/irc-logs / w3c / #html-wg / 2007-05-07 / end

Options:

# Session Start: Mon May 07 00:00:00 2007
# Session Ident: #html-wg
# [00:05] * Quits: hyatt (hyatt@24.6.91.161) (Quit: hyatt)
# [00:25] * Quits: heycam (cam@203.214.12.71) (Ping timeout)
# [00:34] * Joins: hyatt (hyatt@24.6.91.161)
# [00:51] * Quits: hyatt (hyatt@24.6.91.161) (Quit: hyatt)
# [00:51] * Joins: schepers (schepers@72.29.239.177)
# [00:54] <zcorpan_> so... much... noise... on... public-...html... must... breathe
# [00:56] <Dashiva> And 90% of it is echoed on www-html...
# [00:56] <zcorpan_> glad i'm not subscribed to www-html then
# [00:57] <jgraham> Philip`: Dunno. I'll file a bug
# [00:57] <Philip`> Any ideas of some signal to feed in, to help the ratio and maybe give people some idea of what would be actual useful discussion?
# [00:58] <Dashiva> I'm on the "em and i are used with no real difference, so just stop pretending they're different" side
# [01:00] <Philip`> jgraham: Okay, thanks - I just deleted that page from my database so I don't have to worry about it any more :-)
# [01:03] <jgraham> Hmm. I wonder how much signal I would loose just setting up a filter to delete all email containing the word "emphasis"?
# [01:04] * jgraham has no intention of actually doing that
# [01:04] <Dashiva> It would wipe out most of the mail, so you might end up with SNR undefined :)
# [01:07] <jgraham> Philip`: Issue 35 filed http://code.google.com/p/html5lib/issues/detail?id=35
# [01:14] * Joins: heycam (cam@130.194.72.84)
# [01:17] * Parts: hasather (hasather@81.235.209.174)
# [01:18] <Philip`> I'll see if I can reproduce the issue more easily...
# [01:19] * Quits: edas (edaspet@88.191.34.123) (Client exited)
# [01:19] * jgraham wonders if the <em>/<i> debate could be avoided for HTML6 by specifying default presentations in multiple media
# [01:20] <jgraham> Of course, if this debate doesn't stop soon we're all going to be using Silverlight on Windows and there won't /be/ a HTML6
# [01:21] <Dashiva> At this pace, we'll all be dead before HTML6 would come into being anyway
# [01:21] <jgraham> s/6/5/
# [01:21] <jgraham> ;)
# [01:23] <Philip`> The WHATWG could make HTML6 and then implement an HTML6 web browser inside Silverlight so it can run on IE without Microsoft having to get involved
# [01:24] <Dashiva> Replace the entire computer with a thin silverlight client?
# [01:26] <Philip`> I like how the 'customer quotes' on the Silverlight page is from an online advertising company - I think I'd be perfectly happy if I couldn't see their content
# [01:27] * Philip` wonders if it'd be useful as a way to implement <video> and <canvas>
# [01:28] <wilhelm> Sounds painful.
# [01:40] * Quits: heycam (cam@130.194.72.84) (Quit: bye)
# [01:40] * Joins: heycam (cam@130.194.72.84)
# [01:46] * Joins: MrNaz (Naz@203.214.95.166)
# [01:49] * Quits: schepers (schepers@72.29.239.177) (Quit: Free at last!)
# [01:49] <Philip`> Wouldn't people with aural browsers want <i> to be rendered in a different way to normal text so that they can disambiguate sentences like "<i>The Thing</i> was interesting" to the same extent that people with visual browsers can? Throwing away that information seems worse than rendering it in a potentially unsuitable tone of voice
# [01:51] <Philip`> (in which case it'd be almost precisely identical to <em>, and everyone would just deduce the real semantics in their heads because that's what heads are good for)
# [01:57] <jgraham> Philip`: I don't know. If <i> and <em> are really treated identically by all UAs then I don't see how anyone can argue the distinction is anything but academic
# [01:58] * Joins: hyatt (hyatt@24.6.91.161)
# [01:58] <jgraham> I think TV Raman said he had his UA set up like that but I think someone else said they didn't
# [02:02] <Philip`> Would people stop arguing over it just because it's only academic?
# [02:04] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
# [02:10] * Joins: gavin_ (gavin@74.103.208.221)
# [02:19] * Quits: AGraf (Ashe@213.47.199.86) (Quit: Quit)
# [02:25] * Quits: hyatt (hyatt@24.6.91.161) (Quit: hyatt)
# [02:32] * Quits: zdenko (zdenko@84.255.203.169) (Quit: zdenko)
# [02:34] * Quits: zcorpan_ (zcorpan@84.216.42.227) (Ping timeout)
# [02:55] <cying> i wish for more brevity on the html-wg mailing list
# [02:55] <cying> i feel compelled to read everything so i could be somewhat informed in raising future issues without being accidentally redundant
# [02:57] <mjs> we're having a tempest tea party
# [02:57] <mjs> cying: I'm trying to stop reading everything, as it is draining too much time
# [02:58] <cying> mjs: i'll clarify. i've been unsuccessful at reading everything, i can get to roughly 20% of the traffic
# [02:59] <cying> mjs: but i'm biding my time til something i'm knowledgable in comes up, like content authoring or graphics
# [02:59] <mjs> I actually have been reading everything
# [02:59] <mjs> I am hoping it will get easier once we have a spec
# [02:59] <mjs> and then have specific focus areas for review
# [02:59] <cying> ahhhh
# [02:59] <cying> maybe people should tag their emails
# [03:00] <cying> or some kind of communal tagging emaily thing
# [03:02] <mjs> I'm mainly trying to focus a lot until we have a spec and a smoothly running process that's leading to forward progress
# [03:02] <mjs> but it's a lot of work getting over that initial hump
# [03:03] <cying> mjs: indeed
# [03:07] * Quits: tH (r@87.102.20.82) (Quit: ChatZilla 0.9.78.1-rdmsoft [XULRunner 1.8.0.9/2006120508])
# [03:07] <cying> i'm contemplating a simple python or ruby implementation of HTML5 to serve as a model to experiment with features and behavior
