| 00:00 | <annevk> | I think the switch to relying on DOM Core was intended to be more than a reference switch, but I guess we can file specific bugs |
| 00:06 | <Hixie> | annevk: i'm going through http://www.w3.org/Bugs/Public/show_bug.cgi?id=11305 now |
| 00:08 | <annevk> | cool |
| 00:11 | <Hixie> | annevk: how do i file dom core bugs |
| 00:12 | <annevk> | WebApps WG, component DOM Core |
| 00:13 | <annevk> | http://www.w3.org/Bugs/Public/enter_bug.cgi?product=WebAppsWG |
| 00:13 | <Hixie> | ta |
| 00:15 | <annevk> | reading DOM Range it does look like it would be good to have inside DOM Core |
| 00:20 | <annevk> | DocumentFragment sucks |
| 00:21 | <annevk> | well, insertion methods combined with it suck |
| 00:21 | <annevk> | they are build for a single Node, but DocumentFragment can hold a bunch of them |
| 00:21 | <timeless> | heycam: ping |
| 00:21 | <heycam> | timeless, pong |
| 00:21 | <timeless> | heycam: it'd be *incredibly* beneficial for me to be able to link to #whatever-undefined |
| 00:22 | <timeless> | to be able to talk about `undefined` |
| 00:22 | <timeless> | WebIDL doesn't seem to have any anchors for `undefined` |
| 00:22 | <timeless> | it just has <span class="esvalue">undefined</span> |
| 00:22 | <heycam> | undefined in what ense? |
| 00:22 | <heycam> | the JS value undefined? |
| 00:22 | <timeless> | yeah |
| 00:22 | <timeless> | i want to reply to a spec which says: |
| 00:22 | <heycam> | why do you need to link to it? :) |
| 00:22 | <timeless> | foo.STUPID_CONSTANT |
| 00:22 | <timeless> | > This attribute is optional. |
| 00:23 | <timeless> | "Ok, does that mean it's http://w3/WebIDL/#undefined ?" |
| 00:23 | <heycam> | ok. so I don't know if there really should be a #undefined... it's purely a ECMAScript-defined thing |
| 00:23 | timeless | nods |
| 00:23 | <timeless> | i know |
| 00:23 | <heycam> | just like a don't have a link for Number |
| 00:23 | <timeless> | well |
| 00:24 | <timeless> | i'm 99% certain CSS has a link for Number :) |
| 00:24 | <timeless> | and actually, the same applies for Number.. |
| 00:24 | <timeless> | it'd be really useful to be able to link to those things |
| 00:24 | <heycam> | do you just want an anchor somewhere in the #es-attribute and #es-constant sections? |
| 00:24 | <heycam> | no actually I wouldn't even want to say anything about undefined there in webidl... |
| 00:25 | <timeless> | defintely not that section |
| 00:25 | <timeless> | you could just add <undefined> into http://dev.w3.org/2006/webapi/WebIDL/#es-type-mapping |
| 00:25 | <timeless> | it's technically not a type |
| 00:25 | <timeless> | but it'd sure make my life simpler |
| 00:26 | <timeless> | (stick it near void and explain that it really devolves to void...) |
| 00:26 | <heycam> | well, that's only true for operation return values |
| 00:26 | timeless | nods |
| 00:26 | <heycam> | I'm not actually sure where it would go, or what the sentence might say |
| 00:26 | <timeless> | i don't have a good suggestion |
| 00:26 | <timeless> | but basically, this spec is probably supposedly referencing webidl |
| 00:27 | <Hixie> | timeless: why don't you link to the ES spec instead? |
| 00:27 | <timeless> | Hixie: offhand i don't think webidl links to it |
| 00:27 | <timeless> | which means i probably can't find it :) |
| 00:27 | <timeless> | oh, it does |
| 00:27 | <timeless> | but not where i was looking |
| 00:27 | <Hixie> | http://people.mozilla.org/~jorendorff/es5.html |
| 00:27 | <timeless> | right, the fact that es is a pdf |
| 00:28 | <TabAtkins_> | Hixie: Yes, I've gone through it all. |
| 00:28 | <timeless> | is a pretty big reason for me not linking to it... |
| 00:28 | <timeless> | i guess http://people.mozilla.org/~jorendorff/es5.html#sec-4.3.9 isn't a terrible link |
| 00:28 | <Hixie> | TabAtkins_: awesome, more e-mail i can delete! |
| 00:28 | <timeless> | asside from being incredibly unofficial |
| 00:29 | <annevk> | timeless, there's also http://es5.github.com/ |
| 00:29 | timeless | needs to seriously consider changing the primary display (or at least the one that has the taskbar) to be the right one |
| 00:29 | <annevk> | timeless, see http://platform.html5.org/ |
| 00:29 | timeless | keeps hovering the mouse over the right edge of the left display (roughly center of screen) and having w7 hide all windows |
| 00:30 | <timeless> | annevk: hrm, http://es5.github.com/#x4.3.9 isn't terrible |
| 00:30 | <timeless> | not great |
| 00:30 | <timeless> | mostly terrible :) |
| 00:30 | <timeless> | an anchor that has the word `undefined` in it is much better than a random number :) |
| 00:31 | <annevk> | yeah dunno |
| 00:31 | <annevk> | standards organizations are generally pretty crappy |
| 00:31 | <timeless> | +1 |
| 00:31 | <annevk> | ECMA guys publish in PDF |
| 00:31 | <annevk> | IETF in plain text |
| 00:31 | <annevk> | W3C in HTML4 |
| 00:31 | <timeless> | with a req for XHTML, no? :) |
| 00:31 | <annevk> | none of them really seems to care about the web much |
| 00:31 | <timeless> | +1 |
| 00:32 | <Philip`> | WHATWG publishes in pages that crash your browser |
| 00:32 | <timeless> | +1 |
| 00:32 | <timeless> | Philip`: to be fair, i have a number of those myself |
| 00:32 | <jcranmer> | RFC in XML |
| 00:32 | <timeless> | the sad thing is that my aunt has one |
| 00:32 | <Philip`> | and with a zillion different variants so you can never tell which one has which feature |
| 00:32 | <jcranmer> | well, the RFCs are writtin in XMLs |
| 00:32 | <heycam> | timeless, your aunt has an RFC? |
| 00:32 | <jcranmer> | and published as plain text and html |
| 00:32 | <timeless> | heycam: no, a page she publishes that crashes a browser |
| 00:33 | <timeless> | https://bugs.meego.com/show_bug.cgi?id=18260#c8 |
| 00:33 | <timeless> | jcranmer: yeah, i know |
| 00:33 | <timeless> | my colleague (one door over) is writing an RFC :) |
| 00:33 | timeless | keeps trickle feeding him comments |
| 00:33 | <jcranmer> | timeless: if I make a browser that loads every page by doing *(int*)0 |
| 00:33 | <jcranmer> | everybody has a page that crashes a browser! |
| 00:34 | <timeless> | jcranmer: yeah well |
| 00:34 | <timeless> | we tried suggesting to her that perhaps such a long page wasn't a good idea |
| 00:34 | <timeless> | she didn't buy the argument |
| 00:34 | <timeless> | and she definitely wasn't willing to buy the demo'd product |
| 00:34 | <timeless> | (it really really FAILed) |
| 00:35 | <timeless> | heycam: anyway, i think i'd still rather have something in webidl |
| 00:35 | <timeless> | even if it's just a table listing the actual sections in ES |
| 00:35 | <timeless> | it'd be a public service :) |
| 00:36 | <timeless> | (useful named anchors, what a concept!) |
| 00:36 | <heycam> | timeless, send a mail to public-script-coord then, I'll think about it :) |
| 00:37 | <timeless> | ok |
| 00:37 | <jamesr> | ecmascript 'undefined' is reassignable |
| 00:37 | <timeless> | oh, i saw that you posted a list of known emails :) |
| 00:37 | <timeless> | jamesr: yeah, i know |
| 00:37 | <timeless> | one of my favorite features |
| 00:38 | <Hixie> | annevk: is XML doc vs HTML doc affecting "the case-sensitivity of some selectors" no longer true? |
| 00:38 | <heycam> | void 0 ftw |
| 00:38 | <timeless> | yep |
| 00:40 | <annevk> | Hixie, that is still true in general, but we are trying to change a few things |
| 00:40 | <Hixie> | k |
| 00:41 | <annevk> | Hixie, in HTML type and attribute names are case-insensitive |
| 00:41 | <annevk> | Hixie, I'm trying to get the magic list for HTML attributes whose values are matched case-insensitively removed |
| 00:42 | <Hixie> | i love how dom core takes all the easy stuff and leaves all the complicated stuff |
| 00:42 | <timeless> | Hixie: someone had to claim the easy stuff first! |
| 00:42 | <timeless> | (that reminds me of Nokia) |
| 00:42 | <Hixie> | i claimed it first! |
| 00:42 | <Hixie> | then they stole it! |
| 00:42 | <Hixie> | :-P |
| 00:42 | <annevk> | Hixie, what complicated stuff? |
| 00:43 | <Hixie> | like the document's address, and the reload override flag |
| 00:43 | <annevk> | we could do document URL I suppose but you would have to set it correctly etc. |
| 00:44 | <Hixie> | the less you take the easier it is for me :-P |
| 00:44 | <annevk> | it would just be a concept, just like charset / media type |
| 00:44 | <timeless> | aww, this spec didn't have any spelling errors |
| 00:44 | <timeless> | that's no fun |
| 00:44 | <timeless> | well, they misspelled JavaScript once |
| 00:44 | <timeless> | but everyone does that :) |
| 00:44 | <annevk> | document URL is an open issue |
| 00:45 | <annevk> | I raised it on the WHATWG list |
| 00:45 | <timeless> | hrm |
| 00:45 | <timeless> | anyone here understand how paste works? |
| 00:45 | <Hixie> | annevk: some of this stuff is kinda making the spec harder to understand. e.g. not mentioning the HTML parser is what puts a doc into quirks mode |
| 00:46 | <timeless> | whenever i paste something from Notepad into IE/Firefox on pastebin.mozilla.org |
| 00:46 | <Hixie> | annevk: and removing the examples in HTMLCollection |
| 00:46 | <timeless> | I get doubled lines for lines I write |
| 00:46 | <timeless> | but only single lines for lines that have > |
| 00:46 | <annevk> | Hixie, I did not move that stuff, but we can certainly add some examples |
| 00:47 | <timeless> | http://www3.ttc.ca/ |
| 00:47 | timeless | chuckles |
| 00:47 | timeless | wonders why ttc did that |
