| 09:08 | <krijnh> | Ping |
| 09:09 | <krijnh> | Shitty connection today :/ |
| 09:17 | <krijnh> | Pong |
| 11:00 | <krijnh> | Grmbl |
| 11:01 | <hsivonen> | the source line numbering is right |
| 11:02 | <krijnh> | Hey, I'm back |
| 11:02 | <hsivonen> | but the parser line numbering goes into the weed but there's nothing particularly strange in the source at that point... |
| 11:10 | <madness> | Camaban: re: opera, yes, that too |
| 11:44 | <Philip`> | "[whatwg] [html5] Unsubcribe me!" -- that's not fair - artificially inflated subscriber numbers are the most useful outcome of the Ogg issue, but it will be lost if people just unsubscribe :-( |
| 11:44 | <stijntje> | hahaha wtf |
| 11:44 | <stijntje> | :SD |
| 11:45 | <stijntje> | "$$ka-ching$$" |
| 11:47 | <Lachy> | looks like mo0n_sniper just subscribed to send that pointless message and now wants to unsubscribe |
| 11:49 | <othermaciej> | that was pretty sad, yeah |
| 11:59 | <hsivonen> | yay. I managed to minimize Sam's dogfood case to 5 lines of XML |
| 12:00 | <maikmerten> | stijntje, yeah, I totally love this professional unsubscripe request, too |
| 12:00 | <maikmerten> | "hit and run" |
| 12:00 | <maikmerten> | *unsubscribe |
| 12:00 | <hsivonen> | unsophisticated advocacy doesn't reflect well on any community :-( |
| 12:01 | <maikmerten> | yeah, and I always get a slight stomache when seing things like this |
| 12:02 | <stijntje> | did I miss something? I don't seem to have the mail Shannon is replying to in his latest |
| 12:02 | <maikmerten> | err... stomachache |
| 12:04 | <hsivonen> | stijntje: I don't see what (s)he is replying to, either |
| 12:04 | <stijntje> | might be a private mail from Dave |
| 12:05 | <hsivonen> | in-reply-to points to the same domain as message-id |
| 12:05 | <maikmerten> | which reminds me how I often just hit "reply" on a mailing list post and thus accidentially answer off-list |
| 12:05 | <stijntje> | that happens to me too |
| 12:06 | <maikmerten> | back in ye olde days "reply" would always get back to the list |
| 12:06 | <maikmerten> | I'm not sure why all lists seem to have changed this |
| 12:07 | <maikmerten> | (I now hit "reply all" and then kill all non-list addresses so people won't get the same thing twice) |
| 12:08 | <stijntje> | is Shanon a male or female name, by the way? |
| 12:09 | <hsivonen> | stijntje: it is at least female, but I don't know it works as a male name, too |
| 12:09 | <stijntje> | hsivonen: thank you |
| 12:10 | <maikmerten> | on the stomachache thing: It's nice if people show support for free formats and all... but why can some guys not stick to minimum social standards? (even if impulse and passion drives them away) |
| 12:10 | <stijntje> | or read up on the thing about which they're upset :) |
| 12:12 | <maikmerten> | (having said that I must confess I also did write some messages to the list I wished I wouldn't have sent) |
| 15:28 | <hsivonen> | What foreign namespaces are reasonable in XHTML5 other than SVG, MathML and RDF (plus whatever happens in the RDF subtree)? |
| 15:30 | <hsivonen> | is there any good-for-the-Web reason to use XLink in XHTML5? |
| 15:33 | <hsivonen> | the XBL2 spec has an example that embeds XBL2 in SVG. is it reasonable to embed XBL2 in XHTML5? |
| 15:33 | <Philip`> | There's XLink in SVG subtrees |
| 15:34 | <hsivonen> | Philip`: yeah, embedding SVG covers that case |
| 15:35 | <annevk> | it's reasonable |
| 15:36 | <annevk> | XLink should be avoided when possible |
| 15:36 | <hsivonen> | annevk: ok. what's our XBL2 in text/html story? |
| 15:36 | <annevk> | CSS |
| 15:36 | <annevk> | for now anyway |
| 15:36 | <hsivonen> | ok. XHTML5 > HTML5 then :-) |
| 15:37 | <annevk> | euh, run |
| 15:37 | <hsivonen> | does RDF have some kind of notion of correctness that doesn't involve knowledge about a particular vocabulary/ontology/whatever |
| 15:37 | <hsivonen> | ? |
| 15:38 | <hsivonen> | I'm wondering if I should make RDF a black hole or plug in Jena |
| 15:38 | <hsivonen> | (though I'm only going to plug in Jena if it's useful and low-hanging fruit) |
| 15:39 | <hsivonen> | is it legitimate to use XHTML5/SVG/MathML elements inside an RDF subtree? |
| 15:39 | <annevk> | given how widely RDF is deployed I would prioritize other things personally |
