| 00:32 | <webben> | Hixie: ah cool :) |
| 00:33 | <webben> | re v2 list |
| 03:31 | <Hixie> | http://webblaze.cs.berkeley.edu/2009/mime-sniff/mime-sniff.txt |
| 03:31 | <Hixie> | adam rocks |
| 05:02 | <BenMillard> | some better ALA comments are turning up |
| 05:03 | <BenMillard> | http://alistapart.com/comments/semanticsinhtml5?page=5#41 and http://alistapart.com/comments/semanticsinhtml5?page=5#48 imho |
| 05:04 | <BenMillard> | (Benjamin Hawkes-Lewis on existing extensibility mechanisms in HTML, then Ryan Cannon quoting the HTML5 spec of data-* attributes) |
| 05:10 | <BenMillard> | hmm, XHTML5: http://alistapart.com/comments/semanticsinhtml5?page=6#52 |
| 05:12 | <BenMillard> | (has an XML prolog and <doctype html> but <link rel="stylesheet" /> rather than <?xml-stylesheet ...?>) |
| 05:12 | <BenMillard> | (with a space before the />, XHTML-in-text/html style) |
| 05:49 | BenMillard | just saw JohnResig writing about data-* attributes here: http://ejohn.org/blog/html-5-data-attributes/ |
| 05:54 | <BenMillard> | JohnResig, how about this instead of data-* for that example: <b title="John Resig lives in Boston and likes to eat bacon">John</b> says: [...] |
| 05:55 | <BenMillard> | JohnResig, or linking the name to a page which includes these (and possibly more) details: <a href="...">John</b> says: [...] |
| 05:57 | <BenMillard> | JohnResig, perhaps you could show what an hCard for yourself would look like if written using data-* and compare it to the same thing done using Microformats as a more compelling example? |
| 09:02 | <takkaria> | yay for the Internet-Draft |
| 10:50 | <annevk> | webben, fwiw, there's no <h> element in HTML5 |
| 10:50 | <annevk> | re: http://www.alistapart.com/comments/semanticsinhtml5?page=6#53 |
| 10:51 | <Hixie> | webben: (it's spelt <h1>) |
| 10:51 | <annevk> | (those hyperlinks seem bound to break the moment they remove a duplicate comment, grmbl) |
| 10:51 | <webben> | oh, didn't there used to be? |
| 10:51 | <annevk> | nope |
| 10:51 | <webben> | oh okay. |
| 10:52 | webben | will write a correction. |
| 10:54 | <webben> | it's correct to say that the heading algorithm has been redefined to allow h1 to solve the problems I mention, right? |
| 10:55 | <Hixie> | <section> with <h1> solves the problem of content syndication where you don't know what level heading to use ahead of time |
| 10:56 | <Hixie> | that's really the only thing i was trying to solve with it, though it does effectively solve <h7> and copy-and-paste as well |
| 10:56 | <webben> | cool |
| 10:57 | Philip` | thinks that if your document has more than six nested levels of headings, you should probably think about simplifying your document :-) |
| 10:57 | <Hixie> | yeah |
| 10:57 | <webben> | Generally true, yes. |
| 11:00 | <Philip`> | (I'd guess most existing uses of <h6> come from people wanting headings with a very small font, rather than people who've already used all of <h1>-<h5>) |
| 11:02 | <hsivonen> | The use case for h6 is shifting down heading when creating a compilation of documents that have h1-h5 |
| 11:04 | webben | issues his correction. Thanks for catching that Hixie and annevk . |
| 11:20 | webben | is puzzled by how XHTML2's modularity magics away it's deviations from how test/html needs to be parsed. http://lists.w3.org/Archives/Public/www-html/2009Jan/0015.html |
| 11:20 | <webben> | *its |
| 11:21 | <hsivonen> | Hmm. Windows 7 beta tomorrow. I wonder if it installs under Parallels. |
| 11:34 | <annevk> | Hixie, " XXX * simple color picker" is done |
| 11:34 | <Hixie> | yeah several a re done |
| 11:34 | <Hixie> | i'll probably go through the list in late february or march |
| 11:35 | <annevk> | k |