| 00:47 | <annevk> | Hixie, as well as mentioning how quirks mode is set; the reason we moved it is because it's global |
| 00:47 | <annevk> | if you feel any of this is on the wrong side we can certainly reevaluate |
| 00:48 | <Hixie> | annevk: um, you say the default character encoding is UTF-8 |
| 00:48 | <Hixie> | but it's UTF-16 |
| 00:48 | <annevk> | but this allows CSS to hook into the DOM without having to hook into HTML |
| 00:48 | <Hixie> | i think it's the right side |
| 00:48 | <Hixie> | i just think moving it is making the specs harder to read |
| 00:48 | <Hixie> | ideally we would just have one spec and it wouldn't be an issue! :-) |
| 00:49 | <annevk> | 'If the attribute is present in an XML document, its value must be an ASCII case-insensitive match for the string "UTF-8"' |
| 00:49 | <annevk> | "utf-16" (including quotes) yields no results in HTML |
| 00:50 | <Hixie> | ? |
| 00:50 | <Hixie> | the document's character encoding defaults to UTF-16 in WA1 and to UTF-8 in DOM Core |
| 00:50 | <Hixie> | i believe UTF-16 is the correct default |
| 00:50 | <annevk> | never mind me |
| 00:52 | <Hixie> | annevk: should i file a bug? |
| 00:53 | <annevk> | w((document.implementation.createHTMLDocument("test")).characterSet) gives UTF-8 in Gecko |
| 00:56 | <Hixie> | i'll have to test it more later |
| 00:56 | <Hixie> | i gotta go |
| 00:59 | <annevk> | k, I tried testing it a bit more, but I'm getting a bunch of undefined things for these attributes |
| 01:02 | <annevk> | basically it's not DocumentFragment so much as it is having these methods available on Document |
| 01:56 | tantek | notices discussion about fullscreen API (cc: Hixie annevk cpearce) |
| 02:09 | <tantek> | Hixie, I've got it in my queue after CSS3-UI: https://wiki.mozilla.org/Tantek-Mozilla-projects#Full_Screen |
| 02:10 | <tantek> | assuming I get permission to publish it with a CC0+OWFa license, where would it go on the whatwg site? |
| 02:46 | <tantek> | until there's an actual spec draft - feel free to edit / add to https://wiki.mozilla.org/Gecko:FullScreenAPI and I'll incorporate from there. |
| 03:30 | <mahir256> | hello? |
| 04:16 | <Hixie> | tantek: we don't have a system for publication yet but we can figure something out |
| 04:16 | <Hixie> | tantek: easiest for me is to just give you a subdomain, e.g. css.whatwg.org |
| 04:19 | <tantek> | Hixie - I suppose I should research how whatwg specs are organized. |
| 04:19 | <Hixie> | there is no pattern |
| 04:19 | <Hixie> | right now there's only one spec, and it keeps changing :-) |
| 04:20 | <tantek> | for some reason I thought there were multiple specs and editors :) |
| 04:20 | <Hixie> | not currently |
| 04:21 | <Hixie> | though we keep talking about it |
| 04:21 | <tantek> | what about the spec that Aryeh was saying he's working on |
| 04:21 | <Hixie> | usually around the same time an editor who mainly works at the w3c gets frustrated with the w3c :-) |
| 04:21 | <Hixie> | aryeh's stuff was intended to get merged into the html spec |
| 04:21 | <tantek> | oh ok |
| 04:21 | <Hixie> | not sure what we'll end up doing |
| 04:21 | <tantek> | interesting - so this is a bit of a novel situation :) |
| 04:21 | <Hixie> | there's a lot of playing by ear here :-) |
| 04:22 | <tantek> | makes sense :) |
| 04:23 | <Hixie> | process is for weenies :-P |
| 04:27 | <tantek> | heh - I suppose that makes me a bit of a process weenie then: http://microformats.org/wiki/process |
| 04:27 | <Hixie> | :-) |
| 04:28 | <tantek> | Hixie, in other news, it looks like I've picked up the task to integrate / update your work on hit-testing into the CSS3-UI spec. |
| 04:28 | <tantek> | "your work" meaning this email message: http://lists.w3.org/Archives/Public/www-style/2009Feb/0287.html |
| 04:29 | <Hixie> | that was mostly the work of someone else at opera iirc |
| 04:29 | <Hixie> | but man, that was a while ago |
| 04:31 | <tantek> | somewhere I've seen somebody say something about things just taking longer. |
| 04:31 | <Hixie> | hehe |
| 04:32 | <tantek> | if you have any updates on hit-testing, let me know - I'll be starting from that email message, and then analyzing this set of notes from Leif about interop challenges: http://lists.w3.org/Archives/Public/www-style/2010Aug/0407.html |
| 04:32 | <Hixie> | i'm happy to say that i know nothing new |
| 04:32 | <tantek> | good to know. |
| 04:34 | <tantek> | well at least your assumption that "the CSS specifications will at some future point deal with the issue" is being realized (regarding hit-testing) |
| 04:34 | <Hixie> | heh |
| 04:34 | <tantek> | CSS3-UI seems like the logical place to document hit-testing, and there's consensus on that from CSS/SVG WGs. |
| 04:34 | <Hixie> | cool |
| 04:35 | <tantek> | or perhaps just that no-one else wanted to pick it up, and I was foolish enough to do so. |
| 04:36 | <Hixie> | i know _that_ feeling |
| 04:36 | <tantek> | lol |
| 04:37 | <tantek> | I definitely had that feeling of, what, *this* is not defined? Then asked "How is this not an essential part of the open web platform?" (silence) |
| 04:38 | <tantek> | And then I asked why it wasn't defined in HTML5 (what I would have expected) which then led to your email message. |
| 04:38 | <tantek> | And the subsequent realization that there was only one logical course of action. |
| 04:38 | <Hixie> | in a somewhat futile attempt to scope the work i have on my plate i've drawn a line around anything presentational and have left that for, well, apparently, you. |
| 04:42 | tantek | has a feeling he'll be spending more time in #whatwg. |
| 05:23 | <jamesr> | annevk: yt? what's the preferred URL to cite for DOMCore (assuming i can w3 bullshit, for the moment) |
| 05:24 | <annevk> | http://dvcs.w3.org/hg/domcore/raw-file/tip/Overview.html until we reach a more stable state |
| 05:29 | jamesr | global search-replaces links to /TR/2004/REC-DOM-Level-3-Core-20040407/ |
| 05:29 | <annevk> | which spec is this? |
| 05:29 | <annevk> | or is this source code? |
| 05:29 | <jamesr> | http://dvcs.w3.org/hg/webperf/raw-file/tip/specs/RequestAnimationFrame/Overview.html |
| 05:30 | <jamesr> | will update in a sed |
| 05:30 | <jamesr> | sec |
| 05:33 | <annevk> | aah, roc's API |
| 05:33 | <annevk> | nice |
| 05:33 | <jamesr> | yeah it's a good one. just have to write it down concretely enough :) |
| 05:35 | <jamesr> | another noob question: when citing something from whatwg HTML, do you link to the multipage version or the single page version? |
| 05:35 | <jamesr> | in the spec itself all internal links go to the same version |
| 05:35 | <jamesr> | but you can't pull that trick from an external reference. i know hixie would say it should all be in the same spec :) |
| 05:36 | <annevk> | I used http://www.whatwg.org/html for now |
| 05:37 | <annevk> | you can append #fragments to the end of that URL and redirects will happen automatically |
| 05:37 | <jamesr> | ah cool, so since the current default is multipage it'll go to the multipage version? |
| 05:37 | <annevk> | yes that will always go to the multipage |
| 05:37 | <annevk> | see e.g. on http://dev.w3.org/csswg/cssom-view/ |
| 05:38 | <annevk> | search for "queue a task" or some such |
| 05:39 | <jamesr> | http://www.whatwg.org/html/#queue-a-task. any reason for that over http://www.whatwg.org/html#queue-a-task or http://whatwg.org/html#queue-a-task ? |
| 05:39 | <annevk> | no |
| 05:40 | <annevk> | it's just what our scripts outputs |
| 05:40 | <annevk> | we generate those links automatically |
| 05:40 | <annevk> | see https://bitbucket.org/ms2ger/dom-core for the setup |
| 05:40 | <annevk> | Anolis is quite nice for HTML specs |
| 05:45 | <jamesr> | my problem is i only get a few hours every few weeks to work on actually editing specs, which isn't enough time to get up to speed and stay up to speed with how to do it efficiently |
| 05:46 | <annevk> | fair enough |
| 05:46 | <annevk> | I will let you know when we made this easy enough so you can be set up in a couple of minutes |
| 05:47 | <annevk> | (which probably will be a while) |
| 05:48 | <Hixie> | jamesr: that's how i feel about coding :-) |
| 05:50 | <jamesr> | annevk: well for this spec there's currently an xml+makefile+xslt setup that heycam got working, and that seems to do something |
| 05:50 | <jamesr> | i guess i could just start with the .html output of that and then figure out how to munge it into the input anolis expects |
| 05:54 | <annevk> | it's what HTML, XHR, DOM Core, etc. are all using, but if the current setup works than it might not be worth the trouble if you have that little bandwidth |
| 05:56 | <jamesr> | is the input format documented anywhere? |
| 05:56 | <jamesr> | man, do i have to clone https://bitbucket.org/ms2ger/anolis/src in order to view README.html? |
| 05:57 | <annevk> | jamesr, we have https://bitbucket.org/ms2ger/spec-template |
| 06:00 | <annevk> | actually, that might be out of date |
| 06:01 | <annevk> | easiest is just looking at Overview.src.html files |
| 06:35 | <annevk> | what is so hard about specifying commonAncestorContainer ? |
| 06:35 | <annevk> | other than that the Range specification looks really nice |
| 06:56 | <annevk> | http://oreilly.com/catalog/0636920021360 |
| 06:56 | <annevk> | via myakura on G+ |
| 06:56 | <annevk> | book on HTML5 FileSystem API |
| 06:56 | <annevk> | (not that there's such a thing in HTML) |
| 07:28 | <zewt> | 74 pages isn't a "book", heh |
