| 00:53 | <heycam> | Hixie, I've allowed typedefs to specify extended attributes as part of their definition now. you might like to use that for event listener attributes to avoid duplicating [TreatNonCallableAsNone] everywhere |
| 00:54 | <heycam> | Hixie, e.g. `typedef [TreatNonCallableAsNone] Function EventListener; … interface Whatever { attribute EventListener onfoo; }` |
| 01:26 | <Hixie> | heycam|away: nice |
| 04:30 | <MikeSmith> | looking at http://caniuse.com/#feat=fileapi |
| 04:31 | <MikeSmith> | what is partial about File API support in current Firefox? |
| 04:38 | <zewt> | no filesystem API, iirc |
| 09:42 | <annevk> | IE9 has responseType? |
| 10:20 | <annevk> | how does the chunked-text stuff relate to the stream stuff? |
| 10:20 | <annevk> | do we really need chunked-text if we have stream? |
| 10:20 | <annevk> | related keyword: XMLHttpRequest |
| 10:41 | <dating> | http://netdate.dk/ :-) |
| 10:42 | <annevk> | krijn: should we nuke that kind of spam from the logs? |
| 10:42 | <krijn> | Happens a lot? |
| 10:43 | <annevk> | not sure, but maybe once they realize it can work |
| 11:03 | <jgraham> | http://www.rfc-editor.org/rfc/rfc6455.txt |
| 11:07 | <annevk> | http://tools.ietf.org/html/rfc6455 |
| 11:07 | <jgraham> | annevk: But that's not a Real Not-Standard |
| 11:08 | <annevk> | is rfc-editor really the preferred address? |
| 11:08 | <annevk> | doubt that somewhat |
| 11:09 | <annevk> | but tools.ietf.org is by far the most useful imo |
| 11:09 | <annevk> | and I encourage everyone to link to that one from standards |
| 11:14 | <jgraham> | It's the one that gets linked in the announcement email at least |
| 11:14 | <jgraham> | I agree the other one is better in every way |
| 11:15 | <jgraham> | One day we might even get past the whole "but what if I want to print it on my daisy wheel printer from 1970?" thing |
| 11:15 | <jgraham> | (sigh) |
| 11:16 | <annevk> | just need to stop doing things at the IETF |
| 11:22 | <jgraham> | That would route around the damage of course |
| 11:23 | <annevk> | aah, no mike |
| 11:44 | <annevk> | fwiw |
| 11:44 | <annevk> | I started drafting some text on the encoding stuff |
| 11:45 | <annevk> | at this point it's labels and the definition |
| 11:45 | <annevk> | e.g. "iso-8859-1", "windows-1252", and the mapping table for them |
| 13:49 | <annevk> | https://twitter.com/stevefaulkner/status/146204630831996929 maybe Anolis should strip out the <dfn> related title attributes? |
| 14:03 | <zcorpan> | i find the tooltips useful, but then i don't use AT |
| 14:05 | <annevk> | yeah me too |
| 14:05 | <annevk> | I somewhat doubt they disable it because of how standards make use of tooltips |
| 14:05 | <annevk> | at least, I would expect the typical AT user not to read standards |
| 14:06 | <zcorpan> | just like the typical non-AT user doesn't read standards? or more so for AT users? |
| 14:08 | <annevk> | not sure if that's relevant |
| 14:52 | <zcorpan> | ah, so that's how we should deal with feature requests. "just relax during the coming weekend. if you still care about the use case on monday, get back to us" |
| 14:53 | <annevk> | where is that from? |
| 14:55 | <miketaylr> | heh |
| 14:55 | <miketaylr> | xhr responseType json thread |
| 14:55 | <annevk> | http://dvcs.w3.org/hg/encoding/raw-file/tip/Overview.html |
| 14:55 | <zcorpan> | http://lists.w3.org/Archives/Public/public-webapps/2011OctDec/1460.html |
| 14:55 | <annevk> | Encoding Standard ^^ |
| 14:56 | <annevk> | very much work in progress |
| 14:56 | <annevk> | and feedback wanted, too |
| 14:56 | <zcorpan> | oooh. annevk++ |
| 15:02 | <zcorpan> | annevk: it's not clear what the missing rows should be |
| 15:02 | <zcorpan> | oh wait |
| 15:03 | <zcorpan> | "Unless stated otherwise, in an 8-bit encoding the octet's hexadecimal value maps to a code point of the same hexadecimal value." |
| 15:03 | <annevk> | those things can be tidied up somewhat |
