?id=647621 |
| 21:01 | <annevk> | guaranteed job security with the mess that is the platform |
| 21:02 | <annevk> | Hixie: why did you add .charset and .defaultCharset? |
| 21:04 | <annevk> | http://html5.org/tools/web-apps-tracker?from=1459&to=1460 looks like an attempt at defining WebKit's IDL |
| 21:05 | <annevk> | Ms2ger: remove from DOM? |
| 21:06 | <Ms2ger> | Oh, these are ours now? |
| 21:06 | <Ms2ger> | wfm |
| 21:08 | <annevk> | "please take care of this" "ok dropped from the platform" |
| 21:08 | <annevk> | I guess I should first file a bug on WebKit |
| 21:09 | <annevk> | angel spec writer |
| 21:11 | <Ms2ger> | I believe Hixie's thesis was that there are no angels |
| 21:13 | <annevk> | https://bugs.webkit.org/show_bug.cgi?id=74819 |
| 21:15 | <kennyluck> | annevk, how do you know there's no site relying on document.charset? Not I am against removing it. Just curious. |
| 21:16 | <Ms2ger> | kennyluck, it would be broken in Firefox? |
| 21:16 | <annevk> | Gecko would have open bugs and Opera would have had more bugs too |
| 21:17 | <annevk> | we had exactly one bug where we implemented both charset and characterSet, but we actually only needed characterSet |
| 21:17 | <kennyluck> | Ms2ger, I would guess when people see junk, they don't file a bug because are used to turning charset already. |
| 21:17 | <annevk> | I filed a bug to remove .charset |
| 21:18 | <annevk> | kennyluck: you cannot use these attributes for re-encoding something... |
| 21:18 | <annevk> | re-decoding* |
| 21:36 | <annevk> | Ms2ger: you should ping someone regarding https://bugzilla.mozilla.org/show_bug.cgi?id=623610 |
| 21:36 | <annevk> | Ms2ger: also, you might want to patch http://mxr.mozilla.org/mozilla/source/intl/uconv/src/charsetalias.properties#515 |
| 21:38 | <Ms2ger> | Oh, by the way, I was looking at fixing cp1258 earlier, the unicode-style file was really helpful |
| 21:39 | <annevk> | cool |
| 21:39 | <Ms2ger> | Haven't filed yet, I'll get to it soonish, I think |
| 22:54 | <annevk> | fwiw, at some point someone said governments need stable standards or some such |
| 22:54 | <annevk> | they do, but they also adapt |
| 22:55 | <annevk> | e.g. Webrichtlijnen 1.0 referenced HTML 4.01, XHTML 1.0, etc. The second edition stayed away from making such specific requirements |
| 22:56 | <annevk> | and just phrased the requirements in a different way, but thereby allowing <canvas> etc. |
| 22:56 | <zewt> | that adobe guy seems to be trying to say "everything everyone is doing is wrong" |
| 22:56 | <annevk> | (I'm not sure if it was actually ever a problem in practice, as many sites that did not conform to the letter of Webrichtlijnen 1.0 were nevertheless considered compliant...) |
| 22:57 | <annevk> | (At least a sample I looked through at some point.) |