| 07:59 | <hsivonen> | http://www.jenitennison.com/blog/node/160#comment-11188 |
| 09:15 | <hsivonen> | so, it looks like multiple people are so far upset about AryehGregor not submitting the HTML Editing API spec to the W3C and are attributing AryehGregor's email to the "WHATWG" |
| 09:35 | <jgraham> | All Fear The Cabal! |
| 09:44 | <jgraham> | But seriously, unless people get over the WHATWG-as-conspiracy rhetoric and actually look at real issues (W3C Process having a big impedence mismatch with the needs of authors and implementors, W3C high level goals as expressed through e.g. focus of the TAG being very different to the needs of its core consituents), this kind of conflict is going to continue |
| 09:46 | <espadrine`> | hsivonen: I think his choice was reasonable. If you refuse Minix' license, create Linux. If you refuse W3C's license, create stuff your way and choose a different license for it. I guess I don't really understand the fuss around Aryeh Gregor's actions. |
| 09:49 | <hsivonen> | jgraham: what I read into AryehGregor's email was that he needs to become an employee (of Google) in order be able to use time to deal with the W3C |
| 10:09 | <jgraham> | hsivonen: The reason doesn't seem that important. The fact is that someone has done a bunch of difficult work that has the potential to enhance the web platform and people are getting upset because that person isn't willing to do even more work to deal with Process *even though* he has set things up so that anyone who feels strongly about Process can do that part of the work instead |
| 10:10 | <jgraham> | I can't understand the sense of entitlement one needs to have an attitude like that |
| 12:43 | <Dashiva> | Is _anything_ really a natural fit for XML? |
| 12:43 | <Ms2ger> | Pain. |
| 12:44 | <Dashiva> | Obviously you can fit almost anything into XML, but claiming something is a natural fit seems like a strong claim |
| 12:49 | <jgraham> | Well if you have data with just the right level of structure (more than key-value pairs but less than a graph), don't care about representation efficiency and don't care that small errors in the input will break everything, I guess it isn't a bad fit |
| 12:53 | <Philip`> | The alternatives are generally worse, though |
| 12:53 | <jgraham> | For which usecase? |
| 12:56 | Philip` | can't remember what he was thinking now |
| 12:59 | <hsivonen> | Dashiva: DocBook? |
| 13:07 | <JirkaK> | Highly structured documents, that's for what SGML and XML was invented and is natural fit. Today most known and used formats are DocBook, DITA and TEI |
| 13:09 | <Dashiva> | Such documents often have needs that translate into overlapping markup |
| 13:11 | <karlcow> | Dashiva: I think there is also a question of culture (habits). I'm more used to XML toolchain, so it is a lot easier for me to benefit from the "strictness" of XML than trying to reinvent it with the DOM. |
| 13:11 | <hsivonen> | JirkaK: who actually uses DITA these days (except for Eliot Kimber)? |
| 13:11 | <JirkaK> | No, unless you want to do change tracking in elegant way |
| 13:11 | <hsivonen> | is there an elegant way to de change tracking? |
| 13:12 | <karlcow> | Is there an elegant way? (Reductio ad absurdum) |
| 13:13 | <JirkaK> | hsivonen: A lot of software companies are authoring their help systems in DITA. In past years I have seen many companies to migrate from "legacy" systems like FrameMaker to DITA. There was a lot of marketing behind DITA in past years, so it got more traction in industry then DocBook recently. |
| 13:14 | <JirkaK> | probably most elegant way of change tracking is to use plain text and some versioning system ;-) |
| 13:15 | <hsivonen> | JirkaK: interesting. I was unaware of that kind of usage. |
| 13:15 | <hsivonen> | (help system that is. not plain text) |
| 13:16 | <JirkaK> | hsivonen: a lot of XML editors now provide out-of-the box configuration from DITA so it is fairly easy to author in DITA |
| 13:37 | <david_carlisle> | hsivonen: and as you'll see in Norm W's recent posts dita mindset is pushing docbook developments to a certain extent |
| 13:39 | <david_carlisle> | Dashiva: we use xml for essentially everything here (documentation, code generation, fun, ...) |
| 13:39 | <jgraham> | david_carlisle: You also use Fortran :p |
| 13:39 | <david_carlisle> | jgraham: yes but of course we generate it from xml |
| 13:40 | <david_carlisle> | why can't the whole world be like this? |
| 13:40 | <hsivonen> | david_carlisle: I can't tell if you are joking now. are you? |
| 13:40 | <david_carlisle> | no |
| 13:40 | <jgraham> | My humor detector is oscillating wildly |
| 13:40 | <Ms2ger> | Generating Fortran from XML? |
| 13:40 | <jgraham> | I really couldn't tell at all |
| 13:40 | <david_carlisle> | yes (and the other way actually) |
| 13:41 | <david_carlisle> | we have xml saying what the interface should be and we generate fortran, c, c#, matlab, r, python, whatever from that |
| 13:41 | <JirkaK> | using XSLT I guess? |
| 13:42 | <david_carlisle> | what else? 9although some of the youngsters use python) |
| 13:43 | <Ms2ger> | You make me want to jump into a lake, but I'm afraid annevk5 wouldn't let me out :) |
| 13:45 | <david_carlisle> | looking at people transform documents using DOM scripting has a similar effect on me, it works but it's fragile and painful compared to using a language designed for the job |
| 13:48 | <hsivonen> | foolip: is it OK to put the email address you are using in the validator.nu bugzilla in the hg user field? |
| 13:48 | <foolip> | hsivonen, philip at foolip dot org, yes |
| 13:48 | <foolip> | this wasn't work, after all |
| 13:48 | <hsivonen> | foolip: thanks |
| 13:49 | <foolip> | does that mean I'll have an account on the mercurial server, or just that the commit says so? |
| 13:49 | <Ms2ger> | The latter |
| 13:49 | <foolip> | ok, like git then |
| 13:49 | karlcow | has the same feeling than david_carlisle… culture I guess. |
| 13:50 | <foolip> | it turns out there's a lot of friction when moving from git to hg, all the little tings are different :) |
| 13:50 | <foolip> | I imagine the reverse is equally true |
| 13:50 | <Ms2ger> | Git? What's that? :) |
| 13:51 | <foolip> | Ms2ger, https://twitter.com/#!/agnoster/status/44636629423497217 |
| 13:54 | <jgraham> | Are those just difficult sounding words or would it actually mean something if I understood the sentence properly? |
| 13:54 | <foolip> | jgraham, it's technobabble :) |
| 13:54 | <Ms2ger> | A Hilbert space? Come on. |
| 13:55 | <jgraham> | Well pure mathematics is often surprising |
| 13:56 | <Ms2ger> | To physicists, you mean ;) |
| 13:56 | <jgraham> | Possibly :) |
| 13:58 | <Ms2ger> | Did people see http://tools.ietf.org/html/draft-freed-media-type-regs-00, btw? |
| 13:58 | <jgraham> | It did sound rather like some of the explainations of monads you hear from hardcore haskellers |
| 13:59 | <Philip`> | http://tartley.com/?p=1267 too |
| 14:15 | <jgraham> | I still want a proper code review system for hg |
| 14:15 | <Ms2ger> | Don't we all? |
| 14:16 | <jgraham> | Well I'm surprised Mozilla haven't written one already |
| 14:16 | <Ms2ger> | We use bugzilla |
| 14:16 | <jgraham> | I know |
| 14:17 | <jgraham> | I'm surprised Mozilla haven't written one already |
| 14:18 | <hsivonen> | jgraham: gecko developers do stuff the way it has been done since the CVS days and Labs, Webdev, etc. use github |
| 14:19 | <hsivonen> | these days, b.m.o has a link to something called "splinter reviews" but I've never used the newfangled thing |
| 14:21 | <hsivonen> | G+'s real names policy has interesting effects on people whose name doesn't use the Latin script |
| 14:22 | <hsivonen> | do you use the native script, the Romanization of your native name or the (if you have one) your English alias? |
| 14:22 | <jgraham> | hsivonen: One reason I am surprised is that I occasionally see people complain about you having difficulties retaining new contributers, and the review system is often cited as a reason |
| 14:22 | <jgraham> | In particular the need to find the right person to review |
| 14:23 | <hsivonen> | jgraham: yeah, Mozilla isn't good at fixing its problems with attracting and retaining contributors |
| 14:23 | <jgraham> | But that is a purely technical problem that you could solve with software |
| 14:24 | hsivonen | notices 村田真's "Bragging Rights" https://plus.google.com/u/0/110720658734820520378/about |
| 14:28 | JirkaK | Murata-san? |
| 14:29 | <hsivonen> | JirkaK: so it seems |
| 14:30 | <Dashiva> | Yes |
| 14:49 | <timeless> | jgraham: which is purely technical? |
| 14:56 | <jgraham> | timeless: Finding the right reviewer for a change |
| 14:56 | <jgraham> | In general making reviews low-pain |
| 14:56 | <jgraham> | Both for the reviewer and reviewee |
| 14:56 | <timeless> | yeah, i think i outlined how to do that about 8 years ago |
| 14:57 | <timeless> | (programmatically) |
| 14:57 | <timeless> | it was never implemented |
| 14:57 | <timeless> | there's actually more than enough data to figure it out |
| 14:59 | <smaug____> | I haven't seen any good way to do reviews |
| 14:59 | <smaug____> | s/way/software/ |
| 14:59 | <timeless> | oh, i'm just talking about how to find the right reviewer |
| 15:00 | <timeless> | the software i've seen that tries to actually "enable" reviewing itself gives me headaches |
| 15:01 | <timeless> | http://www.usatoday.com/news/nation/2011-07-27-ny-thruway-prayers_n.htm |