# [03:08] <mjs> there's a python implementation of the parsing spec (not up to date though)
# [03:08] <cying> oooh
# [03:26] <Philip`> If you wanted to help update/fix/optimise the current html5lib, I believe that would be quite useful
# [03:27] * Joins: hyatt (hyatt@24.6.91.161)
# [03:38] * Quits: hyatt (hyatt@24.6.91.161) (Quit: hyatt)
# [04:03] * Joins: Zeros (Zeros-Elip@69.140.48.129)
# [04:03] * Quits: Zeros (Zeros-Elip@69.140.48.129) (Quit: Leaving)
# [04:04] <Hixie> man, a lot of threads have author patterns that go: A B A C A D A E A B A C A D A E A B A C A D A E
# [04:04] <Hixie> how does A have so much time!
# [04:05] <Hixie> and how come the non-A viewpoints are in such great majority and so coordinated in their responses!
# [04:05] * Joins: hyatt (hyatt@24.6.91.161)
# [04:05] <Hixie> are you guys taking turns fielding e-mails or something?
# [04:12] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
# [04:17] * Joins: gavin_ (gavin@74.103.208.221)
# [04:20] <cying> Philip`: what's html5lib? the python parser?
# [04:21] * Joins: dbaron (dbaron@72.29.239.177)
# [04:22] <cying> Hixie: i'm actually an amalgam of doug schepers, john boyer, murray maloney, and you. ;-)
# [04:22] * Joins: kazuhito (chatzilla@142.59.90.219)
# [04:23] <Philip`> cying: Yes - http://code.google.com/p/html5lib/
# [04:23] * Quits: hyatt (hyatt@24.6.91.161) (Quit: hyatt)
# [04:23] <cying> Hixie: maciej is also jim jewett and james graham
# [04:23] <cying> Philip`: ooh i'll have a look
# [04:24] <cying> Philip`: ooooh, very nice and clean
# [04:37] * Joins: Zeros (Zeros-Elip@69.140.48.129)
# [04:56] * Joins: hyatt (hyatt@24.6.91.161)
# [05:21] * Quits: kazuhito (chatzilla@142.59.90.219) (Ping timeout)
# [05:28] * Joins: kazuhito (chatzilla@142.59.90.219)
# [05:59] * Quits: hyatt (hyatt@24.6.91.161) (Quit: hyatt)
# [06:00] * Joins: myakura (myakura@125.194.247.196)
# [06:07] * Joins: hyatt (hyatt@24.6.91.161)
# [06:08] * Quits: dbaron (dbaron@72.29.239.177) (Quit: 8403864 bytes have been tenured, next gc will be global.)
# [06:19] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
# [06:21] * Quits: hyatt (hyatt@24.6.91.161) (Quit: hyatt)
# [06:24] * Joins: gavin_ (gavin@74.103.208.221)
# [07:23] * Joins: hyatt (hyatt@24.6.91.161)
# [07:29] <hyatt> Hixie: yt?
# [07:32] * Quits: Zeros (Zeros-Elip@69.140.48.129) (Quit: Leaving)
# [08:21] * Quits: Lachy (Lachlan@203.214.143.196) (Connection reset by peer)
# [08:21] * Joins: Lachy (Lachlan@203.214.143.196)
# [08:26] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
# [08:31] * Joins: gavin_ (gavin@74.103.208.221)
# [08:44] <Hixie> hyatt: vaguely
# [08:45] <hyatt> heh
# [08:45] <Hixie> wassup?
# [08:50] * Joins: loic (loic@90.29.107.221)
# [08:52] <mjs> cying: I don't think I have any secret identities
# [08:53] <cying> mjs: a likely story
# [09:02] * Quits: heycam (cam@130.194.72.84) (Quit: bye)
# [09:14] * Joins: schepers (schepers@208.38.57.47)
# [09:31] * Joins: zdenko (zdenko@193.77.152.244)
# [09:36] * Parts: inimino (chatzilla@75.71.88.233)
# [09:45] * Quits: cying (cying@75.41.252.252) (Quit: cying)
# [09:45] * Joins: ROBOd (robod@86.34.246.154)
# [09:48] <anne> html5lib is further than the current spec, I think, mjs
# [09:48] <anne> although it has some bugs
# [09:48] * anne sees loads of e-mails about cleaning a house
# [09:48] * anne ... spam!
# [09:49] * Joins: heycam (cam@203.214.12.71)
# [09:52] * Lachy wonders which post he could link to from a blog entry, which would give a realitively good overview of the discussion about b and i
# [09:53] <hsivonen> Lachy: one from Tina, one from Murray
# [09:53] <Lachy> I was thinking about one from mjs
# [09:54] <hsivonen> Lachy: ok. then there's mine from January on the whatwg list
# [09:54] <anne> 544 reasons to look at XForms...
# [09:54] <Lachy> I'll do one from each, I just have to find one that illustrates their points of view
# [09:54] <anne> did they suddenly start integrating with HTML4?
# [09:54] <anne> and DOM Level 2 HTML?
# [09:54] * anne missed the announcement
# [09:55] <hsivonen> anne: they? integrate? is the Forms WG now defining how to put XForms in text/html?
# [09:55] <Lachy> hsivonen, do you know specifically which one from you on whatwg?
# [09:56] <anne> hsivonen, I don't think so
# [09:56] <anne> hsivonen, but John thinks that 544 errata + new features (or 700) items make a good reason to look at XForms
