| 00:02 | <Hixie> | with opera's announcement of their new scripting engine, i guess IE is the only one still using a JS engine from the last century |
| 00:03 | <othermaciej> | it will make JS performance shootouts just look laughable |
| 00:14 | <roc> | where's this announcement? |
| 00:14 | <wilhelm> | http://my.opera.com/core/blog/2009/02/04/carakan |
| 00:15 | <roc> | sound good |
| 01:30 | <Hixie> | heycam: regarding the style sheet, i really don't want to go there. historically, we've had problems whenever specs don't stick closely to the official style sheet (e.g. CSS1's marble background). It also dilutes the W3C brand if the specs have different styles. I would recommend instead asking the pubteam to update their styles, so that all specs could benefit, not just hmtl. |
| 01:31 | <takkaria> | http://takkaria.org/life.html works in Safari and Opera now |
| 01:34 | <Hixie> | takkaria: what are the survival and birth lines? |
| 01:41 | <takkaria> | they control the conditions of the game |
| 01:42 | <takkaria> | survival 2 and 3 means that grids with 2 or 3 neighbours survive |
| 01:42 | <takkaria> | birth 3 means that areas with three neighbours get reborn |
| 01:42 | <takkaria> | see http://en.wikipedia.org/wiki/Life-like_cellular_automaton |
| 01:43 | <takkaria> | there are all kinds of funky things you do |
| 01:43 | <Hixie> | ah, it's a count, ok |
| 01:43 | <Hixie> | i thought it was the direction, which confused me |
| 01:44 | <takkaria> | setting everything to full and then turning some off gives a nice symmetry |
| 01:46 | <Hixie> | setting the survival to a high number only and birth to a low number only gives what looks like static |
| 01:51 | <Hixie> | i wonder if larry wants <font> and <frameset> back too |
| 02:24 | Hixie | stares at the cliff that is defining how <frameset> works, and ponders how to climb it |
| 02:29 | <Hixie> | wtf, mozilla does things in framesets differently based on quirks mode |
| 02:50 | <Philip`> | http://lists.w3.org/Archives/Public/www-archive/2009Feb/att-0032/image006.jpg doesn't seem like especially appropriate UI |
| 04:50 | <heycam> | Hixie, what about the scripts on the page (dfn.js, toc.js)? |
| 04:51 | <heycam> | also, what was it that prompted you to remove the styles from the w3c copy some months ago? |
| 05:02 | <Hixie> | heycam: it seems highly unlikely that the w3c would allow me to put scripts on a page when those scripts could hijack user credentials |
| 05:03 | <Hixie> | heycam: i do not recall when the styles on the w3c copy were changed |
| 05:03 | <Hixie> | or why |
| 05:03 | <Hixie> | but in any case i really would much rather all of this was done w3c-wide than have html5 be special cased |
| 05:04 | <heycam> | Hixie, yeah i suppose those scripts could mess around with member-private webapps on w3.ogr |
| 05:04 | <Hixie> | they could snoop around staff-only mailing lists, too |
| 05:04 | <heycam> | right |
| 05:05 | <heycam> | so i don't think it's a matter of html5 being special cased |
| 05:05 | <heycam> | rather that the style requirements are loose enough that most specs have differing styles |
| 05:05 | <heycam> | (but with some commonality) |
| 05:06 | <Hixie> | if the styles are lacking so much that all the specs have to work around it, they should be changed |
| 05:07 | <heycam> | yes, most moderately complex specs do need to have some local styles to produce nice looking content |
| 05:07 | <heycam> | and it would be ideal if the site-wide styles handled these cases |
| 05:07 | <heycam> | but fixing that seems to be a much bigger job than changing this spec |
| 05:07 | <heycam> | basically to what it was like before |
| 05:08 | <heycam> | brb |
| 05:08 | <Hixie> | i'm not looking for a quick fix, i'm looking for a good fix :-) |
| 06:27 | <Hixie> | IE makes no sense |
| 06:27 | <Hixie> | <frameset cols="1.5*,1.5*"> |
| 06:27 | <Hixie> | is treated the same as cols="1.5*,1* |
| 06:27 | <Hixie> | " |
| 06:27 | <Hixie> | decimals on the number of the last entry in the list are ignored |