| 15:01 | smaug____ | needs to close some of his 200+ tabs :/ |
| 15:02 | <timeless> | smaug____: heh |
| 15:02 | timeless | ponders |
| 15:02 | <timeless> | i have no idea how many tabs i have |
| 15:02 | <timeless> | i'm using panorama which makes counting tabs *hard* |
| 15:02 | <jgraham> | smaug____: The system we have is pretty good. I think Gerrit is a bit similar |
| 15:03 | jgraham | discovers hg-review |
| 15:03 | <jgraham> | This looks like it might not obviously be awful |
| 15:03 | <timeless> | oh yeah, 've seen that mentioned |
| 15:03 | <zewt> | quite a standard to set for software |
| 15:03 | <timeless> | i've never tried it |
| 15:03 | timeless | should |
| 15:04 | <timeless> | there's actually a hg repo called `all-extensions` iirc |
| 15:04 | timeless | needs to grab it |
| 15:04 | <timeless> | https://bitbucket.org/abuehl/allextensions |
| 15:07 | <hsivonen> | sadly, reviews don't really scale |
| 15:08 | <timeless> | ? |
| 15:08 | <hsivonen> | you can hire more people to write more code but you can't hire more reviewers |
| 15:08 | <timeless> | well |
| 15:08 | <timeless> | you could hire the people who used to code for you |
| 15:08 | <timeless> | moco chose not to do that |
| 15:09 | <hsivonen> | timeless: :-( |
| 15:09 | <zewt> | ... and you want people experienced with your code to review, but people often aren't going to view being moved from development to reviewing as a promotion, heh |
| 15:09 | <smaug____> | browser vendors should hire some more spec reviewers ;) |
| 15:23 | <timeless> | smaug____: that happened to me 2 months ago :) |
| 15:23 | <timeless> | but yeah |
| 15:23 | timeless | should provide a place where people could suggest specs to review |
| 15:23 | <timeless> | hrm, if people want, please feel free to comment on my G+ page with suggestions :) |
| 15:24 | <zewt> | too bad g+ is broken : | |
| 15:24 | <smaug____> | timeless: specs to review? |
| 15:25 | <smaug____> | like all of them? |
| 15:25 | <zewt> | any new google product that doesn't work with gapps is just poor and broken |
| 15:25 | <timeless> | smaug____: a prioritized list with links is appreciated |
| 15:25 | <smaug____> | yeah |
| 15:25 | <timeless> | gathering a list of specs is actually moderately painful |
| 15:25 | <timeless> | i've tried a couple of times |
| 15:25 | <timeless> | i keep giving up |
| 15:26 | <timeless> | there are probably hundreds of xml specs that i don't want to look at |
| 15:26 | <timeless> | as an example of things to disprove your "all of them" |
| 15:26 | <smaug____> | well, sure. All the relevant ones... |
| 15:27 | <timeless> | right... see "please give me a prioritized list with links" |
| 15:27 | <timeless> | seriously.. |
| 15:27 | <timeless> | technically bugzilla does a better job of this than standards bodies |
| 15:27 | <timeless> | i can go to bugzilla.mozilla.org/request.cgi and get a list |
| 15:28 | <timeless> | it isn't perfect |
| 15:28 | <timeless> | e.g. it'd be nice if i could sort by deadline |
| 15:28 | <timeless> | (which is something bmo doesn't have anyway, but which is definitely relevant for standards work) |
| 15:29 | <smaug____> | well, what does deadline mean in whatwg? |
| 15:30 | <smaug____> | (but sure, with w3c specs it would be useful) |
| 15:31 | <jgraham> | Thoughts on hg-review so far: Who the fuck thought that black on green was a nice colour scheme for displaying code? |
| 15:32 | <jgraham> | I guess it's because these were lines added |
| 15:32 | <zewt> | someone who uses a screen reader, i'd imagine |
| 15:32 | <jgraham> | But seriously it's ugly |
| 15:33 | <jgraham> | zewt: That would also explain the fact that it users overflow:hidden at (presumably) 80 columns |
| 15:33 | <zewt> | ... that would make me stop looking and move on, alone |
| 15:33 | <jgraham> | Well these are fixable issues |
| 15:34 | <zewt> | it speaks of a complete lack of experience with real code |
| 15:34 | <zewt> | 80 columns is used by two people: those who havn't changed their coding habits since the 90s, and people who religiously follow PEP-8 (and their overlap) |
| 15:34 | <jgraham> | If I can work out how to actually share review comments I might suggest using it |
| 15:34 | <zewt> | two classes of people, rather |
| 15:34 | jgraham | likes code that wraps at a sensibly small number of columns |
| 15:35 | <zewt> | me too, such as 120 :) |
| 15:35 | <jgraham> | It means I can have multiple side-by-side windows |
| 15:35 | <jgraham> | I find that people typically use 80 or "however wide my monitor is" |
| 15:36 | <zewt> | my terminal windows are around 240x70 |
| 15:37 | <zewt> | so having two windows would still make 120 columns sane; and even 100 columns is a lot less annoying than 80, which just causes too much line splitting |
| 15:41 | timeless | has at least 4 side by side windows |
| 15:41 | <timeless> | smaug____: anyway... |
| 15:42 | <timeless> | to prioritize for whatwg, i'd probably ask for things which people think are <small> or "urgent" or which vendors seem likely to try to implement "sooner" than others |
| 15:42 | <jgraham> | Oh, I found a scrollbar |
| 15:42 | <jgraham> | But still |
| 16:27 | <Ms2ger> | timeless, not sure if you saw platform.html5.org? It's not really what you need, but it's about the best list of web-relevant specs I've seen |
| 16:37 | <annevk> | didn't know foolip was patching Validator.nu, neat |
| 16:39 | <foolip> | annevk, I had nothing to do this weekend :) |
| 16:40 | <Ms2ger> | Could've written a spec :) |
| 16:41 | <timeless> | Ms2ger: yeah, i've seen it |
| 16:41 | <timeless> | i'm pretty sure it's unusable for my purposes |
| 16:41 | <timeless> | what i need is a list of things which i should be reviewing |
| 16:42 | <timeless> | things which are large and have lots of eyeballs and aren't moving don't count |
| 16:42 | <timeless> | (SVG comes to mind) |
| 16:42 | <hsivonen> | krijnh: the logs have oddities on /me lines |
| 16:42 | <dglazkov> | good morning, Whatwg! |
| 16:43 | <hsivonen> | dglazkov: good morning |
| 16:43 | timeless | looks for a definition of `cancelable` |
| 16:45 | <foolip> | it seems to me that writing specs on algorithm form Hixie-style is a lot like writing code without test compiling it, and then waiting for a very long time before you get the warnings/errors |
| 16:45 | <smaug____> | indeed |
| 16:46 | <smaug____> | though, what would be a better way to write specs? |
| 16:47 | <timeless> | compiling and testing? :) |
| 16:48 | <Ms2ger> | Then feel free to review DOM Core :) |
| 16:48 | <timeless> | heh |
| 16:48 | <timeless> | i will at some point |
| 16:49 | <timeless> | but i'd like to have a priority ordered list |
| 16:49 | <timeless> | if there are things i should look at first, i'd rather do that... |
| 16:49 | <annevk> | Ms2ger, btw, I'm pretty close to have xxxhierarchy sorted out |
| 16:49 | timeless | is currently looking at deviceorientation |
| 16:49 | <Ms2ger> | Yay! |
| 16:49 | <annevk> | Ms2ger, in a style that does not make use of the nodes model |
| 16:50 | <annevk> | making it much more explicit what actually needs to be done which seems somewhat better |
| 16:54 | <timeless> | Ms2ger: wrt seeing that host, it's in my w3 panorama stack (#5/7) |
| 17:22 | <AryehGregor> | Sigh, specdrama. |
| 17:23 | <AryehGregor> | hsivonen, I assume that if I wanted to use Google time to submit the spec to the W3C or whatever, Hixie would probably let me, but he didn't tell me to and I'm not interested in doing it unless I'm told to. |
| 17:57 | <annevk> | AryehGregor, it's the same people that always complain, regardless of what you do |
| 17:58 | <annevk> | AryehGregor, if Hixie made the HTML specification forty pages longer they would complain too |
| 17:58 | <AryehGregor> | annevk, I thought there have been a couple of very productive Google+ discussions lately. |
| 17:58 | <AryehGregor> | I'm trying to continue the trend. :) |
| 17:58 | <AryehGregor> | (but yeah) |
| 17:58 | <AryehGregor> | The only that would have made them happy would be to totally ignore the W3C and work only in the WHATWG, so they didn't know about it. |
| 18:01 | <annevk> | they would know about it |
| 18:01 | <AryehGregor> | They didn't know about it until I just announced it. |
| 18:01 | <AryehGregor> | Even though I announced it on the WHATWG list months ago. |
| 18:02 | <annevk> | Julian commented on the WHATWG list I thought |
| 18:03 | <AryehGregor> | You mean yesterday, or in general? |
| 18:03 | <annevk> | yesterday |
| 18:03 | <AryehGregor> | Yesterday he commented on public-html. |
| 18:03 | <AryehGregor> | On, of course, a purely procedural issue. |
| 18:04 | <annevk> | oh you're right |
| 18:28 | <AryehGregor> | Megapost written: https://plus.google.com/u/0/105458233028934590147/posts/h7nsT7wuNmX |
| 18:32 | Ms2ger | is perfectly happy to stay out of that discussion |
| 18:35 | <annevk> | I think you make some pretty good points in there |
| 18:35 | <annevk> | Both about power of editors and process issues with the W3C |
| 18:40 | <jgraham> | AryehGregor: Nice commen |
| 18:40 | <jgraham> | t |
| 19:08 | <hober> | AryehGregor: I'm sad that I'm limited to +1ing that once. |
| 19:11 | <annevk> | AryehGregor, you mind if I quote that verbatim on my blog with context? |
| 19:11 | <AryehGregor> | annevk, go ahead. |
| 19:11 | <AryehGregor> | I'm optimistic that we can win over some of the more reasonable people to a large degree, given a chance. |