| 11:36 | <hsivonen> | hmm. the promo video listing new exciting features of Windows 7 lists features of which all but one are already in Mac OS X... |
| 11:37 | <Hixie> | what's the missing one? |
| 11:37 | <hsivonen> | Hixie: a window that tells you about the features of your device when you plug it in |
| 11:37 | <Hixie> | o_O |
| 11:37 | <hsivonen> | (I'm not sure that's a feature I want to have.) |
| 11:38 | <hsivonen> | device being a device other than the PC itself, that is |
| 11:40 | <hsivonen> | furthest block always stays on the stack, it seems. right? |
| 11:41 | <Hixie> | the root one? yeah |
| 11:41 | <Hixie> | pretty sure |
| 11:41 | <hsivonen> | Hixie: I mean "furthest block" in AAA |
| 11:41 | <Hixie> | oh |
| 11:42 | <Hixie> | i don't claim to know anything about AAA |
| 11:42 | <hsivonen> | I assume "common ancestor" stays on the stack as well |
| 11:47 | <hsivonen> | hmm. it *looks* like the local variables that are initialized by assigning from the stack or the list are all effectively weak references |
| 11:47 | <hsivonen> | I sure hope my analysis is correct. |
| 11:48 | <Hixie> | put a comment to that effect there so that when debugging you have a lead |
| 11:48 | <hsivonen> | I put "// weak ref" after each one |
| 11:49 | <Hixie> | i meant something like "I believe all these weak refs are correct, but if this is crashing around here, maybe my analysis is wrong" or some such :-) |
| 11:49 | <hsivonen> | ok. done |
| 12:14 | <hsivonen> | bah, the Windows 7 beta requires Vista to begin with. |
| 12:23 | <gsnedders> | hsivonen: What says that? |
| 12:27 | <gsnedders> | hsivonen: All I can see is it only upgrades from Win Vista SP1 |
| 12:28 | <hsivonen> | gsnedders: I thought it meant upgrading was the only supported mode of installation |
| 12:28 | <hsivonen> | gsnedders: is that not the case? |
| 12:28 | <gsnedders> | hsivonen: I'm not sure. I think it's a question of how you read it. |
| 12:29 | <gsnedders> | hsivonen: "you can most certainly do a clean install of the Windows 7 Beta" |
| 12:29 | <gsnedders> | hsivonen: From comments of the Win7 blog |
| 12:29 | <hsivonen> | gsnedders: ok. |
| 14:39 | <takkaria> | does anyone know of something like a <canvas> interactive interpreter? |
| 14:39 | <takkaria> | so you have your canvas, your line history, and a box to type more instructions in |
| 14:49 | <Philip`> | takkaria: http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C!DOCTYPE%20html%3E%0D%0A%3Ccanvas%20id%3Dc%3E%3C%2Fcanvas%3E%3Cscript%3Ewindow.onload%3Dfunction()%7Bvar%20ctx%3Ddocument.getElementById('c').getContext('2d')%3B%0D%0A%20%20%2F*%20write%20your%20stuff%20here%20*%2F%0D%0A%7D%3C%2Fscript%3E |
| 16:47 | <annevk> | http://twitter.com/threedaymonk/statuses/1104618551 welcome to the interwebs |
| 17:40 | <webben> | Does anyone know the current story on incorporating MathML/SVG into text/html? Is the effort dead, deferred to a future version, low priority, or what? |
| 17:40 | <olliej> | annevk: hehehe |
| 17:41 | <gsnedders> | webben: MathML is in current draft; SVG is awaiting action from SVG WG. |
| 17:41 | <webben> | it's in the draft now? |
| 17:41 | <gsnedders> | webben: (Hixie's original proposal for SVG is in the draft, just commented out) |
| 17:41 | <gsnedders> | webben: yeah |
| 17:41 | webben | really needs to start following the draft via CVS again. |
| 17:41 | <webben> | gsnedders: thanks :) |
| 17:41 | <Philip`> | http://hsivonen.iki.fi/html5-gecko-build/ has a browser implementation with MathML and SVG support in text/html |
| 17:42 | <webben> | thanks Philip` |
| 17:48 | <webben> | If I'm reading the draft right, it's basically just <math><!-- paste XML MathML here? --></math> ? |