| 06:56 | <jwalden> | only the last, or anything beyond the first? |
| 07:02 | <Hixie> | only the last |
| 07:02 | <Hixie> | also, i can't believe i'm speccing framesets. what i jip. |
| 07:02 | <Hixie> | what a jip, even |
| 07:07 | heycam | wonders if "jip" is considered more un-PC than "gyp" |
| 07:08 | <heycam> | or if the evolution in spelling has made it different enough from "gypsy" |
| 07:09 | <heycam> | s/more/as/ |
| 07:09 | <heycam> | and s/than/as/ :) |
| 07:11 | jwalden | resists getting into a discussion about slur-ish words or phrases no longer being offensive in that manner when basically nobody thinks about the connotation when they say it |
| 07:20 | <Hixie> | my three-legged cat just brought in a big-ass dead crow. |
| 07:20 | <Hixie> | good times. |
| 07:22 | <nessy> | sounds like the house of a wizard |
| 07:33 | <Hixie> | a highly incompetent wizard, if these cats are anything to go by :-) |
| 07:33 | <Hixie> | they're ridiculously adorable, but they're not the most competent at the stereotypical wild cat behaviours |
| 07:33 | <Hixie> | to put it mildly |
| 07:43 | <Hixie> | oh lord |
| 07:43 | <Hixie> | IE distinguishes between <frameset frameborder> and <frameset frameborder=> |
| 07:48 | <hsivonen> | Perhaps it's just me, but I see the Dreamweaver profile screenshots as arguments *against* profile. |
| 08:50 | <annevk> | ooh fun, we're back to @profile |
| 08:50 | <annevk> | and around it goes |
| 08:55 | <takkaria> | I find it kinda sad that rsayre is talking about things being "no worse than the status quo" |
| 08:56 | <hsivonen> | rel=profile misses the point by a mile |
| 08:56 | <Hixie> | so uh, can anyone explain what noresize does exactly on a <frame> element? |
| 08:56 | <hsivonen> | Swedish mile, even :-) |
| 08:56 | <hsivonen> | (Swedish one being longer than English) |
| 08:57 | <annevk> | Hixie, presumably it prevents the user from resizing the frame |
| 08:57 | <annevk> | Hixie, but it's been a while since I loved frames and knew all the details :) |
| 08:57 | <Hixie> | which frames, exactly? it seems to affect more than just the borders around the frame. |
| 08:58 | <annevk> | sorry |
| 08:58 | <Hixie> | lord, html4 is so bad |
| 08:58 | <Hixie> | "When present, this boolean attribute tells the user agent that the frame window must not be resizeable" |
| 08:58 | <Hixie> | there's a must! |
| 08:58 | <Hixie> | i've no idea what the must means, but there is one! |
| 08:58 | <Hixie> | an implementation requirement, no less |
| 08:58 | <annevk> | http://www.htmlcodetutorial.com/frames/_FRAME_NORESIZE.html |
| 08:59 | <Hixie> | (...for rendering) |
| 08:59 | <annevk> | guess you can't implement HTML4 on top of a table then |
| 08:59 | <annevk> | or what was that joke again? |
| 08:59 | <Hixie> | i already read that page |
| 08:59 | <Hixie> | and no, a table can't resize anything |
| 08:59 | <Hixie> | so that's fine |
| 08:59 | <Hixie> | it still complies |
| 09:01 | Hixie | tries to get IE to let him resize a frame, any frame, so that he can test which frames IE doesn't let him resize when he sets the attribute |
| 09:02 | <Lachy> | Hixie, IIRC, noresize prevents the user from moving any frame border surrounding that frame. |
| 09:02 | <Hixie> | http://software.hixie.ch/utilities/js/live-dom-viewer/?%3Cframeset%20cols%3D%221*%2C2*%2C1*%22%3E%3Cframe%20src%3Ddocument%3E%3Cframeset%20rows%3D%221*%2C1*%22%3E%3Cframe%20src%3Ddocument%3E%3Cframeset%20cols%3D%221*%2C1*%22%3E%3Cframe%20src%3Ddocument%3E%3Cframe%20src%3Ddocument%3E%3C%2Fframeset%3E%3C%2Fframeset%3E%3Cframe%20src%3Ddocument%3E%3C%2Fframeset%3E |
| 09:02 | <Hixie> | Lachy: it seems to affect other borders too, especially in the case of nested framesets |
| 09:02 | <Hixie> | it may be that i decide that's a bug i'm going to ignore, though |
| 09:04 | <Lachy> | which other borders? In my tests, it only affected borders of that frame itself |
| 09:04 | <Hixie> | (nsFrameSetFrame.cpp has some pretty ridiculous code) |