# [09:57] <Lachy> this is the blog entry draft http://lachy.id.au/temp/b-and-i
# [09:58] <hsivonen> Lachy: http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2007-January/009060.html
# [09:59] <hyatt> god this is getting really really complicated
# [10:00] * hyatt beats his head against the desk
# [10:00] * hyatt should just go back to the old slow way he was doing @font-face
# [10:01] <Lachy> hsivonen, thanks
# [10:02] <hyatt> it boggles my mind that this much email has been generated over <i>
# [10:02] <Lachy> hyatt, it's called bikshedding
# [10:02] <Lachy> bikeshedding*
# [10:02] <hyatt> we could be advancing the web by developing all sorts of cool new features
# [10:02] <hyatt> but no, let's waste 10000 messages talking about... <i>.
# [10:03] <hyatt> Lachy: yeah, seriously.
# [10:03] <hyatt> some serious bike shed painting going on
# [10:08] <hyatt> yes, this works
# [10:08] <hyatt> unbelievable
# [10:08] <Lachy> what works?
# [10:08] <hyatt> @font-face
# [10:08] * hyatt has been hacking it up for webkit
# [10:09] <Lachy> oh, I didn't think it was going to get implemented in favour of something easier
# [10:09] <hyatt> meh tina pissed me off
# [10:09] <hyatt> so i implemented it :)
# [10:09] <Lachy> what happened to the idea of using font-family: url(font);
# [10:09] <hyatt> plus it proves i'll implement a spec i hate
# [10:09] <hyatt> :)
# [10:09] <hsivonen> hyatt: what are you doing about panose?
# [10:09] <hyatt> although after actually learning how @font-face works
# [10:10] <hyatt> i see that it's way better than the URL solution
# [10:10] <hyatt> the problem with the @font-face spec really is just that i don't think it needs to have as many features
# [10:10] * Quits: loic (loic@90.29.107.221) (Ping timeout)
# [10:10] <hyatt> hsivonen: nothing, not implementing that part
# [10:10] <hsivonen> :-)
# [10:11] <hyatt> i implemented (in the descriptor) font-weight, font-style, font-family
# [10:11] <hyatt> and then fully implemented src (local support, format, url)
# [10:11] <hyatt> i'll probably implement unicode-range and then declare victory
# [10:11] <hyatt> the rest just seems way too geeky to me
# [10:11] <anne> heh
# [10:12] <hyatt> it's really neat that you can override serif, monospace, etc. from the user sheet
# [10:12] <anne> says the guy who implement obscure CSS features :p
# [10:12] <Lachy> what font files will be supported? Are you adding support for Microsoft's EOT files?
# [10:12] <hyatt> Lachy: if i knew how to read them i would
# [10:12] <hyatt> Lachy: but i don't
# [10:12] <hyatt> Lachy: so .ttf
# [10:12] <Lachy> ok
# [10:12] <Lachy> cool
# [10:12] <hyatt> basically just plain old .ttf
# [10:13] <Lachy> what about .otf?
# [10:13] <hyatt> yeah that will work too
# [10:13] <Lachy> cool
# [10:13] <hyatt> truetype or opentype
# [10:13] * hsivonen hopes apple's ttf hintin engine protects against malicious fonts
# [10:13] <hyatt> most of the time those files both have .ttf extensions, which is why i just said "ttf"
# [10:13] <hyatt> hsivonen: yeah i hope so too
# [10:13] <hyatt> hsivonen: but i think these fonts cancome in through PDF right now
# [10:14] <hsivonen> ok
# [10:14] <hyatt> hsivonen: so there is already an attack vector i think
# [10:14] <hyatt> it's really cool though
# [10:14] * Joins: inimino (inimino@75.71.88.233)
# [10:14] <hsivonen> good point
# [10:14] <hyatt> while the font is downloading i return a placeholder font
# [10:14] <hyatt> that will measure
# [10:14] <hyatt> but not render
# [10:14] <hyatt> so tha i don't hold up the display of the page
# [10:15] <hyatt> WinIE just goes ahead and draws the placeholder font it picks
# [10:15] <hyatt> which looks terrible
# [10:15] <hyatt> when the real font arrives
# [10:15] <hyatt> really fun feature to work on
# [10:16] <hyatt> wonder how hard it would be to code up in gecko
# [10:16] <anne> what about all the copyright issues involved with fonts?
# [10:16] * anne thought that was mostly slowing down adoption
# [10:16] <hyatt> anne: i'm nervous about that
# [10:16] <hyatt> anne: i'm not sure if there are any issues here that i'm not aware of
# [10:17] <hyatt> anne: i.e., if the browser is liable for supporting downloadable .ttf
# [10:17] <hyatt> anne: there are thousands of free fonts on the web though
# [10:17] <anne> howcome run the font story for a while but copyright issues mostly slowed discussion down at least in the CSS WG
# [10:17] <anne> yeah, that's a fair point
# [10:17] <hyatt> really?
# [10:18] <hyatt> mostly just a concern that nobody would use it without some kind of protection?
# [10:18] <anne> I think the font people were mostly concerned
# [10:18] <anne> That people could steal fonts and such (much like people can already do with images, etc.)
# [10:19] <anne> I don't think I recall all though
# [10:19] <schepers> implement SVG fonts in HTML... the shapes of fonts aren't copyrighted, just the "program" (the format)
# [10:22] <Lachy> http://lachy.id.au/log/2007/05/b-and-i
# [10:25] <anne> can't find the discussion
# [10:25] * Lachy apologises for posting one last mail in the b and i debate
# [10:25] <hyatt> anne: i guess my attitude is that we should support an unencumbered open font format like ttf
# [10:26] * Lachy notes that it can be ignored by most people
# [10:26] <hyatt> anne: if we need to also supports some kind of lightly-encrypted mess like .eot
# [10:26] <hyatt> anne: that shouldn't preclude us from supporting .ttf
# [10:26] <hyatt> err prevent us from supporting .ttf
# [10:26] * Joins: loic (loic@90.29.152.121)
# [10:27] <hyatt> mainly i wanted ot make sure there was no weird liability issue or legal issue with a browser supporting ttf openly