| 19:12 | <AryehGregor> | Not the W3C itself, though. |
| 19:14 | <AryehGregor> | Maybe someone should also copy it to the discussion on the chairs list. The answer in the end is obviously going to have to be "we're not going to ever change things enough to satisfy him", but I imagine some of the participants there would find it informative. |
| 19:17 | tantek | tries to catch-up on the drama. |
| 19:17 | <AryehGregor> | Hixie, I'll respond to some of the DOM Range stuff, at least, since there's not much of it. |
| 19:19 | <AryehGregor> | Maybe some of the editing stuff too. |
| 19:20 | <annevk> | AryehGregor, was your editing work going to take in Selection in due course? |
| 19:20 | <AryehGregor> | annevk, yeah, that's the theory. |
| 19:20 | <AryehGregor> | I don't want to do it right now because I just asked for implementer review and don't want to add a massive chunk of spec text just now. |
| 19:20 | <AryehGregor> | I'll work on other things for the moment. |
| 19:23 | <annevk> | okay |
| 19:23 | <AryehGregor> | Unless you want to absorb DOM Range into DOM Core and this is blocking your work. |
| 19:23 | <tantek> | AryehGregor - what's the spec you're developing? |
| 19:23 | <AryehGregor> | tantek, http://aryeh.name/spec/editing/editing.html |
| 19:24 | <tantek> | AryehGregor - seems reasonable to me. |
| 19:25 | <timeless> | AryehGregor: so... |
| 19:25 | tantek | is confused by why there is any drama about this at all. |
| 19:25 | <timeless> | i'd like to note that there are too many mailing lists |
| 19:25 | <timeless> | and no useful announce lists |
| 19:25 | <timeless> | if you could have posted to one useful announce list |
| 19:25 | <tantek> | timeless - you can shorten that to: there are too many mailing lists |
| 19:25 | <tantek> | :) |
| 19:25 | <timeless> | and if i could have been subscribed to that one list |
| 19:25 | <timeless> | tantek: +1 |
| 19:25 | <timeless> | i'd probably be reviewing it now |
| 19:25 | <tantek> | timeless - I think that's called "choose your feeds wisely" :) |
| 19:25 | <timeless> | tantek: i haven't found any useful feeds! |
| 19:26 | <AryehGregor> | I could make a post on the WHATWG blog, if that would be helpful. |
| 19:26 | <timeless> | i've been soliciting for them all day |
| 19:26 | tantek | commends AryehGregor for his tireless hard work and working in the open with a liberal license. Keep up the good work. And don't waste time on the haters. |
| 19:26 | <timeless> | if you did it and got others to do the same, that'd encourage me to read that blog |
| 19:26 | <timeless> | so, +1 :) |
| 19:28 | <erlehmann> | a bit off-topic: is a „liberal“ a „free“ license (e.g. not proprietary) ? or does it mean MIT/X1-style (not copyleft) ? |
| 19:28 | <timeless> | s/X1/X11/ ? :) |
| 19:29 | <timeless> | i think it leans to MIT/.. |
| 19:32 | timeless | will +1 the actual plus comment later from home.. |
| 19:33 | <erlehmann> | he, i use „add“ for „+1“ to confuzzle my not-yet-deleted friends. |
| 19:33 | <erlehmann> | „i added that“ |
| 19:33 | <TabAtkins_> | "incremented" |
| 19:34 | <Ms2ger> | "inc'd" |
| 19:36 | <AryehGregor> | erlehmann, specs should be permissively licensed, like CC0 or MIT-style. Copyleft for specs is a bad idea, IMO. |
| 19:36 | <AryehGregor> | It means projects can't, e.g., include parts of it in their documentation unless their license happens to be compatible. |
| 19:36 | <timeless> | right |
| 19:36 | <AryehGregor> | Proprietary forks are not a big problem in practice for specs anyway. |
| 19:36 | <timeless> | similarly, examples shouldn't be copyleft |
| 19:37 | <annevk> | draft: http://annevankesteren.nl/2011/07/editor |
| 19:37 | <timeless> | heck, header files generally shouldn't be copy left |
| 19:37 | <annevk> | not sure about the title |
| 19:37 | <timeless> | (ignoring the minor detail that they generallly don't have sufficient substance to have anything for which one can claim copyright!) |
| 19:38 | <timeless> | annevk: is it unreasonable for me to wish your <editing APIs of the web platform> was a link? |
| 19:38 | <erlehmann> | AryehGregor, i fully agree with that. differentiation through implementations makes more sense. |
| 19:38 | <annevk> | timeless, no |
| 19:38 | <timeless> | :) |
| 19:39 | <timeless> | annevk: might as well try to get eyeballs for something useful after all :) |
| 19:40 | <erlehmann> | timeless, why that stuff with the header files? |
| 19:41 | <timeless> | erlehmann: if i want to be compatible with an api |
| 19:41 | <timeless> | why should the api be able to influence the license on *my* product? |
| 19:41 | <erlehmann> | isn't there at least one program that is free because of readline being GPL? |
| 19:41 | <timeless> | generally speaking commercial products get this right and ensure their licenses on headers are non tainting |
| 19:41 | <Ms2ger> | Obviously you lack a cite attribute |
| 19:41 | <timeless> | i'm sure there are a dozen crappy programs ou tthere |
| 19:42 | <timeless> | s/u t/ut / |
| 19:42 | <timeless> | but personally i don't believe that's a valid reason for anything |
| 19:42 | <timeless> | and besides, readline is a library, not just a header |
| 19:42 | <erlehmann> | so you are more talking about stuff susceptible to network effects. |
| 19:42 | <timeless> | if you want to require someone to have <foo-license> because they include <foo-library> |
| 19:42 | <timeless> | that's ok |
| 19:43 | <timeless> | but just for a header file? |
| 19:43 | <timeless> | that's not ok in my book |
| 19:43 | <annevk> | timeless, added links |
| 19:43 | <timeless> | thanks |
| 19:43 | <annevk> | Ms2ger, I could link to the thread that way I suppose |
| 19:43 | <erlehmann> | well, i think its simply a matter of strategy. if you make it too restrictive, few people will use it. |
| 19:44 | timeless | shrugs |
| 19:45 | <erlehmann> | oh well. |
| 19:46 | <Ms2ger> | Hmm |
| 19:46 | <timeless> | you object? |
| 19:46 | <Ms2ger> | Not at all |
| 19:46 | <timeless> | do you want me to get -emeritus added? |
| 19:46 | <Ms2ger> | I wish it was still true :) |
| 19:46 | <timeless> | it's better than a /maemo or /meego hostmask :) |
| 19:47 | <timeless> | that'd be really problematic :) |
| 19:47 | <timeless> | but yeah, we're a webkit shop :) |
| 19:47 | <Ms2ger> | Boo :) |
| 19:47 | <timeless> | amusingly the tools team internally has firefox addons |
| 19:47 | <timeless> | and they also fix compatibility issues w/ nightly :) |
| 19:47 | <timeless> | yay tools teams! |
| 20:01 | <tantek> | erlehmann, timeless, yes, liberal as in MIT or CC0 (preferred). |
| 20:02 | <tantek> | AryehGregor, please consider also saying the IP in the spec is available under OWFa 1.0 as well (handles patent stuff) |
| 20:02 | <AryehGregor> | What does it do about patents? |
| 20:02 | <AryehGregor> | I don't own any patents, so I can't see that I could do much. |
| 20:02 | <tantek> | it basically is the way to do a patent non-assert |
| 20:02 | <tantek> | in an open / portable way |
| 20:03 | <tantek> | the same way that CC is the way to do a copyright non-assert |
| 20:03 | <tantek> | in an open / portable way |
| 20:03 | <AryehGregor> | I have no patents to non-assert, though. |
| 20:03 | <tantek> | sure, but people often say "but w3c gets you patent /ip protection etc." |
| 20:03 | <tantek> | and using OWFa allows you to do it independently |
| 20:03 | <wilhelm> | Your employer may have. |
| 20:03 | <tantek> | http://www.openwebfoundation.org/legal/the-owf-1-0-agreements/owfa-1-0 |
| 20:05 | <tantek> | more on the topic of doing open standards development in general: http://tantek.com/2011/168/b1/practices-good-open-web-standards-development |
| 20:05 | <gsnedders> | AryehGregor: In a country with submarine patents, why should I believe you? |
| 20:05 | <tantek> | (probably more than anyone here needs, and most of that will just be duh/assumed material for the folks here) |
| 20:06 | <tantek> | (hopefully useful as a reference though) |
| 20:06 | <AryehGregor> | gsnedders, because if you suspect me of secretly holding patents, you may as well suspect everyone in the world of it, give up on software development, and take up some safe occupation like coal mining instead. |
| 20:06 | <AryehGregor> | People who worry about patents are not going to worry about me. |
| 20:06 | <AryehGregor> | Also, patent filings are public from day one, I think. |
| 20:06 | <AryehGregor> | And submarine patents as such don't exist anymore, at least as the term was formerly used. |
| 20:09 | <gsnedders> | AryehGregor: My understanding (which is limited, because, well, I'm not in the US for a start) is that the last submarine patents are still not all published. Regardless of whether you hold patents or not, people have good reason to prefer a promise you won't sue them for violating a patent (which you know they won't, because you don't hold any). |
| 20:09 | <AryehGregor> | I think submarine patents only exist if they were filed before, like, 1996. |
| 20:09 | <AryehGregor> | When I was eight. |
| 20:09 | <gsnedders> | AryehGregor: It's about insurance that the editor of the spec hasn't deliberately put something in the spec so they can sue for it. |
| 20:10 | <AryehGregor> | Okay, that's all very well, but I don't think anyone is going to be credibly worried about it. |
| 20:10 | <AryehGregor> | It's not high on my to-do list. |