| 17:49 | <annevk> | well, XML without prefixes or namespaces, but yeah |
| 18:28 | <BenMillard> | webben, you are Benjamin Hawkes-Lewis? |
| 18:35 | <hsivonen> | rubys: I committed code for releasing memory in the HTML5 parser earlier today. I expect the code for releasing element nodes to change in the future to make it possible to dispatch memory release request to a different thread. |
| 18:36 | <hsivonen> | s/request/requests/ |
| 18:39 | <hsivonen> | rubys: I hope the code for releasing local names can remain as-is. I'll have to check if atoms are thread-safe in Gecko. |
| 18:40 | <hsivonen> | (I'm not really satisfied with the manual nature of the memory management and having to know where stuff gets released, but alternatives seemed like overkills.) |
| 18:43 | <rubys> | I've only just gotten back to looking at the code, the first assumption that the code makes is that unicode is stored as utf-16. |
| 18:44 | <rubys> | For Ruby, at least, I'd like to use utf-8. |
| 18:45 | <rubys> | My immediate plans at to try to compile the generated code outside of Mozilla, stubbing everything I need to to get it to compile (not execute or even link, simply compile). That will give me an initial list of things I need to worry about. |
| 18:46 | <hsivonen> | I guess I should write some kind of document pointing out the memory ownership assumptions |
| 18:49 | <rubys> | don't worry too much about it, I'll discover what I need to from the code, when I see things that need addressed, I'll discuss them with you, and then over time we will evolve this into something that can handle multiple environments... at that point documentation would be helpful. |
| 19:10 | <takkaria> | rubys: there is a C HTML5 parser around which assumes UTF-8 :) |
| 19:13 | <rubys> | takkaria: url? |
| 19:14 | <takkaria> | http://www.netsurf-browser.org/projects/hubbub/ |
| 19:14 | <takkaria> | I spent my summer working on it |
| 19:14 | <takkaria> | it hasn't had a formal release yet either |
| 19:14 | <takkaria> | but it is being used as a parser in a browser for minority OSs |
| 19:15 | <rubys> | good choice of license |
| 19:35 | <hsivonen> | http://en.wikipedia.org/w/index.php?title=XHTML&diff=262789758&oldid=prev |
| 19:37 | <BenMillard> | hsivonen, ouch. |
| 19:38 | <smedero> | whatwg has almost always referred to it as HTML5 |
| 19:38 | <smedero> | and w3c has almost said HTML 5 |
| 19:38 | <smedero> | why that needs to be corrected in an wikipedia article is beyond me |
| 19:39 | <hsivonen> | smedero: it's a convenient way to also brush off the observation that XHTML 1.2 is not chartered |
| 19:40 | <hober> | whatever happened to xhtml2 rebranding? |
| 19:40 | <smedero> | indeed |
| 19:41 | <gsnedders> | smedero: No, we haven't. |
| 19:41 | <BenMillard> | hsivonen, and changing xmlns="http://www.w3.org/2002/06/xhtml2/"; to xmlns="http://www.w3.org/1999/xhtml"; |
| 19:42 | <smedero> | i see that the whatwg specs do in fact have the space as well |
| 19:42 | <gsnedders> | smedero: WHATWG has for years called it Web Applications 1.0, and the combination of that with WF2 has often been called "HTML 5". The spec noted it was often called that too |
| 19:42 | <smedero> | though I swore they didnt always |
| 19:42 | <smedero> | not that this matters one damn bit |
| 19:42 | <smedero> | hober: that's a great question... I was lead to believe that a consensus had been reached in 2007 to undertake that work... |
| 19:42 | <smedero> | (within the w3c) |
| 19:42 | <smedero> | perhaps i misunderstood |
| 19:43 | <BenMillard> | hsivonen, I'm also confused by the removal of .xml when .html and .htm are retained. |
| 19:44 | <smedero> | huh, what happened to: http://www.w3.org/TR/xhtml12 |