# [10:27] <hyatt> everyone seems to have shied away from doing it
# [10:27] <Lachy> don't browsers already support ttf? What legal issues could there be?
# [10:28] <hyatt> well i think ttf has bits in the format itself that say "you can't embed me" or something
# [10:28] <hyatt> which i am not paying attention to at all
# [10:28] <hyatt> i'm not sure if i'm legally obligated to or something
# [10:28] <anne> if people would expose bought fonts in this way nobody would buy the fonts
# [10:29] <hyatt> right, this would obviously only work for free fonts
# [10:29] * hyatt finds the whole concept of selling fonts silly though
# [10:29] <hyatt> :)
# [10:29] * anne would be fine with the market collapsing
# [10:29] <anne> although we might get less nice fonts
# [10:29] <Lachy> ah, sounds like some primitive form of DRM
# [10:29] <hyatt> maybe the web needs a generic drm scheme
# [10:30] <anne> xrm
# [10:30] <Lachy> maybe the web needs absolutely no drm at all
# [10:30] <hyatt> :)
# [10:30] <anne> Extensible Rights Managements
# [10:31] <anne> Initially it's just from you, until someone extends it
# [10:32] <inimino> http://www.xrml.org/
# [10:34] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
# [10:34] <anne> http://piecesofrakesh.blogspot.com/2006/02/interviewing-ui-designers.html is funny
# [10:34] <anne> inimino, crap
# [10:35] <inimino> most of the bad ideas, like the good ones, are already taken
# [10:35] <Lachy> xrml is defective by design
# [10:39] * Joins: gavin_ (gavin@74.103.208.221)
# [10:39] <anne> http://flickr.com/photos/tantek/487522534/ !
# [10:39] <Lachy> I wish more interviews focussed on technical abilities like that. Most interviews I've been through focus more on you're ability to describe your own personality
# [10:41] <anne> "Gareth Hay has left the HTML Working Group sysbot+ipp@w3.org"
# [10:42] <Lachy> heh
# [10:43] <Lachy> oh well, looks like I won't need to bother sending my response to him I have in my drafts
# [10:44] <Lachy> maybe I'll just turn it into a blog entry instead
# [10:57] * hyatt is now known as hyattSopranos
# [11:01] * Lachy wonders if defining predefined class names with an _ prefix would resolve the debate
# [11:01] <Lachy> so we'd have _copyright, _note, _error, etc.
# [11:02] <Lachy> that would be consistent with predefined targat values
# [11:02] <Lachy> and Selectors can still use class selectors without escaping. so ._copyright { ... }
# [11:17] <Philip`> Either I've made a stupid mistake in my code or else zero pages out of ~2500 from a totally non-random sample use any underscore-prefixed class name
# [11:18] <Philip`> (so it seems that would avoid any conflicts)
# [11:21] <anne> i kind of like it
# [11:21] <anne> as Jonas said, it's compatible with target=
# [11:23] <hsivonen> it doesn't support bootstrapping with existing content
# [11:24] <anne> yeah, maybe i don't really care :)
# [11:24] <anne> although bootstrapping with existing content is nice
# [11:24] <anne> (we could always argue that the DOCTYPe determines the language in use :p )
# [11:25] <anne> (and that therefore people who don't trust existing content can use the DOCTYPE to determine the type of content they're looking at)
# [11:26] <anne> XHR LC2 almost out of the door, once it's published XHR2 will be started :)
# [11:28] <Philip`> Maybe authors should/must not use non-prefixed names that match the predefined ones, but content consumers may interpret non-prefixed names as equivalent to the prefixed name (so they can work with existing content, but they shouldn't do that if false positives are bad in that particular case)?
# [11:28] <Philip`> (Oops, I missed one - http://www.roomster.net/ has _box_white_green_border)
# [11:28] <anne> (that's actually relevant)
# [11:28] <anne> (although maybe not)
# [11:31] * xover likes Lachy's last on em/i...
# [11:31] * hyattSopranos is now known as hyatt
# [11:32] <Philip`> Urgh, http://www.monstertones.com/ has <section> and breaks html5lib with NotImplementedError again
# [11:33] <Lachy> xover, thanks
# [11:34] <anne> Philip`, my suggestion is that you go to the InBody state and remove <section> and go from the method dispatcher
# [11:34] <anne> Philip`, so they'll be handled by the anything else case
# [11:35] * Joins: tH (r@87.102.20.82)
# [11:35] <anne> that should be quite a trivial change
# [11:35] <Philip`> Ah, thanks, sounds like that could work - though I have to leave in about two minutes so I won't bother now and I'll probably forget by the time I get back
# [11:36] <anne> heh
# [11:36] <anne> s/go/co/
# [11:38] * Quits: loic (loic@90.29.152.121) (Ping timeout)
# [11:38] <Philip`> (I can see how these new elements are a problem for anyone who doesn't want to break the parsing of ~0.1% of sites)
# [11:42] * Joins: tH_ (r@87.102.4.173)
# [11:42] * Quits: tH (r@87.102.20.82) (Ping timeout)
# [11:42] * tH_ is now known as tH
# [11:44] <xover> Lachy: I'm like to still disagree, of course, but I still quite liked it. Clear, well-reasoned, balanced. I'm not sure that's even allowed on public-html any more. :-)
# [11:45] * Joins: hyatt_ (hyatt@24.6.91.161)
# [11:45] <xover> I suspect the entire difference in opinion boils down to the relative importance of the separation of structure and presentation.
# [11:46] * Quits: hyatt_ (hyatt@24.6.91.161) (Client exited)
# [11:46] * Quits: hyatt (hyatt@24.6.91.161) (Ping timeout)
# [11:46] * Joins: hyatt (hyatt@24.6.91.161)
# [11:47] <xover> Where I — and, I'm guessing, Tina — feel that separation is far more important than the opposing position holds.
# [11:47] <xover> Hmm.
# [11:48] <xover> There is also that (unfortunate, imo) combination of «Here's how you handle “tagsoup”» and «Here's what the language /should/ be.» in the WHATWG spec.
# [11:50] <xover> I don't think anyone would argue — if you stipulate a need to specify how to deal with “tagsoup” — that b/i shouldn't be specified for that purpose.
# [11:51] <Lachy> xover, I'm fairly sure we got past the debate about whether or not to specify processing requirements
# [11:51] <xover> IOW, I've often advocated that HTML 4.01 should be updated to drop the Transitional bits.
# [11:52] <xover> Lachy: Yeah, but I wonder whether it may not be one of the things fuelling the current i/em tempest.
# [11:53] <xover> Similar to mjs' Prescriptive/Descriptive points of view observation.
# [11:53] <Lachy> the only thing that seems to be fuelling that debate is the semantic purists who object to anything even remotely presentational, despite having clear use cases
# [11:53] <hsivonen> the fact remains that Dreamweaver and Opera emit <em> when the user asked for italic
# [11:53] <hsivonen> and Tidy encourages people to map <i> to <em>
# [11:53] <Lachy> yeah, that's unfortunate
# [11:54] <xover> hsivonen: I can predict the response to be “See why b/i shouldn't be allowed? It confuses people!”
# [11:54] <Lachy> the validator beta, which has Tidy integrated into it as a way to clean up your source, will map <i> to <em> automatically
# [11:55] <hsivonen> Lachy: clearly, that makes the documents much more semantic :-)
# [11:55] <Lachy> xover, the other alternative would be to define <em> and <strong> like <i> and <b> are currently defined
# [11:56] <Lachy> i.e. make their semantics context-sensitive
# [11:56] <xover> Hmm. I comment on the ongoing discussion, and get sucked right into it.
# [11:57] <xover> It's a black hole, and the event horizon seems to be at the edge of every communication channel the HTML WG posesses.
# [11:57] <Lachy> that's why people need to stop arguing about whether or not it should be included, and just accept the reality fo the situation
# [11:58] <xover> Lachy: My point was, the em/i debate is a proxy for the bigger, more fundamental, issues.
# [11:58] <xover> That's why they're all so hell bent on not backing down.
# [11:58] * Joins: AGraf (Ashe@213.47.199.86)
# [11:58] <Lachy> what's the more fundamental issues?
# [11:58] <xover> (me included, btw)
# [11:59] <xover> Uhm. See above?
# [12:00] <xover> I'd argue til I was blue about separation of structure and presentation; but I'm not entirely sure I care whether b/i are included.