| 20:10 | <gsnedders> | AryehGregor: It's about the risk. Sure, I could take you at your word, but then someday someone will lie. |
| 20:11 | <AryehGregor> | Find me someone who's actually worried about my secret patent portfolio and I'll consider spending time on the issue. |
| 20:12 | <gsnedders> | AryehGregor: Waste as much time on it as you did the text of the spec? |
| 20:12 | <AryehGregor> | I'm doing other things, this is a distraction based on a theoretical issue. |
| 20:16 | <Hixie> | hey where's innnerText supposed to live? |
| 20:16 | <Ms2ger> | In hell |
| 20:17 | <Hixie> | dom core? dom parsing? html? editing? |
| 20:17 | <Ms2ger> | Not in a spec I edit |
| 20:17 | <foolip> | surely not dom core? |
| 20:17 | <Ms2ger> | So one of the latter |
| 20:17 | <Ms2ger> | It's not clear anybody's interested in a spec, though |
| 20:17 | <gsnedders> | HTML, I'd guess |
| 20:18 | <annevk> | depends on the details |
| 20:18 | <Hixie> | k i'll just continue to ignore this problem for now |
| 20:18 | <erlehmann> | academics that deal with patents are interesting. i have no idea how the two cultures i like to think represented by cerf vs. brandenburg (the mp3 guy) are intermingled or opposed to each other. |
| 20:19 | <Hixie> | annevk: http://platform.html5.org/ needs updating for dom traversal, right? |
| 20:19 | <Hixie> | i wish http://platform.html5.org/ didn't list each section from WA1 separately |
| 20:19 | <Ms2ger> | Yes |
| 20:19 | <annevk> | yes |
| 20:20 | <annevk> | I haven't reached Mike in a while |
| 20:20 | <Hixie> | (and i wish it had css more expanded) |
| 20:21 | <jgraham> | foolip: Well it sort-of is DOM core if it applies to all namespaces. And if it doesn't apply to all namespaces, that's pretty weird |
| 20:21 | <jgraham> | OTOH I widh it would die :) |
| 20:21 | <Ms2ger> | Anybody can define stuff on Element |
| 20:21 | <foolip> | jgraham, does it exist on SVG element nodes? |
| 20:21 | <annevk> | Hixie, some of the feedback you gave me was about way different things |
| 20:22 | <Ms2ger> | That doesn't mean it needs to be in Core |
| 20:22 | <foolip> | oh, I'm stupid, what else is new |
| 20:22 | <jgraham> | Ms2ger: Yes but having gratuitously split interfaces isn't helpful |
| 20:22 | <annevk> | Hixie, like disabled attribute for <iframe> |
| 20:22 | <foolip> | I thought it was innerHTML :-/ |
| 20:22 | <jgraham> | *wish |
| 20:24 | <Hixie> | annevk: the use case was dragging something, the solution is mouse capture |
| 20:24 | <Hixie> | annevk: the iframe bit is a distraction |
| 20:24 | <Hixie> | i'll move that thread to a separate bucket on my end though |
| 20:24 | <annevk> | the keyboard stuff is also not for DOM Core |
| 20:25 | <annevk> | btw, the SVG WG wants to follow the on* pattern for event handlers |
| 20:25 | <Hixie> | one e-mail from hallvord? |
| 20:25 | <annevk> | several emails on keyboard stuff |
| 20:25 | <annevk> | and dblclick |
| 20:26 | <Hixie> | k |
| 20:27 | <Hixie> | ok after taking out event-specific stuff, i have 16 e-mails left |
| 20:28 | <Hixie> | want me to send you that bucket again so you can make sure i got the right ones? |
| 20:28 | <annevk> | cancelling keydown is one of those |
| 20:28 | <annevk> | sure |
| 20:28 | <annevk> | there was some relevant event stuff about Apple-specific extensions to the Event interface |
| 20:28 | <annevk> | (though I doubt that will result in spec changes) |
| 20:29 | <zewt> | i'm surprised more sites aren't really really evil about cancelling keydown |
| 20:29 | <Hixie> | annevk: if there's nothing in what i just sent you that needs any response and it's all been taken into account already, i can just delete the bucket on my end |
| 20:32 | <annevk> | Node having an item() method looks like something that might need updating in DOM Core |
| 20:32 | <annevk> | "Node item(in long index); // and Node is also indexable as array" |
| 20:32 | <Ms2ger> | Who does that? |
| 20:32 | <annevk> | this was an email from Maciej |
| 20:34 | <annevk> | Safari also has a bunch of stuff on Event apparently |
| 20:34 | <annevk> | srcElement - clone of the nonstandard IE extension srcElement, an alias for target |
| 20:34 | <annevk> | returnValue - clone of the nonstandard IE extension returnValue, this is a read-write property that lets you get and set whether default is prevented. This lets you un-prevent default, something not allowed in the standard DOM event model. |
| 20:34 | <annevk> | cancelBubble - similar to returnValue, but for stopPropagation |
| 20:34 | <annevk> | dataTransfer - this appears to be applicable only to drag & drop events, probably a mistake that it is in the Event interface |
| 20:34 | <annevk> | clipboardData - similarly, but for clipboard events |
| 20:35 | <annevk> | Hixie, you can delete that bucket, I will file a bug on Node.item() and looking at Event extensions |
| 20:35 | <Hixie> | cool |
| 20:35 | <Hixie> | thanks dude |
| 20:38 | <annevk> | hmm I cannot find item in document |
| 20:42 | <Hixie> | ok twice now i have explained to plh how the w3c process is broken, he claims it is not, i then respond in detail explaining how it is, and he goes silent |
| 20:42 | <Hixie> | http://lists.w3.org/Archives/Public/public-web-perf/2011Jul/0090.html and http://lists.w3.org/Archives/Public/public-web-perf/2011Apr/0013.html |
| 20:43 | <erlehmann> | well, egoless discussions are rare. and you are like, the W3Cs nemesis personified? ;) |
| 20:43 | <zewt> | at least he didn't say "let's agree to disagree" as if that settles everything |
| 20:43 | <AryehGregor> | plh just added me on Google+ after my big post about how I'm not interested in working with the W3C. |
| 20:43 | <zewt> | that's pretty much the worst internet-disagreement expression there is |
| 20:43 | <Hixie> | i just wish he would either explain why i'm wrong, or stop claiming things that i've shown are wrong |
| 20:43 | <erlehmann> | “i think both parties have made their points today. now for something completely different” |
| 20:45 | <erlehmann> | such people exist. i regard overly broad criticism that is not told as backstory to narrow issues as trolling anyway. |
| 20:53 | <erlehmann> | annevk, not wanting to troll you, but why don't you link directly to the fragment of the comment on the page? |
| 20:53 | <erlehmann> | (i know people hate scrolling) |
| 20:55 | <Hixie> | AryehGregor: fwiw, the editing stuff looks fantastic. |
| 20:56 | <AryehGregor> | Hixie, thanks. |
| 20:57 | <timeless> | AryehGregor: i think the current problem with patents is that you can revise claims after filing.. |
| 20:57 | <AryehGregor> | timeless, I think that in most cases, the current problem with patents is that they exist. |
| 20:57 | <Hixie> | so this crypto thread on whatwg -- anyone know what spec it's supposed to affect? |
| 20:57 | <AryehGregor> | But that's a futile discussion. |
| 20:57 | <AryehGregor> | Hixie, about the new API? I assumed people were making a new spec. |
| 20:58 | <Hixie> | do you know who specifically? |
| 20:58 | <AryehGregor> | No clue, haven't been following closely. |
| 20:59 | <timeless> | annevk: mike was around on monday iirc |
| 21:01 | <timeless> | zewt: ePost.ca seems pretty evil about form input |
| 21:01 | <timeless> | i hit a couple of other .ca sites which were pretty evil |
| 21:02 | <zewt> | that's one place it's easy to do really stupid things, but not necessarily willfully |
| 21:02 | <zewt> | like "preventDefault if the key isn't [0-9]" |
| 21:02 | <zewt> | but evil i'm talking about things like preventing escape for browser stop, alt-left/right, f5, etc |
| 21:02 | <Ms2ger> | Hixie, ddahl is writing up something he calls a spec :) |
| 21:02 | <zewt> | eg. deliberately being obnoxious to users |
| 21:03 | <Hixie> | Ms2ger: for crypto? |
| 21:03 | <Hixie> | Ms2ger: k |
| 21:03 | <timeless> | Hixie: i presume it's the mozilla hosted crypto spec |
| 21:03 | <Hixie> | do you have a link? |
| 21:03 | <zewt> | there was some proposal a while back but it was sort of a mess and i didn't look very closely |
| 21:03 | <Ms2ger> | Somewhere on wikimo, I'll look |
| 21:03 | <timeless> | zewt: oh, i think i've hit some of those |
| 21:03 | <timeless> | Hixie: yeah, i think |
| 21:04 | <timeless> | grumble, it isn't googleable |
| 21:04 | <timeless> | lemme ask gmail |
| 21:04 | <Ms2ger> | https://wiki.mozilla.org/Privacy/Features/DOMCryptAPISpec/Latest |
| 21:04 | <zewt> | the other day i hit a "wait 30 seconds to download" site that stopped the countdown when the window loses focus, to try to force you to sit there and wait :( |
| 21:04 | <zewt> | another one of those really gross evil obnoxious things that I'm surprised I hadn't seen before |
| 21:05 | <timeless> | oh right, it doesn't have 'o' after 'Crypt' |
| 21:05 | <timeless> | it intentionally breaks any google searches i might try |
| 21:05 | <timeless> | zewt: those will be *more* common |
| 21:05 | <Ms2ger> | timeless, what's the bugzilla component to add new repos to mxr? |
| 21:05 | <timeless> | webtools/mxr i think |
| 21:06 | <zewt> | those have been possible for a long long time, and those download sites have always been obnoxious, so it's puzzled me that it took so long |
| 21:06 | <timeless> | zewt: http://www.w3.org/TR/2011/WD-page-visibility-20110721/ |
| 21:06 | <Ms2ger> | OK, I thought that was about issues in MXR |
| 21:06 | <timeless> | Ms2ger: i don't think we have a second place |
| 21:06 | <timeless> | you can use server operations |