| 19:44 | <gsnedders> | smedero: AFAIK they never published a WD… |
| 19:45 | <smedero> | yeah but there were editor's draft... |
| 19:45 | <gsnedders> | smedero: That won't be a TR |
| 19:45 | <smedero> | ahh |
| 19:45 | <smedero> | then the editor's draft is wrong |
| 19:45 | <smedero> | as it points to that URL |
| 19:45 | <gsnedders> | what ED? |
| 19:46 | <smedero> | well there this: http://www.w3.org/MarkUp/2008/ED-xhtml12-20081023/xhtml12.html |
| 19:46 | <gsnedders> | Ah, the latest stable version link always points to such a URL even when it doesn't exist |
| 19:46 | <smedero> | there's actually a newer one here: http://www.w3.org/MarkUp/2008/ED-xhtml12-20081223/ |
| 19:47 | <hsivonen> | they removed the flaming subtitle, it seems |
| 19:48 | <gsnedders> | what subtitle? |
| 19:48 | <hsivonen> | gsnedders: meaningful, accessible XHTML |
| 19:48 | <hsivonen> | or something like that |
| 19:48 | <gsnedders> | ah |
| 19:53 | <BenMillard> | hsivonen, there's a section in the Talk page on the Wikipedia article about that, too: http://en.wikipedia.org/wiki/Talk:XHTML#Change_title_for_XHTML_1.2 |
| 19:54 | <BenMillard> | it lists the entry on this page as how you find the XHTML 1.2 draft: http://www.w3.org/MarkUp/Drafts/#xhtml12 |
| 19:55 | <BenMillard> | (the date is what you click to get the spec) |
| 19:56 | <BenMillard> | hmm, XHTML Media Types updated: http://www.w3.org/MarkUp/2009/ED-xhtml-media-types-20090107/ |
| 20:07 | gsnedders | needs a word that means "not introspection". |
| 20:07 | <Philip`> | gsnedders: How about using two words? |
| 20:07 | <gsnedders> | Philip`: No. |
| 20:09 | <hsivonen> | extraspection? :-) |
| 20:10 | <BenMillard> | gsnedders, what's the context? |
| 20:10 | <Philip`> | Introspectionless? |
| 20:11 | <gsnedders> | hsivonen: It's s/a/o/, but no, that's still more specific. There are things that are neither. |
| 20:11 | <Philip`> | Antiintrospectionarianism? |
| 20:11 | <gsnedders> | BenMillard: tag for blog, to separate display of posts on the home page up |
| 20:11 | <gsnedders> | BenMillard: (Most of those tagged with introspection are also tagged with lust :)) |
| 20:12 | <BenMillard> | outgoing? |
| 20:12 | <gsnedders> | BenMillard: That too leaves things that are neither |
| 20:12 | <BenMillard> | I think I understand what you're trying to convey and are similarly struggling to figure out the word. :) |
| 20:12 | <gsnedders> | (think of code, that isn't really observing anything) |
| 20:13 | <gsnedders> | BenMillard: http://en.wiktionary.org/wiki/introspection :) |
| 20:13 | <gsnedders> | BenMillard: Anything that isn't that. |
| 20:13 | <BenMillard> | perhaps the absence of the introspection tag is sufficient to indicate it is "not introspection"? |
| 20:13 | <gsnedders> | BenMillard: Then I have the issue of a column header. |
| 20:14 | <BenMillard> | gsnedders, do you have a demo page with placeholder text for it I could see? |
| 20:15 | <gsnedders> | BenMillard: does sftp://gsnedders:/dev/head work for you? :) |
| 20:16 | <BenMillard> | gsnedders, I may have uninstalled the only app I had which supported sftp... |
| 20:16 | <gsnedders> | BenMillard: Then you have the issue of reading my head, though. |
| 20:16 | gsnedders | thinks BenMillard missed the joke |
| 20:17 | BenMillard | gives up. |
| 20:20 | gsnedders | tries to come up with a sample of what he wants, then realizes he can't decide how to mark it up ;P |
| 20:21 | <gsnedders> | Should post+summary be marked up in a dl? |
| 20:31 | <gsnedders> | BenMillard: http://secret.gsnedders.com/g6/home.html |
| 21:07 | <BenMillard> | gsnedders, <dl> makes sense there as it's a list of title+summary pairs. |