| 15:03 | <annevk> | for now I just wanted to get the basics out there |
| 15:04 | <zcorpan> | yep. looks nice |
| 15:14 | <MikeSmith> | it would be great if caniuse.com provided a number to indicate how many browser engines support a particular feature |
| 15:15 | <MikeSmith> | and then expose an API for getting just that number |
| 15:16 | <MikeSmith> | the percentages it currently lists don't really seem very useful in practice |
| 16:42 | <ksweeney> | what's the best way to inform spec author of a typo? |
| 16:42 | <ksweeney> | this room? |
| 16:42 | <Ms2ger> | Which spec? |
| 16:43 | <ksweeney> | html5 |
| 16:43 | <Ms2ger> | File a bug with the widget |
| 16:44 | <ksweeney> | thanks |
| 17:33 | <dglazkov> | good morning, Whatwg! |
| 17:35 | <Ms2ger> | Morning dglazkov |
| 18:00 | <divya> | dglazkov: can I trouble you with my queries over understanding http://dvcs.w3.org/hg/webcomponents/raw-file/tip/explainer/index.html |
| 18:10 | <david_carlisle> | annevk: do you know whetehr there are differences from the mappings on the unicode site, eg http://unicode.org/Public/MAPPINGS/VENDORS/MICSFT/WINDOWS/CP1254.TXT |
| 18:10 | <david_carlisle> | spelling:-( |
| 18:13 | <divya> | i guess dglazkov has a script running to chime in good morning every morning :| |
| 18:14 | <zewt> | it's annoying that this window gets highlighted for a null message every day, heh |
| 18:17 | <Ms2ger> | /ignore dglazkov |
| 18:20 | <zewt> | david_carlisle: well, 0x8D (and so on) are undefined according to unicode, but to U+008D in that page aaaaaand where'd he go |
| 18:23 | <zewt> | annevk: fwiw, isn't the U+FFFD mapping correct in those cases, and the unity mapping the one that's wrong? |
| 18:24 | <zewt> | (no idea if there's content expecting it; i guess if IE does it then it's likely) |
| 18:59 | <jgraham> | divya: Maybe you should just ask the room and we can puzzle it out together. Then dglazkov can tell us how wrong we were later on |
| 18:59 | <divya> | woot then let me do just that jgraham |
| 19:00 | <divya> | so here is the early draft i am reading http://dvcs.w3.org/hg/webcomponents/raw-file/tip/explainer/index.html |
| 19:00 | <divya> | there is this sentence "If the document changes so that the decorate property no longer applies, the decorator is unapplied, returning the rendering of the element back to the pre-decoration state." |
| 19:01 | <divya> | I am wondering why does it simply not say if the selectors no longer match rather than document changes so that… |
| 19:05 | <jgraham> | Are the two not equivalent? |
| 19:06 | <divya> | that is what I wanted clarification on. |
| 19:07 | <divya> | I assumed as much, but the wording seems to imply more at play than typical css selector matching? maybe its just me. |
| 19:10 | <divya> | "The script is run once when the decorator element is parsed or inserted into the document." |
| 19:12 | <divya> | nvm that quote, just ignorant me. |
| 19:20 | <divya> | dglazkov: pardon me and my flood of bugs |
| 19:21 | <Ms2ger> | annevk, did you retest Gecko for your encoding work? |
| 19:22 | <Ms2ger> | I thought I fixed the 7F-9F stuff... |
| 19:26 | <TabAtkins> | divya: They're equivalent. File a bug on the document to make him change the language, as your wording is more natural. |
| 19:27 | <divya> | k done that TabAtkins |
| 19:30 | <dglazkov> | divya: no, this is great! |
| 19:30 | <dglazkov> | divya++ |
| 19:30 | <dglazkov> | sorry, I was absent. irccloud makes me feel omnipresent |
| 19:30 | <divya> | haha okay dglazkov expect more!! |
| 19:31 | <dglazkov> | divya: this is great stuff |
| 19:31 | <divya> | i assumed you were putting out fires else where as per your godly duties. |
| 19:32 | <dglazkov> | well. I got the flu. so duties are a bit ... mucousey. |
| 19:32 | <divya> | was that what seattle was the recipient of this morning then |
| 19:33 | <dglazkov> | divya: but please do file bugs. I will go through them and edit the 'splainer. This is great stuff. |