| 09:04 | <Hixie> | Lachy: let's see |
| 09:05 | <Hixie> | Lachy: first, in IE, none of the borders in that example resize for me at all. |
| 09:06 | <Hixie> | oh, hm, no, i think you're right (testing webkit here) |
| 09:06 | <Hixie> | ok |
| 09:06 | <Lachy> | I'm testing Firefox and Opera |
| 09:06 | <Lachy> | Opera seems to be having trouble dealing with frames in live dom viewer |
| 09:06 | <Lachy> | I got it to crash |
| 09:07 | <Lachy> | gotta go, back soon. |
| 09:08 | <Hixie> | yeah it crashed for me early on and i didn't go back :-) |
| 09:09 | <annevk> | like anyone cares if <frameset> makes a browser crash |
| 09:09 | <annevk> | we call it a feature |
| 09:10 | <Hixie> | crashes are often security bugs, and are at the very least denial of service bugs |
| 09:11 | <Hixie> | woohoo, i finished speccing frames |
| 09:15 | <annevk> | your sarcasm detector is in need of sleep :) |
| 09:15 | <Hixie> | :-P |
| 09:15 | <Hixie> | i blame frames |
| 09:18 | <annevk> | wow, does Dreamweaver really use "browse" for profile="", that's funny |
| 09:21 | <Hixie> | hsivonen: <font> isn't obsolete for ideological reasons, it's obsolete because it leads to media-dependent page design. |
| 09:22 | <Hixie> | (which leads to a poor user experience) |
| 09:23 | <hsivonen> | Hixie: you get the same design unless you can send media independence ninjas to remove a font color picker from every authoring tool starting with dreamweaver and word |
| 09:24 | <Hixie> | that's why <font> was allowed for wysiwyg uas for a whole |
| 09:24 | <Hixie> | people (including you) didn't like that approach |
| 09:24 | <Hixie> | s/whole/while/ |
| 09:24 | <hsivonen> | I didn't, yes. |
| 09:24 | <hsivonen> | But it's not like people who don't use GUI tools would do <font face> instead of CSS these days |
| 09:24 | <Hixie> | right now the state of the art in editor UA design unfortunately doesn't really have answers for writing accessible media-independent documents. |
| 09:25 | <hsivonen> | so I think it's unlikely that allowing wysiwygism as conforming would regress non-wysiwyg authoring practices |
| 09:25 | <Hixie> | i wish i could share your optimism |
| 09:25 | <annevk> | making <font> conforming would make many people upset |
| 09:26 | <annevk> | in whatever form you pick |
| 09:26 | <hsivonen> | annevk: yes |
| 09:26 | <annevk> | as it is now only a few people complain |
| 09:26 | <annevk> | that's not really good criteria though :) |
| 09:26 | <hsivonen> | annevk: so is the policy for HTML4 features checking the level of complaints? |
| 09:27 | <Hixie> | that isn't the criteria for any feature, as far as i'm concerned |
| 09:28 | <jwalden> | annevk: dunno if you can answer, but why doesn't 10a1 include <video> support? there was a test build way back that did support it, and I was surprised to see it gone when I picked up 10a1 recently |
| 09:29 | <annevk> | jwalden, our test build only had the simple features, not the advanced API |
| 09:29 | <jwalden> | ah |
| 09:29 | <annevk> | jwalden, we are in process of completing our implementation though and it will likely be in the one after 10 though you never know :) |
| 09:29 | <jwalden> | I can understand that, wanting to do it right rather than halfway |
| 09:30 | <jwalden> | cool |
| 09:32 | virtuelv | quite liked the idea of training an army of media/device independence ninjas |
| 09:50 | <annevk> | http://code.google.com/p/es-operating-system/ |
| 09:52 | <Hixie> | i'm mostly down to form controls now |
| 09:52 | <Hixie> | but i guess i'll do those tomorrow |
| 09:56 | <annevk> | Hixie, wouldn't blockquote, figure { margin:1em 40px } be better? |
| 09:57 | <annevk> | seems to me <section> also needs margin:1em 0 or equivalent |
| 09:57 | <Lachy> | hmm, Rob's suggesting the use of aural stylesheets. I wasn't aware that there were any implementations of those in screen readers. |
| 09:59 | <annevk> | I thought Emacspeak did |
| 10:01 | jgraham | wonders how many users emacsspeak actually has |