| 21:06 | <zewt> | it's the sort of thing that just needs browser overrides, really |
| 21:07 | <zewt> | or, for my purposes, at least enough functionality in things like GM to kill it |
| 21:07 | <Hixie> | looks like ddhal and abarth are coauthoring it |
| 21:08 | <Hixie> | abarth: what should i do with crypto-related feedback sent to whatwg? |
| 21:10 | <timeless> | Ms2ger: technically you couldmisfile in websites/Other |
| 21:11 | <Ms2ger> | I (mis)filed in ServerOps |
| 21:11 | timeless | can't *find* serverops |
| 21:12 | <Ms2ger> | mozilla.org |
| 21:14 | <Ms2ger> | Huh, Google tells me Hixie +1d dev.w3.org/2006/webapi/FileAPI/ |
| 21:15 | <Ms2ger> | And that page is completely broken |
| 21:15 | <Hixie> | wow, what happened to that page |
| 21:15 | <Ms2ger> | It starts with |
| 21:15 | <Ms2ger> | Overview.html |
| 21:15 | <Ms2ger> | Web IDL |
| 21:15 | Ms2ger | looks for heycam |
| 21:17 | <Hixie> | wtf is that page |
| 21:18 | <Ms2ger> | A botched attempt at using XSLT |
| 21:18 | <Hixie> | sicking: File API spec got broken dude |
| 21:20 | <Ms2ger> | Probably because Arun's last edit introduced a wellformedness error |
| 21:20 | <Ms2ger> | "This note is normative." |
| 21:21 | <Ms2ger> | What? |
| 21:21 | <erlehmann> | haha |
| 21:32 | <annevk> | :) |
| 21:32 | <annevk> | erlehmann, give me the fragment |
| 21:33 | <erlehmann> | annevk, i believe you are implying g+ is somehow too dense for that. point made. |
| 21:34 | <annevk> | I mean I couldn't get to the fragment |
| 21:36 | <annevk> | Ms2ger, so I cannot find Node.item() anymore |
| 21:38 | <sicking> | Hixie: ugh :( Poke Arun |
| 21:38 | <annevk> | that's why people should use Anolis to make specs |
| 21:38 | <Ms2ger> | Me neither |
| 21:38 | <annevk> | good |
| 21:43 | <annevk> | http://trac.webkit.org/browser/trunk/Source/WebCore/dom/Event.idl does seem to have quite some extensions |
| 21:44 | <AryehGregor> | Hixie, okay, I responded to all of the stuff you sent me where I could give a useful answer, and that was sent to whatwg (not other places). |
| 21:44 | <AryehGregor> | And where I hadn't already responded months ago. |
| 21:45 | <AryehGregor> | Some of those threads are things I started. |
| 21:45 | <annevk> | http://mxr.mozilla.org/mozilla-central/source/dom/interfaces/events/nsIDOMEvent.idl seems to match the specification |
| 21:45 | <annevk> | but Mozilla sometimes additional stuff elsewhere |
| 21:46 | <AryehGregor> | Okay, can I post stuff to the WHATWG blog somehow? I can't remember if I have an account. |
| 21:46 | AryehGregor | doesn't see anywhere to try logging in |
| 21:47 | <Ms2ger> | annevk, try nsIDOMNSEvent |
| 21:47 | <Ms2ger> | (I'm removing it, but haven't got around to it yet) |
| 21:50 | <annevk> | originalTarget |
| 21:50 | <annevk> | explicitOriginalTarget |
| 21:50 | <annevk> | and a similar list of constants as Safari |
| 21:50 | <annevk> | bah |
| 22:01 | <timeless> | AryehGregor: ... |
| 22:01 | <timeless> | about nested lists |
| 22:01 | <AryehGregor> | What about them? |
| 22:01 | <timeless> | have you seen http://support.microsoft.com/kb/2489283 ? |
| 22:01 | <timeless> | (random kb article) |
| 22:01 | <AryehGregor> | What about it? |
| 22:01 | <timeless> | just that they tend to have <ul>'s in <li>'s |
| 22:02 | <AryehGregor> | It nests <ul> inside <li>. That's not a problem. |
| 22:02 | <AryehGregor> | I never said that should be invalid. |
| 22:02 | timeless | considers trying to reread the message |
| 22:02 | <AryehGregor> | But I don't think execCommand() should produce it, under normal circumstances. |
| 22:02 | <AryehGregor> | For execCommand(), it's simpler to only have lists nested directly inside lists. |
| 22:02 | timeless | ponders |
| 22:02 | <timeless> | so... |
| 22:02 | <AryehGregor> | It doesn't have to let you write arbitrary HTML. If you want to do something fancier, write the HTML yourself. |
| 22:03 | <annevk> | emailed www-dom about extensions to Event |
| 22:03 | <AryehGregor> | FWIW, I wasn't able to get Word or OpenOffice to produce lists like that one. |
| 22:03 | <timeless> | the problem is that i think the kb article is a pretty natural thing to want to write |
| 22:03 | <timeless> | really? |
| 22:03 | <AryehGregor> | In a WYSIWYG editor? |
| 22:03 | <AryehGregor> | I mean, you can get something that looks visually something like that. |
| 22:04 | <AryehGregor> | You can do that with my execCommand() spec too. |
| 22:04 | <AryehGregor> | As long as you only put the nested list at the end of the <li>. |
| 22:04 | <timeless> | but you're saying you couldn't get them to generate content that's logically like that? |
| 22:05 | <AryehGregor> | I'm saying we should treat </li><ol>...</ol> logically the same as <ol>...</ol></li>. |
| 22:05 | <AryehGregor> | We can make up whatever semantics are convenient. |
| 22:05 | <annevk> | Ms2ger, please clean up Gecko http://lists.w3.org/Archives/Public/www-dom/2011JulSep/0039.html |
| 22:05 | <AryehGregor> | annevk, how can I post something on the WHATWG blog? |
| 22:05 | <Ms2ger> | File a bug |
| 22:06 | <annevk> | I guess I can |
| 22:06 | <AryehGregor> | Hmm, actually, I guess you'll mention my announcement anyway. |
| 22:06 | <AryehGregor> | In the WHATWG Weekly. |
| 22:06 | <AryehGregor> | So I don't see any need. |
| 22:06 | <timeless> | annevk: my favorite is explicitOriginalTarget |
| 22:06 | <annevk> | AryehGregor, please post it separately |
| 22:06 | <annevk> | I'll make an account for you |
| 22:06 | <AryehGregor> | Okay, why? |
| 22:06 | <timeless> | (there actually was one worse than that) |
| 22:06 | timeless | thinks it had `real` in it |
| 22:06 | <annevk> | AryehGregor, because six months of work deserves a little more than a note |
| 22:07 | <AryehGregor> | MediaWiki has or used to have series of methods in the same class called "X", "doX", "reallyDoX". In at least one place. |
| 22:07 | <AryehGregor> | annevk, k. |
| 22:08 | <timeless> | annevk: http://mxr.mozilla.org/mozilla-central/source/dom/interfaces/events/nsIDOMNSEvent.idl#90 |
| 22:08 | <timeless> | fwiw |
| 22:08 | <annevk> | you should have an invite for the blog AryehGregor |
| 22:08 | <timeless> | 105 [noscript] readonly attribute nsIDOMEventTarget tmpRealOriginalTarget; |
| 22:08 | <timeless> | is the best part |
| 22:08 | <AryehGregor> | annevk, got it. |
| 22:09 | <annevk> | timeless, yeah, but it has [noscript] so I decided to not make jokes about it :) |
| 22:09 | <timeless> | annevk: yeah well |
| 22:09 | <timeless> | what i like is the comment |
| 22:09 | <timeless> | http://mxr.mozilla.org/mozilla-central/ident?i=tmpRealOriginalTarget |
| 22:09 | <timeless> | 100 /* XXX This is TEMPORARY. |
| 22:10 | <annevk> | when was it added? |
| 22:10 | <annevk> | :) |
| 22:10 | <timeless> | hold on |
| 22:10 | <timeless> | i need to ask bonsai(cvs) |
| 22:10 | <timeless> | http://bonsai.mozilla.org/cvsblame.cgi?file=mozilla/security/manager/ssl/src/nsSmartCardEvent.cpp&rev=1.1&root=/cvsroot |
| 22:10 | <timeless> | around 2005 in that file |
| 22:10 | <timeless> | http://bonsai.mozilla.org/cvsview2.cgi?diff_mode=context&whitespace_mode=show&root=/cvsroot&subdir=mozilla/editor/libeditor/text&command=DIFF_FRAMESET&file=nsPlaintextDataTransfer.cpp&rev2=1.35&rev1=1.34 |
| 22:10 | <timeless> | 2003 there |
| 22:11 | <timeless> | but that was probably just the rename from `explicitOriginalTarget` to `tmpRealOriginalTarget` |
| 22:12 | <timeless> | http://bonsai.mozilla.org/cvsblame.cgi?file=mozilla/dom/public/idl/events/nsIDOMNSEvent.idl&rev=1.4.32.1#100 |
| 22:12 | <annevk> | heh |
| 22:12 | <annevk> | https://bugzilla.mozilla.org/show_bug.cgi?id=674696 |
| 22:12 | <timeless> | the change is from 2003 |
| 22:12 | <timeless> | sorry, may 2003 |
| 22:12 | <timeless> | so, it's only 8 years old |
| 22:12 | <annevk> | guess I'll file one on WebKit too |
| 22:13 | <timeless> | anyway, it looks like explicit could be made [noscript] trivially |
| 22:13 | <timeless> | which would be sufficient for your requirements |
| 22:14 | <annevk> | no the constants, originalTarget etc. are all proprietary |
| 22:15 | <jcranmer> | AryehGregor: there is an interface in Thunderbird which has "name", "prettyName", and "prettiestName" (with about 3 or 4 other "name"s) |
| 22:15 | <AryehGregor> | As the saying goes, there are only two hard problems in computer science: cache invalidation and naming things. |
| 22:16 | <Ms2ger> | And off-by-one errors |
| 22:16 | <annevk> | and off by one errors |
| 22:16 | <annevk> | :) |
| 22:16 | <Ms2ger> | :) |
| 22:18 | <annevk> | http://daringfireball.net/linked/2011/07/27/html5-fullscreen hmm |
| 22:32 | <annevk> | filed bugs ms2ger |
| 22:33 | <annevk> | and good night |
| 22:42 | <AryehGregor> | k, posted. |
| 22:42 | <AryehGregor> | timeless, subscribe to the WHATWG blog to get weekly-ish updates on interesting stuff that's happening in WHATWG-land. |
| 22:43 | <Hixie> | AryehGregor: cool, thanks |
| 22:46 | <annevk> | AryehGregor, could use some <code> love |
| 22:46 | <AryehGregor> | Yeah, maybe. |
| 22:46 | <AryehGregor> | Feel free if you like. :) |
| 22:47 | <AryehGregor> | I was content with adding links. |
| 22:47 | <annevk> | meh |
| 22:53 | <AryehGregor> | Oh, so Google Plus lets you make a post that's officially public but you can secretly block arbitrary people from posting there? |
| 22:53 | <AryehGregor> | That seems sinister. |
| 22:53 | <annevk> | TabAtkins, yes it does |
| 22:54 | <annevk> | TabAtkins, but you get initial containing block and maybe that gets in the way |