# [12:01] <xover> Or rather, they need to be included for the “tagsoup” spec; so it just boils down to in which section of the doc they should be, and what their accompanying prose should say.
# [12:02] <xover> And that rather follows from what is eventually decided will be the level of importance assigned to separation of presentation and structure.
# [12:02] <Lachy> xover, see the last paragraph in my last email. The separation doesn't need to be strictly adhered to for the sake of it. It's only useful when it actually achieves the real goal
# [12:03] <xover> Lachy: My point exactly. /That's/ what we disagree on. b/i are just proxies in that fight.
# [12:03] <Lachy> what exactly does a strict separation in all circumstances achieve?
# [12:04] <Lachy> you have to be more pragmatic about the issue
# [12:04] <hsivonen> xover: the thing is that even if the semantic markup proponents get their way in the spec, they don't have the power to make the world comply
# [12:05] * Joins: edas (edaspet@88.191.34.123)
# [12:05] <xover> hsivonen: Don't they?
# [12:05] * Quits: hyatt (hyatt@24.6.91.161) (Quit: hyatt)
# [12:05] <xover> Have you never authored a requirements specification that mandated the conformance to som standard?
# [12:05] <Lachy> xover, if you have a real argument to dispute the more pragmatic approach to separation of presentation and semantics, I'd like to hear it.
# [12:06] <xover> Lachy: It's a big issue. One very much worth debating. ...
# [12:06] <Lachy> xover, putting something in a spec doesn't make people comply
# [12:06] <Lachy> I don't think it's worth debating, I'm just interested to hear what possible arguments there are
# [12:07] <xover> ... But it'd be as big as the current em/i thing. Which, as a point in favor, might actually kill the em/i thing now and avod similar proxy fights in the future.
# [12:08] <xover> Lachy: You don't think other points of view are wirthwhile. You just want to hear the arguemnts so you can shoot them down?
# [12:08] <hsivonen> xover: they don't
# [12:08] <Lachy> I just don't believe there are any more real arguments, just mof the same.
# [12:08] <Lachy> s/mof/more of/
# [12:09] <xover> For em/i, sure, that's going around in circles.
# [12:09] <hsivonen> xover: past experience suggests that the masses of authors aren't going to ponder the finer points of semantics
# [12:10] <xover> hsivonen: Perhaps not all authors are equally important (in this sense, please don't misconstrue this statement)...
# [12:11] <xover> There will always be a great mass of hobbyist producers that have neither the skills the time nor interest for investing in these sorts of issues.
# [12:11] <xover> ANd that's fine.
# [12:11] <xover> I hope those will eventually be covered by authoring tools that get them much of the way there.
# [12:11] <hsivonen> xover: I'm thinking in terms of what a consumer can realistically infer from Web content in general. Semantic proponents seem to consider communication inside a small club of semanticist, so the masses don't matter
# [12:11] <hsivonen> xover: on the Web, we also have a great mass of unskilled professionals
# [12:12] <xover> But even the best of tools cannot help them produce structured content when their way of conceiving of that content is not structured.
# [12:12] <hsivonen> xover: exactly
# [12:12] <xover> But the professional web worker can be influenced to do much better by a variety of means.
# [12:13] <xover> And these are the ones producing, say, governement web sites providing services to the public.
# [12:13] <hsivonen> xover: yeah, I'd be happy with gov't providing HTML, PDF and ODT instead of .doc
# [12:14] <xover> hsivonen: Amen!
# [12:14] * Joins: loic (loic@90.29.152.121)
# [12:14] <anne> lol, someone from Germany called me to buy links on my site for 1000 euro a month
# [12:14] <anne> this Google page rank thingie is apparently quite popular
# [12:14] <hsivonen> anne: did you sell out?
# [12:15] <anne> neh
# [12:15] <xover> And as we've seen with the transition to CSS for formatting for even commercial web sites, even the cold hard capitalists can be persuaded to improve in this sense.
# [12:15] <anne> I already have a job
# [12:15] <anne> with which I can pay bandwidth and all that
# [12:15] <anne> (and more :) )
# [12:15] <Lachy> I got emailed an offer to do a link exchange on my site too. I initally deleted it as spam, though it turned out to be a legit request cause the guy chatted to me on msn
# [12:16] <anne> oh, yeah, I get those more frequently too now
# [12:16] <anne> but link exchange seems kind of silly
# [12:16] <Lachy> perhaps I should just make my policy about advertising on my site more clear somehow
# [12:19] <xover> Ugh. “The and Elements”...
# [12:20] <Lachy> what?
# [12:20] <xover> Either your feed is borked, or my feed reader is.
# [12:20] <Lachy> it would be your feed reader
# [12:21] <xover> Hmm. Yeah. Looks nicely wrapped to me.
# [12:21] <xover> Unless there needs to be some double escaping or something else weird there.
# [12:21] <Lachy> there are, unfortunately, some that require double escaping of entity refs
# [12:21] <xover> right
# [12:21] <Lachy> those are broken
# [12:21] <Lachy> though it appears that /log/feed is sending RSS2.0 instead of Atom. I should fix that
# [12:22] <Lachy> I need to upgrade my Word Press installation too
# [12:22] <xover> It's an old version of NetNewWire. May be a bug fixed in a later version.
# [12:22] <Lachy> yeah, anyway, that blog entry is roughly what I wrote in my last mail to the list
# [12:28] <xover> «A cool off timer for this thread will prevent you from replying to any email in this thread for the next one hour. [Cancel]»