| 10:02 | <jgraham> | Although I guess there might be a strong correlation with people interested in reading the HTML5 spec |
| 10:05 | <Lachy> | the other issue is that letting authors specify aural presentation isn't likely to end up with great results because the vast majority have very little understanding of the concepts involved and are unlikely to know enough to specify appropriate values |
| 10:06 | <Lachy> | it's better to leave such things to the experienced people developing the text to speech systems |
| 10:06 | <jgraham> | Indeed. |
| 10:28 | gsnedders | finally bothers to install Py3 |
| 10:57 | <jgraham> | gsnedders: For any good reason? |
| 11:05 | <annevk> | I'm not really sure what to tell RB. XML does codepoint comparison, not Unicode Normalization, period. See all XML parsers... |
| 11:15 | <jgraham> | annevk: Pointer? Can you just ask him for proof by example? e.g. make a testcase and show how expat, libxml2, etc act? |
| 11:15 | <jgraham> | See if he can find any library that does normalization? |
| 11:16 | <annevk> | he's arguing from his interpretation of the Unicode spec... |
| 11:16 | <annevk> | http://lists.w3.org/Archives/Public/www-style/2009Feb/0149.html |
| 11:16 | <Lachy> | annevk, I wouldn't even bother trying to convince him. |
| 11:16 | <annevk> | it's like debating the semantics of HTML4 tables with him, I've the feeling |
| 11:17 | <annevk> | yeah, wasn't planning spending more time on this one, apart from maybe asking the XML Core WG, though maybe zcorpan can do that now :) |
| 11:18 | <Lachy> | when did zcorpan join the XML Core WG? |
| 11:18 | <jgraham> | Lachy: This week. Yesterday maybe? |
| 11:20 | <zcorpan> | hmm, i thought display:table would be a nice default style for <figure> |
| 11:21 | <zcorpan> | Hixie: ^ |
| 11:24 | <zcorpan> | http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C!DOCTYPE%20html%3E%0D%0A%3Cstyle%3Efigure%20%7B%20display%3Atable%3B%20margin%3A0%2040px%20%7D%20figure%20%3E%20div%20%7B%20text-align%3Acenter%20%7D%3C%2Fstyle%3E%0D%0A%3Cp%3EFoo%20bar%20baz%3C%2Fp%3E%0D%0A%3Cfigure%3E%3Cimg%20src%3Dimage%3E%3Cdiv%3EFoo%20bar%20baz%3C%2Fdiv%3E%3C%2Ffigure%3E%0D%0A%3Cp%3EFoo%20bar%20baz%3C%2Fp%3E |
| 11:28 | <Lachy> | zcorpan, styling figure with display:table and legend with display:table-caption has some nice benefits, like being able to set caption-side too |
| 11:28 | <zcorpan> | Lachy: having display:table-caption by default makes it always appear on top by default |
| 11:29 | <zcorpan> | unless there's a rule figure > legend:last-child { caption-side:bottom } |
| 11:30 | <annevk> | maybe display:inline-block ? |
| 11:30 | <zcorpan> | annevk: for figure? |
| 11:30 | <annevk> | ja |
| 11:30 | <zcorpan> | i wouldn't expect it to render inline with other text... |
| 11:33 | <zcorpan> | actually it would maybe be nice if figure shrink-wrapped ignoring the <legend> |
| 11:33 | <zcorpan> | because i think usually the legend is no wider than the figure |
| 11:37 | <annevk> | maybe figure { width:min-intrinsic } |
| 11:37 | <annevk> | or whatever the latest version is |
| 11:44 | heycam | ponders the significance of the destinations listed on http://70.32.90.124/images/ajira_boardpass.jpg |
| 11:46 | <annevk> | http://tantek.com/log/2006/06.html#d17t2231 via http://www.w3.org/2009/Talks/02wdn/slides |
| 12:40 | <myakura> | it's kinda surprising, but rather interesting to see that <h1> isn't closed by <p> in webkit http://dev.w3.org/2006/webapi/selectors-api/disposition-of-comments.html |
| 12:41 | <myakura> | or html5 defines so? |
| 12:41 | <annevk> | Validator.nu does that too |
| 12:42 | <myakura> | interesting. |
| 12:42 | <annevk> | seems correct per HTML5 on first glance |
| 12:43 | <jgraham> | html5 changed here recently |
| 12:43 | <zcorpan> | ie nests the p |
| 12:43 | <zcorpan> | jgraham: i thought that was about nested headings |
| 12:45 | annevk | changes topic to 'WHATWG (HTML5) -- http://www.whatwg.org/ -- Logs: http://krijnhoetmer.nl/irc-logs/ -- Please leave your sense of logic at the door, thanks!' |