| 22:54 | <annevk> | TabAtkins, like with a checkbox being the root |
| 22:58 | <annevk> | ok people |
| 22:58 | <annevk> | so here is the original email from Maciej |
| 22:58 | <annevk> | http://lists.w3.org/Archives/Public/public-webapi/2006Apr/0161.html |
| 22:58 | <annevk> | anyone who can find |
| 22:58 | <annevk> | Node item(in long index); |
| 22:58 | <annevk> | in WebKit and when/if it was removed gets a cookie |
| 22:59 | <jamesr> | can you do that? |
| 22:59 | <Hixie> | sure you just use Set-Cookie right? |
| 23:00 | <annevk> | AryehGregor, you just block people |
| 23:00 | <timeless> | will it be ;secure? :) |
| 23:00 | <AryehGregor> | annevk, yes, but it seems sinister that the discussion looks like it's public but some unknown list of people is blocked. |
| 23:01 | <AryehGregor> | Although I guess the discussion starter can also delete posts and things. |
| 23:01 | <AryehGregor> | So it's really controlled by them anyway. |
| 23:01 | <annevk> | AryehGregor, seems normal to me for discussions |
| 23:01 | <annevk> | pretty much any public forum has some amount of moderation |
| 23:02 | <annevk> | so they keep functioning |
| 23:02 | <AryehGregor> | Yeah, but it's usually some official moderator people or such, not the person who happened to start the thread. |
| 23:02 | <AryehGregor> | It means I might need to respond differently based on who started the thread, which seems weird to me. |
| 23:02 | <AryehGregor> | But my background is mailing lists and forums, not Facebook-style stuff. |
| 23:04 | <AryehGregor> | "To Areyh, I understand you are not a subject matter expert. Consequently, I put its importance back to major." |
| 23:04 | <AryehGregor> | Meh, see if I bother doing anything with a11y bugs but ignore them from now on. |
| 23:04 | <abarth> | Hixie: maybe ddhal would like to see the crypto-related feedback to whatwg? he's driving. i'm just helping him out |
| 23:05 | <timeless> | AryehGregor: which browser? |
| 23:05 | <jamesr> | so..when do i have to use adoptNode()? |
| 23:05 | <AryehGregor> | timeless, what? |
| 23:05 | <AryehGregor> | jamesr, AFAICT, you don't. |
| 23:05 | <timeless> | "To Areyh, I understand you are not a subject matter expert. Consequently, I put its importance back to major." |
| 23:06 | <jamesr> | AryehGregor: hm, ok |
| 23:06 | <jamesr> | i have a page where i create an iframe, put a canvas inside of it, then move the iframe to a popup window's document |
| 23:06 | <timeless> | oh, w3.org |
| 23:06 | <AryehGregor> | timeless, not a browser: http://www.w3.org/Bugs/Public/show_bug.cgi?id=13176#c5 |
| 23:06 | <timeless> | yeah, google :) |
| 23:06 | <jamesr> | the canvas itself doesn't survive the transition in firefox or opera |
| 23:06 | <jamesr> | why? |
| 23:07 | <AryehGregor> | I mean, I'm pretty sure severity is supposed to be set by the editor and/or editorial assistants, but frankly I don't care, I'll just ignore them from now on. |
| 23:07 | <jamesr> | hah, rich schwerdflkjerlkj |
| 23:07 | <jamesr> | AryehGregor: that guy is just a dick |
| 23:07 | <timeless> | AryehGregor: generally people don't understand how those fields work |
| 23:07 | <AryehGregor> | timeless, I object to the condescension. |
| 23:07 | <timeless> | and bugzilla at some day should be changed to explain it |
| 23:08 | <timeless> | ideally they're the reporter's realm until a qa comes |
| 23:08 | <timeless> | once the qa claims it, they're the qa's until an assignee touches the bug |
| 23:08 | <timeless> | at which point, they're the assignee's |
| 23:08 | <AryehGregor> | If they want to be jerks, their bugs can sit around forever without resolution so they can't escalate without chair intervention. I don't care. |
| 23:08 | <timeless> | but bugzilla doesn't enforce that |
| 23:08 | <timeless> | heh |
| 23:08 | <AryehGregor> | jamesr, you're right, I was too hasty. I'll just ignore his bug reports, not other a11y people's. |
| 23:09 | <AryehGregor> | timeless, we have guidelines for severity: http://wiki.whatwg.org/wiki/Bugzilla_conventions#Severities |
| 23:09 | <AryehGregor> | "major" is unused. |
| 23:09 | <AryehGregor> | But I don't know if severity is a useful field anyway. |
| 23:09 | <AryehGregor> | In my experience, it's almost always inaccurate enough that it's useless, and changing it starts fights. |
| 23:10 | <timeless> | the last bit is always true :) |
| 23:10 | <AryehGregor> | Often of the form "developer sets severity to enhancement, reporter gets offended because they interpret that as 'not important'". |
| 23:10 | <timeless> | be happy it isn't a private bug tracker |
| 23:10 | <timeless> | at nokia, we had a group who kept adding `escalate` everywhere |
| 23:11 | timeless | thinks that was the l10n testing group as opposed to the a11y group, but.. |
| 23:12 | <timeless> | do they have a sample page demoing a real use case for this? |
| 23:12 | <timeless> | it's way too abstract |
| 23:13 | <jamesr> | i don't know if you want to start going down the 'can we get use cases' path |
| 23:13 | <timeless> | i don't mean use case as in `tell me that you need accessibility` |
| 23:13 | <timeless> | i mean `show me a site that's trying and has something so i can see how the problem really manifests` |
| 23:13 | <timeless> | problems you can't touch, grasp, or hold are much harder to address properly |
| 23:14 | <jamesr> | http://lists.w3.org/Archives/Public/public-canvas-api/ is full of people asking them (richard schwesdflkjsdlkfj and charles prichard) for exactly that and them writing novellas that don't quite give concrete use cases but sure do use a lot of words |
| 23:14 | <timeless> | heh |
| 23:15 | <timeless> | i'm not an expert, but i do actually *use* accessibility agents |
| 23:16 | <timeless> | and i file bugs against them (sometimes even with patches) |
| 23:17 | timeless | waves goodbye to irccloud.com |
| 23:17 | <clair> | Hi guys, I want to check before I do so, is it ok for non w3c members to sign up to get CC notifications of w3c bugs? |
| 23:18 | <timeless> | clair: i don't see why it'd be a problem |
| 23:18 | <timeless> | other than the fact that the bugs will probably be 99% noise |
| 23:18 | <clair> | It's just one particular bug I'm interested in, really, I assume you can subscribe to a single one? |
| 23:19 | <timeless> | clair: although, i'm not sure `cc notifications` means what you want it to mean |
| 23:19 | <timeless> | yes |
| 23:19 | <timeless> | you can get an account and cc yourself to the bug |
| 23:19 | <clair> | Yeah that's what I'm thinking - I assume I then get email replies to the ticket? |
| 23:20 | <timeless> | yes |
| 23:20 | <timeless> | you can control which types |
| 23:20 | <clair> | OK, thanks :) |
| 23:21 | <timeless> | http://www.bugzilla.org/docs/4.0/en/html/userpreferences.html#emailpreferences |
| 23:21 | <timeless> | is vaguely relevant |
| 23:21 | <timeless> | sadly, the doc doesn't actually have much to say directly about email |
| 23:21 | <timeless> | it should have a section about it |
| 23:21 | <timeless> | note to self: file bug |
| 23:39 | <Hixie> | abarth: k |
| 23:43 | <Hixie> | hm, if we're going to have other specs on whatwg.org maybe we should have a subdomain |
| 23:43 | <Hixie> | like .specs.whatwg.org |
| 23:43 | <Hixie> | so e.g. fullscreen.specs.whatwg.org |
| 23:43 | <Hixie> | or fullscreen.s.whatwg.org |
| 23:43 | <Hixie> | for tantek's |
| 23:43 | <Hixie> | and wa1.s.whatwg.org or html.s.whatwg.org or whatever we call the complete spec |
| 23:44 | <nessy> | leave out the specs part |
| 23:44 | <abarth> | would those domains point to whatwg servers, or how would we push specs there? |
| 23:44 | <Hixie> | i'd give the editor of each spec a separate account |
| 23:44 | <nessy> | fullscreen.whatwg.org , html.whatwg.org work for me |
| 23:44 | <Hixie> | or we could point them to whever is most appropriate |
| 23:44 | <Hixie> | nessy: the idea would be to keep them separate from blogs. forums. status. and so on |
| 23:45 | <nessy> | ah, hmm |
| 23:45 | <nessy> | .doc. ? |
| 23:45 | <abarth> | Hixie: ok, i'd like to try that for the sniffing spec. Pete Resnick seems pretty opposed to it moving forward at IETF |
| 23:45 | <Hixie> | .s. :-) |
| 23:45 | <Hixie> | abarth: ah, k |
| 23:45 | <nessy> | .s. ist a bit meaningless that's all |
| 23:45 | <Hixie> | abarth: do you have a timeline on this? |
| 23:45 | <nessy> | looks like a typo |
| 23:45 | <Hixie> | short for .spec. |
| 23:46 | <abarth> | Hixie: whenever is convenient for you. there's no real rush |
| 23:46 | <nessy> | yeah, I know, but still looks like a typo ;-) |
| 23:46 | <Hixie> | html.spec.whatwg.org fullscreen.spec.whatwg.org |
| 23:46 | <Hixie> | maybe .spec. is short enough |
| 23:46 | <nessy> | WFM |
| 23:46 | <abarth> | html.TR.whatwg.org |
| 23:46 | <Hixie> | where are ms2ger and anne when you need them |
| 23:46 | <Hixie> | abarth: hahaha |
| 23:47 | <Hixie> | tempting |
| 23:50 | <Hixie> | jeez, whatwg.org has a lot of subdomains already |
| 23:52 | <Hixie> | blog, c, demos, developers, forums, help, history, images, n, status, svn, validator, wiki, xn--7ca. (history and help are dead, demos and status are just subdomains for other subsubdomains) |
| 23:54 | <jamesr> | what's the correct name for "all unicode characters whose code points are <= 255"? |
| 23:55 | <jamesr> | ascii plus some |
| 23:55 | <Hixie> | "Unicode codepoints in the range U+0000 to U+00FF" |