| 21:07 | <gsnedders> | Yeah, that was my conclusion |
| 21:08 | <BenMillard> | "Extrospection" seems to work in the context of it being a personal blog of a geeky person. |
| 21:09 | <hsivonen> | gsnedders: text in the top left corner overlaps other text |
| 21:09 | <gsnedders> | hsivonen: Yeah, I know there are issues. |
| 21:09 | <gsnedders> | (Heck, it's almost completely broken in Firefox, yet alone IE) |
| 21:10 | <BenMillard> | gsnedders, your current blog is like that in Fx2 (which is what I use) |
| 21:10 | <gsnedders> | BenMillard: Screenshot? |
| 21:15 | <BenMillard> | gsnedders, http://projectcerbera.com/!dev/temp/g-fx2.png |
| 21:15 | <gsnedders> | Um, that's odd. |
| 21:15 | <BenMillard> | I think it's to do with parsing unknown block-level tags |
| 21:15 | <gsnedders> | Ah. |
| 21:15 | gsnedders | continues to ignore Fx2 :) |
| 21:19 | BenMillard | returns to ignoring gnsedders's blog. :P |
| 21:20 | <BenMillard> | hmm, lunchtime |
| 21:22 | <gsnedders> | lunchtime? :\ |
| 21:22 | gsnedders | wonders where BenMillard |
| 21:22 | <gsnedders> | is |
| 21:35 | <smedero> | at lunch? |
| 22:29 | <hsivonen> | hmm. no border-radius in Opera 10 :-( |
| 22:58 | <roc> | hsivonen: no -o-border-radius? |
| 23:13 | <maodun> | Looking at the python version of html5lib - why do some of the entity names in constants.py not have versions which are without trailing semicolons. E.g., why do we gracefully parse "È" as "È" but not "&Lambda" as "Λ"? |
| 23:13 | <hsivonen> | roc: no, it seems it's not supported with -o-, either |
| 23:14 | <takkaria> | maodun: that's what the spec says |
| 23:14 | <takkaria> | maodun: some of them only work in attribute values for compat with IE, I believe |
| 23:14 | <takkaria> | or the other way round. I can't remmber |
| 23:14 | <takkaria> | anyway, it's IE-compat stuff |
| 23:15 | <maodun> | takkaria: ah, now I understand. thank you. |
| 23:20 | <maodun> | takkaria: actually, wait - so is the goal not to produce a very robust real-world parser but rather to reproduce the behavior of commonly used browsers? or is the assumption that reproducing the behavior of commonly used browsers will produce a very robust real-world parser? my intution would be that parsing &Lambda as Λ would be the more robust solution, regardless of what IE does. |
| 23:22 | <takkaria> | depends what you mean by robust |
| 23:22 | <takkaria> | if you mean handling existing content as people expect it to be handled |
| 23:22 | <takkaria> | then html5lib is pretty robust |
| 23:23 | <maodun> | that is what i mean. |
| 23:24 | <Hixie> | people expect their pages to be handled like IE handles them, typically |
| 23:24 | <maodun> | so is the thought process that, because of IE, people tend to expect things such as &Lambda to pass through? |
| 23:24 | <Hixie> | i don't mean that they think "haha! i am going to omit a semicolon and that will trigger odd behavior!" |
| 23:24 | <Hixie> | i mean they just write web pages with mistakes, test them in their browser (usually IE, since IE has more market share), and assume all is well if it works |
| 23:25 | <Hixie> | yeah, what you said |
| 23:25 | <Hixie> | see also the topic of the channel :-) |
| 23:25 | <Hixie> | web browsers often make very little sense |
| 23:26 | <maodun> | hahaha, ok, thank you for explaining, i now understand why it is implemented this way. |
| 23:26 | <jmb> | browsers make no sense. you have to be certified insane to work on one |
| 23:32 | <Hixie> | ok can anyone see anything obviously wrong with http://www.whatwg.org/specs/web-apps/current-work/.ietf-websocket-protocol/draft-hixie-thewebsocketprotocol-00 other than the two missing sections? |