| 12:45 | <jgraham> | zcorpan: Oh, you could be right |
| 12:45 | zcorpan | fails to see where http://dev.w3.org/2006/webapi/selectors-api/disposition-of-comments.html has <h1><p> |
| 12:47 | <zcorpan> | myakura: ah, it has <h1></h2> |
| 12:47 | <zcorpan> | myakura: the </h2> should close the h1 |
| 12:48 | <zcorpan> | surprising that webkit fails to do so :) |
| 12:48 | <myakura> | zcorpan: :) |
| 13:15 | <Lachy> | oh, I'll fix that |
| 13:16 | <Lachy> | fixed |
| 13:42 | <Philip`> | "Reducing keystrokes helps with legibility of the code" - that's why Perl is so much more legible than Python |
| 13:47 | <myakura> | compares "getElementsBy*******" to "$" |
| 13:47 | <myakura> | yeah that makes sense :) |
| 13:51 | <jgraham> | The proposal seems to have the rather significant disadvantage of being incompatible with DOM and CSS |
| 14:48 | <annevk> | is making requirements on servers useful? |
| 14:48 | <annevk> | i guess it's the same as author reqs |
| 14:48 | <annevk> | opinions? |
| 14:50 | <Lachy> | annevk, in which spec? |
| 14:50 | <Lachy> | I'm guessing you mean in CORS |
| 14:52 | <annevk> | yes |
| 14:52 | <Lachy> | sometimes it's appropriate to specify server requirements, depending on the circumstances |
| 14:52 | <Lachy> | HTTP specifies server requiements, for instance |
| 14:52 | <annevk> | i guess i'll just do it and see what happens |
| 15:20 | <hsivonen> | annevk: the Origin draft specifies requirements on servers, FWIW |
| 15:22 | <annevk> | yeah, I'm doing it too now |
| 15:24 | <annevk> | still need to say that UAs actually need to use the ABNF for parsing and such |
| 15:25 | <annevk> | and fix a bunch of other non-normative things :/ |
| 15:26 | <annevk> | i don't really like non-normative stuff; it's hard to tell whether it's correct |
| 15:28 | hsivonen | wonders if Google Latitude has anything to do with the Jaiku team and whether it works together with Yahoo! FireEagle |
| 15:33 | <hsivonen> | I wonder if Philip TAYLOR realizes that my native language isn't invariant under Unicode normalization |
| 15:34 | <gsnedders> | jgraham: Writing from the ground new code; need stuff only in 2.6+, so may as well go for 3 |
| 15:41 | <Lachy> | I wonder if anyone has done a study to see how many class names and IDs use non-ASCII characters, and if any, how many of those potentially suffer from normalisation issues |
| 15:43 | <Lachy> | Philip`, http://canvex.lazyilluminati.com/survey/2007-07-17/analyse.cgi/index appears to be broken. Is there a more updated version of that survey that's isnt broken? |
| 15:43 | <gsnedders> | jgraham: How are you getting on with html5lib and Py3? |
| 15:44 | <Philip`> | Lachy: Oops, that shouldn't be broken |
| 15:45 | <Philip`> | Oh, yes it should |
| 15:45 | <Philip`> | Well, it wasn't directly intentional |
| 15:46 | <Philip`> | (I turned off Postgres because I didn't think I was using it for anything and wanted to save memory) |
| 15:46 | <Philip`> | Lachy: It should work now |
| 15:47 | <Lachy> | Philip`, does that one have all the most recent data, despite having 2007-07-17 in the URL? |
| 15:47 | <Philip`> | Lachy: No, it has data from 2007-07-17 |
| 15:48 | <Lachy> | where's the latest one? |
| 15:48 | <Philip`> | Lachy: See e.g. the first paragraph |
| 15:48 | <Philip`> | Lachy: It's not online |
| 15:48 | <Lachy> | oh, I thought you had published more recent data |
| 15:48 | <Philip`> | (That online database thing is unscalable when I have 130K pages) |
| 15:48 | <Lachy> | that doesn't appear to have collected any data on class and id values |
| 15:48 | <jgraham> | gsnedders: html5lib seems to mostly work in python 3 for some definition of mostly which means "all the bits I cared enough to get working and only if you don't do certian things like try to pass it a file not opened in binary mode" |
| 15:49 | <Philip`> | Lachy: I've published random bits of data, a few indexed at http://philip.html5.org/data.html and the rest just mentioned on IRC or mailing lists |