# [12:28] <xover> heh heh
# [12:32] * Quits: sbuluf (itqakvg@200.49.140.7) (Quit: sbuluf)
# [12:44] * Parts: anne (annevk@213.236.208.22)
# [12:45] * Joins: zcorpan_ (zcorpan@84.216.41.43)
# [12:48] * Joins: anne (annevk@213.236.208.22)
# [12:56] * Quits: inimino (inimino@75.71.88.233) (Quit: ChatZilla 0.9.78.1 [Firefox 2.0.0.3/2007031001])
# [13:01] * Quits: ROBOd (robod@86.34.246.154) (Quit: http://www.robodesign.ro )
# [13:04] <Lachy> I just noticed bikeshed.org gets a random background colour everytime it's loaded. Awesome!
# [13:08] * Quits: kazuhito (chatzilla@142.59.90.219) (Connection reset by peer)
# [13:11] <anne> Do people realize that role= implies RDF nonsense?
# [13:11] <anne> Or do they think it's just a new kind of class=?
# [13:11] <Bob_le_Pointu> Second option.
# [13:11] <hsivonen> anne: the latter
# [13:11] <hsivonen> anne: please explain how RDF is involved
# [13:12] <anne> The XHTML2 role= takes qnames besides predefined values which, when resolved, point to RDF resources which describe the meaning of the value
# [13:12] <anne> as far as I understand it
# [13:12] <Lachy> role introduces qnmaes for attr values, which is a bigger problem
# [13:12] <anne> There's has been some debate about renaming role= to rdf:type= too
# [13:13] <anne> See http://www.w3.org/2007/04/25-xhtml-minutes#item02 for instance
# [13:13] <Lachy> is there an article somewhere that discusses the problems with the role attribute that we could point people to?
# [13:14] <anne> I discussed it on my blog, but that's far from useful "pointing people to" material
# [13:16] <Lachy> http://annevankesteren.nl/2005/04/role-attribute
# [13:17] * anne wonders if that's the one he had in mind
# [13:18] <Lachy> there's also this one that briefly mentions it http://annevankesteren.nl/2006/06/accessibility-ideas
# [13:19] <anne> Yeah, that's probably it
# [13:19] <anne> It nicely contradicts the previous one
# [13:20] <Lachy> it's interesting to see how your opinion of XHTML2, role attr, etc. have gradully gone from supportive to unsupportive :-)
# [13:25] <anne> I appreciate it as well that I'm making progress rather than stagnating at some point in time.
# [13:35] <Philip`> Lachy: See e.g. http://lightgoldenrodyellow.bikeshed.org/ too
# [13:36] <anne> ah, it works exactly like that with comments on my blog
# [13:36] <anne> simple topic: noise
# [13:36] <anne> complex topic: signal
# [13:37] <Lachy> anne, instead of disabling comments completely on the simple topic, you should raise the barrier to entry so that only those with important comments will bother taking the effort to post
# [13:38] <Philip`> Require comments to be valid SVG, perhaps?
# [13:38] <Lachy> heh
# [13:38] <Lachy> yeah, valid XHTML is just too simple these days
# [13:40] <anne> it's some weird subset of XHTML actually
# [13:41] <anne> without namespace support etc.
# [13:41] <anne> but it's indeed to trivial
# [13:43] <anne> maybe someone an point http://lightgoldenrodyellow.bikeshed.org/ out to DanC as an argument of why this debating is wasting our time
# [13:43] <anne> s/an point/can point/
# [13:45] <Philip`> Actually, I think http://papayawhip.bikeshed.org/ is a much better site
# [13:45] <anne> oh lol
# [13:46] * anne gets the domain name now :)
# [13:46] <gsnedders> :)
# [13:50] <Lachy> it doesn't work properly with hex codes :-( http://09f911.bikeshed.org/
# [14:25] * Joins: olivier (ot@128.30.52.30)
# [14:59] * Joins: jdandrea (jdandrea@68.192.161.254)
# [15:09] * Quits: loic (loic@90.29.152.121) (Quit: hoopa rules)
# [15:13] * Joins: ROBOd (robod@86.34.246.154)
# [15:20] * Quits: tH (r@87.102.4.173) (Ping timeout)
# [15:42] * Lachy wishes the other Philip Taylor would *stop* CC'ing www-html!
# [15:42] * Quits: schepers (schepers@208.38.57.47) (Quit: Free at last!)
# [15:52] * Joins: tH (r@87.102.4.173)
# [16:04] * Quits: MrNaz (Naz@203.214.95.166) (Quit: Leaving)
# [16:09] * Quits: edas (edaspet@88.191.34.123) (Ping timeout)
# [16:12] * Joins: kazuhito (chatzilla@72.29.239.177)
# [16:40] * Joins: billmason (billmason@69.30.57.156)
# [16:52] * Joins: dbaron (dbaron@72.29.239.177)
# [16:53] * Joins: schepers (schepers@72.29.239.177)
# [17:10] * Joins: MikeSmith (MikeSmith@mcclure.w3.org)
# [17:17] * Quits: zdenko (zdenko@193.77.152.244) (Quit: zdenko)
# [17:34] * Joins: edas (edaspet@88.191.34.123)
# [17:37] * Joins: hasather (hasather@81.235.209.174)
# [17:50] * Joins: zdenko (zdenko@84.255.203.169)
# [18:08] <anne> @role is endorsed by the Mozilla Foundation?
# [18:08] * anne wonders if that claim is accurate
# [18:10] <Lachy> anne, there is some implementation of it in mozilla
# [18:10] <myakura> for its implementation, maybe
# [18:10] <anne> I'm aware of the implementation done by IBM
# [18:11] <myakura> yeah
# [18:12] <Lachy> I don't think the implementatation translates into support, though
# [18:17] <anne> the chair speaks
# [18:18] <mjs> so who here has been on W3C Working Groups besides Web API, WAF, SVG and CSS?
# [18:19] <mjs> (and of course HTML?)
# [18:19] <anne> I've been on CDF
# [18:19] * MikeSmith points to MikeSmith
# [18:19] <anne> I'm on CDF
# [18:19] <anne> i mean :)
# [18:20] <mjs> MikeSmith: which groups have you been on?
# [18:21] <MikeSmith> Web Security Context wg, now team contact for Mobile Web Best Practices and Device Description WGs
# [18:23] <MikeSmith> mjs - why do you ask? (just curious)
# [18:24] <mjs> MikeSmith, anne: on these groups, when a decision has to be made and there's a vote, whether formal or informal, is it normal for all no votes to automatically count as Formal Objections?