| 15:39 | <hsivonen> | annevk: yeah. |
| 15:39 | <hsivonen> | black hole it is |
| 15:52 | <Dashiva> | RDF is like prolog, it's all backwards |
| 15:53 | <Dashiva> | Talking about correctness is tricky |
| 15:54 | <gsnedders> | I. am. correct. |
| 15:54 | <gsnedders> | That wasn't hard. |
| 15:54 | gsnedders | looks up logs to see what he just said |
| 16:38 | <gsnedders> | god. another 51 emails on whatwg. |
| 16:38 | <G0k> | i have a solution for you |
| 16:38 | <stijntje> | mostly ogg drama |
| 16:38 | <G0k> | if contains 'ogg' -> junk |
| 16:42 | <gsnedders> | :P |
| 16:51 | <oxygenws> | in globalStorage, what happen to IP browsing sites? |
| 16:59 | <G0k> | so has anyone even considered implementing the network connections stuff (section 6.3) ? |
| 17:40 | <Hixie> | wow, lots of mail |
| 17:41 | <inimino> | I have replied to some off-list |
| 17:43 | <Hixie> | the internet gods aren't wanting me to reply to this mail |
| 17:43 | <Hixie> | every time i try, my connection dies |
| 17:44 | <inimino> | it seems there are widely divergent opinions on the coercive value of a SHOULD in the spec |
| 17:44 | <Hixie> | it seems pretty academic, since all the browser vendors who would be coerced have already implemented Theora, and the others have said they won't do it |
| 17:44 | <inimino> | and perhaps of the value of <video> without a real agreed-upon baseline |
| 17:45 | <Hixie> | i think it's pretty clear that <video> without a common codec would be a failure |
| 17:45 | <inimino> | right |
| 17:45 | <inimino> | I think a token SHOULD mentioning Ogg could be as great of a failure |
| 17:46 | <Hixie> | sure, since it wouldn't be a common codec |
| 17:47 | <inimino> | specifically if some other format is actually interoperable across any sizeable group of UAs |
| 17:48 | <Hixie> | if some format is interoperable, and it doesn't suck too much, it'll be successful |
| 17:48 | <Hixie> | regardless of what the spec says |
| 17:48 | <inimino> | right |
| 17:48 | <inimino> | and regardless of whether it is interoperable by free software |
| 17:48 | <inimino> | s/interoperable/implementable/ |
| 18:36 | <gsnedders> | Flow 5.0: <http://www.engadget.com/2007/12/14/art-installation-made-up-of-hundreds-of-case-fans-is-full-of-air/>; |
| 21:29 | <Philip`> | Is it bad to write web content using __defineGetter__? |
| 21:30 | <Philip`> | (particularly for something around canvas-3d, so I can make some assumptions about what browsers users will have) |
| 21:31 | <Dashiva> | It's part of es4, so sure |
| 21:31 | <gavin_> | is it part of es3? |
| 21:32 | <gavin_> | does it work in safari/opera? |
| 21:32 | <Philip`> | I guess I only really care about Firefox 3 and Opera 10 and Safari 4, since my code won't work in anything else |
| 21:32 | gavin_ | is just curious |
| 21:33 | <Philip`> | I vaguely remembering hearing that Opera 9.5 added it |
| 21:33 | <Philip`> | s/ering/er/ |
| 21:37 | <Philip`> | Woah, my code works :-o |
| 21:38 | Philip` | can now automatically create a shader object to say e.g. shader.lightPosition = [1,2,3] to set the correspondingly-named GLSL variable, instead of lots of ugly getUniformLocation/uniformf calls |
| 21:47 | <Dashiva> | gavin: Not in es3, mozilla started it in their js 1.7 or so. Opera added support in 9.5. Not sure about Safari |
| 21:51 | <gavin_> | er, __defineGetter__? |
| 21:51 | <gavin_> | pretty sure that existed long before js1.7 |
| 21:52 | <gavin_> | it was landed in Mozilla CVS on 2000-03-02 15:19 with the comment "Added ECMA3 compliant getter/setter syntax." |
| 21:53 | <gavin_> | now, that commit message may not mean __defineGetter__ itself |
| 21:53 | <gavin_> | but that does suggest that it was pre-1.7 :) |
| 21:58 | <Dashiva> | I expect "ECMA3 compliant" means it doesn't extend the syntax, like { get a : function(){} } and function getter a() {} |
| 22:17 | <Philip`> | "most users don't install third-party products" ... except for Flash? |
| 22:17 | <Hixie> | most users don't install flash, it comes pre-installed |
| 22:18 | <Philip`> | Oh, does Windows do that? |
| 22:18 | <Hixie> | not sure what the status is these days. it used to, i believe. |
| 22:18 | <Hixie> | most OEMs do too |