| 15:49 | <Philip`> | Lachy: I could get a list of class and id attribute values from my latest data fairly easily |
| 15:50 | <Lachy> | ok, I'm specifically looking for an answer to my previous question about non-ASCII values and potential normalisation issues |
| 15:51 | <Philip`> | Lachy: It would probably be hard to get meaningful results, since I don't have many pages from interesting languages |
| 15:53 | <Lachy> | ok |
| 17:28 | <Philip`> | Hmm, maybe making an XML file containing an <attribute uri="..." element="..." name="..." value="..."/> for every single attribute in all my pages wasn't the bestest idea ever |
| 17:29 | <Philip`> | since the output file is about 5GB |
| 17:34 | <Philip`> | Hooray for gzip |
| 17:34 | <Philip`> | Now only ~300MB, and only takes about thirty seconds to grep |
| 17:45 | <Philip`> | Lachy: Out of 130K pages, I see 277 with non-ASCII id values, and 260 with non-ASCII class values |
| 17:46 | Philip` | checks whether any of his code is silently normalising stuff |
| 17:59 | <rubys1> | "The server at wiki.whatwg.org is taking too long to respond." -- anybody else see this? |
| 17:59 | <takkaria> | aye |
| 18:01 | <yorick> | why isn't event.dataTranfer available during a dragover event? |
| 18:08 | Philip` | finds that his code wasn't normalising stuff, but he was forgetting to decode the input as UTF-8 before testing whether it was normalised |
| 18:08 | <Philip`> | Lachy: As far as I can see, all of the non-ASCII id and class values are in NFC |
| 18:09 | <Philip`> | (Some are also in NFD, and NFKC, and NFKD, but they're all still NFC too) |
| 18:22 | <Philip`> | Lachy: http://philip.html5.org/data/non-ascii-class-values.txt / http://philip.html5.org/data/non-ascii-id-values.txt |
| 18:24 | <Philip`> | (Obvious problems are the lack of data for most interesting languages, and frequent encoding detection failures) |
| 22:38 | Hixie | considers defining the rendering section using xbl |
| 22:43 | <Lachy> | Hixie, for which elements would you need XBL? |
| 22:43 | <Hixie> | <details>, all the form controls... |
| 22:43 | <Lachy> | ok |
| 23:07 | <Philip`> | That doesn't sound like necessarily the most helpful way of defining it, since few people know XBL and nobody's going to really implement it with XBL |
| 23:08 | <Hixie> | actually implementing it with xbl has been on the long term plans for mozilla for years now |
| 23:08 | <Philip`> | and I'd guess it introduces constraints that interfere with the likely goal of making the definition understandable |
| 23:09 | <Hixie> | i have an idea of a half-way solution that might work |
| 23:09 | <Philip`> | "long terms plans ... for years now" doesn't convince me that anybody is really going to do that ;-) |
| 23:10 | <Hixie> | the same could have been said for the rendering section in html5 |
| 23:10 | <Hixie> | yet here we are |
| 23:12 | <Philip`> | One example of long term plans being carried out still doesn't convince me that other long term plans will be :-) |
| 23:12 | <Hixie> | :-) |
| 23:16 | <Lachy> | it's unfortunate that browser vendors haven't started implementing XBL yet. |
| 23:18 | <Philip`> | Presumably it's much less important to their customers than rewriting their JS and regexp engines yet again |
| 23:21 | <Lachy> | yeah, making existing web sites work better is generally a higher priority than providing new toys for developers to play with |
| 23:22 | <Lachy> | I wish Philip TAYLOR (not you Philip`) would be more specific about which features he wants to know the justifications for |
| 23:22 | <Hixie> | all of them |
| 23:23 | <Philip`> | Lachy: If he was specific, that might result in someone spending time finding and explaining the justifications to him, so maybe it's best if he's not specific |
| 23:23 | <Lachy> | well, ok, but writing up all of that would be a huge effort. I mean, he could at least let us know which ones he wants to know the most so we can priorise |
| 23:23 | <Lachy> | *prioritise |
| 23:24 | <Lachy> | *prioritize |