# [18:24] <mjs> I know that's not the case for SVG, Web API, WAF or CSS
# [18:24] * Quits: gavin (gavin@63.245.208.169) (Ping timeout)
# [18:24] <mjs> I've been discussing the matter with our chairs off-list
# [18:24] <mjs> http://lists.w3.org/Archives/Public/www-archive/2007May/0040.html
# [18:24] <mjs> so I wanted to do a broader check of what existing practice is
# [18:25] <anne> This is not the case for CDF
# [18:25] <anne> CDF is in fact the group I used as example yesterday, where real FOs are raised during LC if needed
# [18:25] * Joins: gavin (gavin@63.245.208.169)
# [18:25] <anne> And voting simply goes by majority
# [18:26] <MikeSmith> true that not normal for other WGs, but then the HTML WG isn't a normal WG so not sure it can be compared
# [18:26] <MikeSmith> most other WGs are maybe 15 or 20 people who all know each other
# [18:27] <MikeSmith> and straw polls are the norm
# [18:27] * zcorpan_ doesn't like polls
# [18:29] <mjs> MikeSmith: I think the much larger size of the HTML WG actually makes it *less* suited to a model where every no vote is a Formal Objection, since 100% consensus in a group that large is almost impossible
# [18:31] <MikeSmith> mjs - I suppose. but as anne notes, it is the same as for LC process -- where anybody (public) can raise an FO (not just WG members)
# [18:32] <mjs> MikeSmith: yes, exactly
# [18:43] * Quits: myakura (myakura@125.194.247.196) (Ping timeout)
# [18:46] * Joins: myakura (myakura@125.194.247.196)
# [18:53] * Quits: mjs (mjs@64.81.48.145) (Quit: mjs)
# [19:16] * Quits: myakura (myakura@125.194.247.196) (Quit: Leaving...)
# [19:21] * Joins: MrNaz (Naz@203.214.95.166)
# [19:31] * Parts: asbjornu (asbjorn@84.48.116.134)
# [19:39] <anne> heh, some of the people who haven't followed WHATWG make arguments about not knowing history
# [19:39] * anne isn't sure how this all makes sense anymore
# [19:45] <gsnedders> what I don't understand is how @role will be any better. we can't have any proper namespaces in HTML, and people will add invalid values to the attribute (thereby making under the arguments for not using @class impossible to add any further values in HTML6)
# [19:50] <anne> People seem to want different things from @role
# [19:50] <anne> 1. Semantics in attribute values to keep structure and semantics somewhat separate.
# [19:50] <anne> 2. Easier to extend.
# [19:50] <anne> 3. Tying in RDF like meaning to HTML.
# [19:51] <anne> 4. Solving accessibility problems with HTML because people use <div class=slider> as opposed to some native control like <input type=slider> which would handle that for them.
# [19:51] <anne> (My point 4 is a bit biased, but you get the idea.)
# [19:57] * Lachy wonders how RDF actually helps with anything
# [19:58] <Lachy> John Foliot is claiming that it can provide extensibility, but didn't explain how exactly
# [19:58] <anne> It provides extensibility in theory, but in practice that hasn't been accomplished yet afaict.
# [19:59] <anne> I'm of the opinion that new features should have clear practical benefits.
# [20:00] * Joins: mjs (mjs@17.255.104.133)
# [20:24] * Joins: hyatt (hyatt@24.6.91.161)
# [20:25] * Quits: hyatt (hyatt@24.6.91.161) (Client exited)
# [20:25] * Joins: hyatt (hyatt@24.6.91.161)
# [20:27] * Quits: dbaron (dbaron@72.29.239.177) (Quit: 8403864 bytes have been tenured, next gc will be global.)
# [20:30] * Quits: edas (edaspet@88.191.34.123) (Ping timeout)
# [20:46] <anne> oh, more www-archive
# [20:49] <jgraham> Seriously, I don't follow the argument that because something is new it won't be (ab)used in exactly the same way as current markup
# [20:50] * Quits: hasather (hasather@81.235.209.174) (Client exited)
# [20:50] <anne> maybe you need some non-logic
# [20:50] * Joins: hasather (hasather@81.235.209.174)
# [20:50] <zcorpan_> jgraham: because the new stuff will break in a drocanian way when you abuse it :)
# [20:51] <zcorpan_> and all implementations will be perfect
# [20:51] * Joins: ddailey (david_dail@24.144.172.117)
# [20:51] <anne> that would require some language interpreter in new implementations
# [20:51] <jgraham> And authors will suddenly start to understand RDF?
# [20:52] <anne> right after they bite through namespaces, sure
# [20:52] <ddailey> finally got IRC working in Opera
# [20:52] <anne> heh, sorry if the UI was not intiutive
# [20:52] <ddailey> never did chat before -- so who knows?
# [20:53] <ddailey> it looks okay to me
# [20:53] <ddailey> user error
# [20:53] * anne though the latest in UI was that the user can't do wrong
# [20:54] <ddailey> I like that
# [20:54] <ddailey> then it's not my fault
# [20:56] <ddailey> anne, I liked what you said about non-logic -- I think there is a study to be had there
# [20:56] <mjs> zcorpan_: there
# [20:56] <mjs> er
# [20:56] <mjs> nevermind, I don't even want to talk about "role"
# [20:57] <ddailey> It looks like one more thing to add to my reading list
# [20:58] <ddailey> I got the idea from Dan that a lot of semantics in the "semantic web" sense can be done through GRDDL rather painlessly without adding markup to HTML
# [20:58] * Quits: hyatt (hyatt@24.6.91.161) (Quit: hyatt)
# [20:59] * anne has to go
# [21:01] <ddailey> maciej, I wonder if we might be able to categorize the prescriptivist arguments.
# [21:01] <ddailey> in some sort of descriptivist way
# [21:02] <zcorpan_> mjs: sorry, forgot to use the <sarcasm> tag :)
# [21:02] <gsnedders> Opera_9.20_Setup.dmg refuses to mount on Mac OS 10.4.9
# [21:03] <gsnedders> zcorpan_: I thought the conclusion was that both start and end tags though required may be omitted? :)