| 22:21 | <Philip`> | Hmm, apparently at least XP does |
| 22:21 | <Philip`> | That sounds boringly non-anti-competitive |
| 22:21 | <Hixie> | XP is from before SilverLight |
| 22:21 | <Hixie> | :-) |
| 22:21 | <parcelbrat> | i didn't think MS was allowed to do that |
| 22:21 | <Hixie> | anyway, gotta go get lunch |
| 22:21 | <parcelbrat> | (be non-anti-ompetitive) |
| 22:26 | <Philip`> | parcelbrat: Indeed - I think we should sue them for it |
| 22:26 | <parcelbrat> | You and me, sholdn't be a problem ;) |
| 22:27 | Philip` | starts writing the press release and preparing interviews |
| 22:27 | parcelbrat | hides his money |
| 23:07 | Philip` | hates rotateX/rotateY/rotateZ |
| 23:12 | <stijntje> | my XP didn't pre-install Flash last time I installed it (2 weeks ago( |
| 23:12 | <stijntje> | ) |
| 23:12 | Philip` | tries to find someone's affine decomposition code to steal |
| 23:12 | <dbaron> | Was Apple's CSS animations proposal discussed on the WHATWG list in addition to www-style? I sent some comments in response to the www-style post; not sure if I should send them elsewhere. |
| 23:13 | <Hixie> | i don't think so |
| 23:13 | Philip` | doesn't think so either |
| 23:14 | Philip` | really doesn't want to port http://www.soe.ucsc.edu/~pang/160/f98/Gems/GemsIV/polar_decomp/Decompose.c to JS |
| 23:15 | <othermaciej> | dbaron: no, it wasn't discussed on the WHATWG list |
| 23:20 | <parcelbrat> | C to JS OUCH! |
| 23:22 | <gsnedders> | hmm… |
| 23:22 | <gsnedders> | nobody sending any unacceptable Content-Length values. |
| 23:23 | <Philip`> | gsnedders: Are there any >256KB? (I set that as the maximum to accept, so hopefully HttpClient still downloaded all the headers and first 256KB properly and just stopped before downloading the rest, but I'm not positive) |
| 23:24 | <Philip`> | (At least, I *think* it was 256KB...) |
| 23:25 | gsnedders | needs to decide what to do with invalid Etag values |
| 23:26 | <gsnedders> | "59584b8bdef9e2e442f03cf8945a7229" does not match the invalid value 59584b8bdef9e2e442f03cf8945a7229 |
| 23:26 | <gsnedders> | Which means the standard quoted-string error handling |
| 23:26 | <gsnedders> | …doesn't work |
| 23:28 | Philip` | wonders whether rotateX goes clockwise or anticlockwise |
| 23:28 | Philip` | also wonders which direction is clockwise |
| 23:29 | Philip` | uses his standard 3d graphics approach of guessing blindly, then fiddling with minus signs until it renders correctly |
| 23:29 | gsnedders | points in a clockwise direction |
| 23:29 | <gsnedders> | silly IRC limitations |
| 23:30 | <gsnedders> | Philip`: max Content-Length value is 9992 |
| 23:31 | <gsnedders> | (which is amazing small, methinks) |
| 23:31 | <Philip`> | If you turn a clock around, clockwise goes backwards, and if you don't have an actual clock face to start with then you can't really tell which way is backwards :-( |
| 23:31 | <Philip`> | gsnedders: Try sorting numerically, not alphabetically :-p |
| 23:31 | <gsnedders> | Philip`: that was :P |
| 23:32 | <gsnedders> | if header["value"] > max:\n\tmax = header["value"] |
| 23:32 | <Philip`> | Put some int(...) in there to make sure it's numbers |
| 23:32 | <gsnedders> | Philip`: they all are |
| 23:33 | gsnedders | opens the file and finds a larger one straight away |
| 23:33 | gsnedders | wonders how Python does > on a string |
| 23:34 | <Philip`> | 238242 is the maximum |
| 23:34 | <Philip`> | (9992 is the maximum if you do string comparisons) |
| 23:35 | gsnedders | is nowadays used to >, + and the like casting to int/float implicitly |
| 23:35 | <Philip`> | Hmm, it looks suspiciously like I dropped anything >256KB |
| 23:36 | <gsnedders> | now I get 3487 as the max value with int() :\ |
| 23:36 | <gsnedders> | oh. wait. |
| 23:36 | <gsnedders> | I printed the wrong value. |
| 23:36 | <gsnedders> | I printed the last content-length value :P |
| 23:36 | <Philip`> | http://www.untraveledroad.com/USA/Nevada/Elko.htm - "Content-Length: 00006330" |
| 23:37 | <gsnedders> | well, as a normal integer, all leading zeros are pointless. |
| 23:37 | <Philip`> | Are they allowed in theory/practice? |
| 23:38 | <gsnedders> | in theory allowed, in practice as far as I know they are too |