| 19:33 | <divya> | k awesomes doing that now dl |
| 19:33 | <divya> | err dglazkov |
| 19:33 | <TabAtkins> | Hm, I wonder if I can construct a timing attack using feComponentTransfer and <foreignElement>. |
| 19:33 | dglazkov | parades divya and tells everyone that they should be like her and read the 'splainer and file bugs. |
| 19:34 | divya | hides from everyone |
| 19:34 | <dglazkov> | :) |
| 19:44 | <TabAtkins> | Philip`: I can't find a link to your demonstration of a timing-channel attack on canvas to extract font data. |
| 19:44 | <Ms2ger> | foreignContent? |
| 19:44 | <TabAtkins> | Ms2ger: Yes. |
| 20:00 | <tantek> | TabAtkins - everytime I see <foreignElement> I hear the character "Mo" from Wall•E saying "FOREIGN ELEMENT" (in the same tone of voice that it said "FOREIGN CONTAMINANT"[1]) [1] http://www.youtube.com/embed/l6gZ9Sf_0HU |
| 20:01 | <TabAtkins> | Hehe |
| 20:09 | <Philip`> | TabAtkins: http://philip.html5.org/demos/canvas/font-timing.html |
| 20:09 | <Philip`> | TabAtkins: (Also related: http://philip.html5.org/demos/canvas/img-timing-1.html http://philip.html5.org/demos/canvas/img-timing-2.html http://philip.html5.org/demos/canvas/img-timing-3.html ) |
| 20:10 | <Philip`> | (Some only work in some browsers) |
| 20:12 | Philip` | wonders if anyone has tried fixing canvas timing attacks yet, or at least filed bugs about it |
| 20:12 | <Ms2ger> | Mozilla has a bug, at least |
| 21:17 | <zewt> | http://meobets.com/~bai/google-font-fail.png really, google? |
| 21:19 | <TabAtkins> | TEXT SO TINY |
| 21:19 | <Ms2ger> | "Dismiss" is clearly the right response |
| 21:19 | <zewt> | should say "dismiss angrily" |
| 21:22 | <miketaylr> | i get that warning on my wife's chromebook too, at 100% |
| 21:22 | <Ms2ger> | That's why you shouldn't get chromebooks |
| 21:23 | <zewt> | google products not working well on google products is ... heh |
| 21:23 | <zewt> | very corporate? |
| 22:05 | gsnedders | blinks |
| 22:05 | <gsnedders> | A new XSLT bug report. |
| 22:05 | <gsnedders> | Wow. |
| 22:06 | <divya> | from who |
| 22:06 | <gsnedders> | No idea. |
| 22:11 | <jgraham> | They reported XSLT as a bug? |
| 22:13 | <miketaylr> | :D |
| 22:14 | <divya> | :D |
| 22:15 | <hober> | jgraham++ |
| 22:30 | <gsnedders> | jgraham: Sadly not. |
| 22:30 | <gsnedders> | jgraham: On the face of it an HTML5 parser bug, so I'm blaming you. |
| 22:33 | <jgraham> | You could do something more constructive and CC me on the bug instead :) |
| 22:34 | <gsnedders> | jgraham: I was in the process of doing so |
| 22:39 | <jgraham> | Oh, that's totally a HTML parser bug, yes |
| 22:39 | <gsnedders> | Presumably caused by namespacing HTML elements now |
| 22:42 | <jgraham> | Hmm, I'm not sure why you think that |
| 22:43 | <Philip`> | When in doubt, blame namespaces |
| 22:44 | <jgraham> | Who is to blame for XSLT in the platform anyway? |
| 22:44 | <gsnedders> | Oh, I'm being stupid. |
| 22:47 | <gsnedders> | annevk5: Or html5 should use data-anolis-xref or whatever the attribute is called |
| 22:50 | <jgraham> | It seems Peter Van der Beken did the Mozilla implementation in 2000. Dunno if they were first or not |
| 22:51 | <jgraham> | Oh, IE had it in 1998 |
| 22:51 | <gsnedders> | Yeah, IE was first. |
| 22:51 | <gsnedders> | Mozilla was still too busy, well, rewritting the whole browser. |
| 22:51 | <gsnedders> | *rewriting |
| 22:52 | <jgraham> | This isn't a prize you want to *win* |
| 22:59 | <Hixie> | oh hey i didn't know about [ArrayClass] |
| 22:59 | <Hixie> | can i sprinkle that about? |
| 23:08 | <heycam> | Hixie, feel free -- but you have to do the work of index getters and a length attribute yourself |
| 23:27 | <heycam> | Hixie, the term "relevant namespace object" is gone now that modules have been removed |
| 23:27 | <Hixie> | heycam: well i'm doing that work already anyway |
| 23:28 | <Hixie> | heycam: can you file a bug on me to fix the use of the term? |
| 23:28 | <heycam> | ok |
| 23:28 | <Hixie> | thanks |