| 23:24 | Lachy | can't spell today |
| 23:25 | <Hixie> | i wish the people complaining would volunteer to do something once in a while |
| 23:27 | <Hixie> | Philip`: see the spec now to see what i ended up doing (w.r.t. xbl) |
| 23:28 | <Philip`> | Hixie: Oh, I just wanted to complain, I didn't want to actually look at the spec and give constructive feedback |
| 23:28 | <Hixie> | :-) |
| 23:29 | <Hixie> | dude the other pt is leaking into you |
| 23:29 | <Hixie> | careful |
| 23:29 | <Hixie> | :-P |
| 23:29 | <heycam> | Hixie, did something happen to the whatwg.org styles too? |
| 23:29 | <Hixie> | ? |
| 23:30 | <Hixie> | lots of things have happened, can you be more specific? :-) |
| 23:30 | <heycam> | the background green of element definitions was gone, and the background white of the status marker boxes on the left was transparent |
| 23:30 | <heycam> | but i just reloaded the page and it's fine... |
| 23:30 | <Hixie> | oh i temporarily had a stray <style> element in the document |
| 23:31 | <heycam> | oh, k |
| 23:31 | <Hixie> | you happened to reload during the 5 seconds that it was broken |
| 23:31 | <heycam> | =) |
| 23:33 | <Lachy> | Hixie, is the details binding link supposed to link to a binding that hasn't yet been written? It's currently linking to the element, which is confusing |
| 23:33 | <Hixie> | oops, not supposed to be a link at all |
| 23:36 | <Lachy> | ok, so you're just going to rely in the description of the binding in the prose then? |
| 23:36 | <Lachy> | and the binding itself is imaginary |
| 23:36 | <Hixie> | yeah |
| 23:38 | <Hixie> | woo, the end is in sight |
| 23:38 | <Hixie> | only a few more days of this |
| 23:38 | <Lachy> | really? I thought you once told me the rendering section was going to be huge and a significant amount of effort. I didn't think that meant just a week's worth of work |
| 23:38 | <Hixie> | i split the rendering section in two, rendering and obsolete elements |
| 23:39 | <Lachy> | oh |
| 23:39 | <Hixie> | a lot of the effort went into the obsolete elements section |
| 23:39 | <Hixie> | also, this IS a lot of work! :-P |
| 23:39 | <Hixie> | have you seen the frameset section? |
| 23:39 | <Hixie> | sheesh |
| 23:39 | <Hixie> | :-P |
| 23:39 | <Hixie> | (i'm getting pretty good at cranking out this stuff; a few years ago, this would be taking me way longer) |
| 23:40 | <Lachy> | yeah, I noticed you've been getting faster. I've been having difficuty keeping up with many of your changes |
| 23:41 | <Lachy> | *slow down!* |
| 23:41 | <Philip`> | Lachy: Maybe you're just slowing down in your old age |
| 23:43 | <Hixie> | don't worry, you'll get better at keeping up :-P |
| 23:43 | <Lachy> | I'm not old, you young whipper snapper!... Oh, and Get Off The Lawn! ;-) |
| 23:49 | <Hixie> | didn't windows have dialogs with a button at the bottom that said "More details" or something which when clicked would grow the dialog and show more details? |
| 23:49 | <Hixie> | i can't find any examples in my xp install anymore |
| 23:49 | <Lachy> | OS X have those |
| 23:50 | <Lachy> | they're called disclosure triangles |
| 23:50 | <Philip`> | Hixie: Like http://www.juergentreml.de/wp-content/uploads/2008/01/long-copy.jpg ? |
| 23:50 | <Lachy> | I think many apps on windows do lots of different things for that kind of UI |
| 23:52 | <Philip`> | Hixie: Also like http://i.technet.microsoft.com/Cc700803.authen01(en-us,TechNet.10).gif ? |
| 23:52 | <Hixie> | the latter is the closest to what i meant |
| 23:52 | <Hixie> | i specifically don't mean actual triangles though |
| 23:53 | <Hixie> | i mean a button with text in it |
| 23:53 | Philip` | doesn't remember any dialogs where that text is actually inside the button |
| 23:55 | <Hixie> | well then good, that means i don't need anything but an example of a triangle in the spec |
| 23:56 | <Philip`> | Then again, I've never actually used XP on my own computers, so my memory probably isn't a good guide |
| 23:57 | <Hixie> | bbiab |
.gif){kind=link}