# [21:03] <zcorpan_> gsnedders: required and may be omitted at the same time?
# [21:03] <mjs> ddailey: not sure what you mean
# [21:04] <gsnedders> zcorpan_: like <html>, <head>, and <body>
# [21:04] <ddailey> wondered how many different points of view they actually have
# [21:04] <gsnedders> n^2 (where n is the number of members of the WG)
# [21:04] <zcorpan_> gsnedders: ah, ok. so you meant that the element is required but the tags may be omitted :)
# [21:04] <gsnedders> zcorpan_: yes :)
# [21:04] <ddailey> I think some are saying -- look how scary the web will be if we don't clamp down on grammar now
# [21:05] <ddailey> some seem to have something else going on
# [21:05] <mjs> well, I don't want to be in the business of classifying arguments I don't really believe in
# [21:05] <ddailey> if it's only n^2 it can be managed with a fairly simple grammar
# [21:06] <gsnedders> mjs: anytime I try and say what I think they're saying I just get told that I'm trying to put words in their mouth…
# [21:07] <ddailey> one might be able to hash out a compromise or a synthesis or something ifall those other points of view can be enumerated
# [21:09] <ddailey> I am softening on design principles -- one really DOESN't want to have this particular debate again and again forever
# [21:10] <ddailey> maybe if we had a token economy of some sort and we handed out finitely many words that any side of any debate could use
# [21:17] <ddailey> ddailey should learn how to use 8-) and so forth (except I don't know what they mean) -- there are lots of humor tags inserted in some of the above
# [21:17] <Philip`> Ooh, my message arrived on www-html, only four days late
# [21:18] * Parts: ddailey (david_dail@24.144.172.117)
# [21:18] <Philip`> *messages
# [21:31] * Joins: ddailey (david_dail@24.144.172.117)
# [21:31] <Philip`> Aha, now I can run XPath queries on ~2000 HTML documents in ~50 seconds, and get totally statistically insignificant results from a heavily biased sample, which might be handy
# [21:32] * Quits: tH (r@87.102.4.173) (Ping timeout)
# [21:34] * Joins: tH (r@87.102.4.173)
# [21:34] * Joins: inimino (inimino@75.71.88.233)
# [21:40] * Quits: olivier (ot@128.30.52.30) (Quit: Leaving)
# [21:43] <ddailey> html-wg irc looks a lot more active when one just reads the logs -- nearly festive at times
# [21:45] <Philip`> It's rather bursty, and seems to be going through a quiet patch now
# [21:47] <ddailey> I tried dinking around with XPATH a little but was unable to write any code that worked across browsers.. Any recos on a good tutorial?
# [21:49] <Philip`> Unfortunately I have no idea how to use it in browsers and not much idea how to use it out of browsers either - I'm just doing stuff in Python and copying-and-pasting examples
# [21:50] * Joins: dbaron (dbaron@72.29.239.177)
# [21:55] <ddailey> The XSLT code that Dan wrote to summarize the tasks volunteer questionnaire was pretty nifty looking. I was going to try to learn XSLT this semester until this WG came along. Maybe this summer. Shouldn't be much activity this summer should there be? Don't all you folks take big summer vacations? ;)
# [22:15] * Quits: dbaron (dbaron@72.29.239.177) (Ping timeout)
# [22:19] * Joins: dbaron (dbaron@72.29.239.177)
# [22:39] * Quits: beowulf (carisenda@91.84.50.132) (Ping timeout)
# [22:42] * Joins: beowulf (carisenda@91.84.50.132)
# [22:47] <hsivonen> do the accessibility folks really believe that RDF will give AT more understanding?
# [22:48] * Quits: dbaron (dbaron@72.29.239.177) (Ping timeout)
# [22:49] <ddailey> I would guess that they are not all of one point of view
# [22:49] <Lachy> hsivonen, that's what I'm wondering
# [22:50] * Joins: dbaron (dbaron@72.29.239.177)
# [22:51] * Quits: ROBOd (robod@86.34.246.154) (Quit: http://www.robodesign.ro )
# [22:51] <mjs> hsivonen: people mistakenly think that the ability to write declarations for a custom vocabulary and put it in a namespace gives understanding without prior agreement
# [22:52] <mjs> (well, mistakenly IMO anyway)
# [22:52] <ddailey> mjs: by without prior agreement you mean
# [22:53] <ddailey> oops hit the return button by accident
# [22:53] <mjs> I mean that if you declare your own custom vocabulary in some formal way, that in no way aids anyone in understanding that vocabulary if they are not specifically aware of it
# [22:54] <ddailey> Okay, yes that's what I thought
# [22:54] <Lachy> it's like defining your own FooML and then expecting UAs to instantly understand it
# [22:55] * gsnedders got told a few days ago that browsers already support XHTML2 :P
# [22:55] <Lachy> gsnedders, of course the do! There's a working example on the web ;-)
# [22:55] * mjs facepalms
# [22:55] <mjs> anyway, Tina joined the XHTML2 WG, so that's one positive step from the flamage
# [22:56] <Lachy> yeah, she joined last week
# [22:56] <Lachy> gsnedders, http://w3future.com/html/examples.xml?notransform
# [22:57] <Lachy> (they don't like telling anyone this, but that relies on proprietary extensions in major browsers to actually be functional)
# [22:57] <gsnedders> Lachy: what extensions?
# [22:57] <Lachy> view the CSS
# [22:57] <mjs> does it work in Safari?
# [22:58] <gsnedders> mjs: does for me
# [22:59] <mjs> is that page itself allegedly XHTML2?
# [22:59] <Lachy> ah, it's been changed since I last looked at it
# [22:59] <gsnedders> Lachy: where? I see nothing proprietary skimming through it
# [22:59] <gsnedders> mjs: with XSL, though
# [22:59] <Lachy> they now use XSLT to transform it to HTML
# [22:59] <mjs> it's XSLT transformed
# [23:00] <Lachy> it used to contain properties like '-o-link' (I think that's it) for making the links work, and binding XBL1 for Mozilla
# [23:01] <mjs> I wonder if their transform would actually handle xml-events and xforms (they declare the namespaces but don't use them)
# [23:02] * mjs guesses no, since scripting would have to work on the post-transform DOM)
# [23:02] <Lachy> yeah, that's weird
# [23:02] <Lachy> XMLEvents is just another poorly designed technology though
# [23:02] <Lachy> XBL makes it obsolete
# [23:03] <ddailey> Does that that mean I don't have to learn it? goodie
# [23:03] <mjs> xml-events makes easy things hard and hard things hard
# [23:06] <Lachy> ddailey, you never have to learn anything that comes out of the XHTML2 and related groups
# [23:06] <Lachy> did the XHTML2WG make xml events?
# [23:06] <gsnedders> g'nite
# [23:07] <Lachy> yep, they did
# [23:07] <ddailey> The less the better
# [23:09] <Lachy> wow, I love how they replace <img ... onclick=",,,"> with something even easier: ...
# [23:09] <Lachy> <img src="button.gif" alt="OK">
# [23:09] <Lachy> <script ev:event="activate" type="application/x-javascript">
# [23:09] <Lachy> doactivate(event);
# [23:09] <Lachy> </script>
# [23:09] <Lachy> </img>
# [23:13] <jgraham> ddailey: X-Path in browsers (well Firefox at least) http://www-xray.ast.cam.ac.uk/~jgraham/mozilla/xpath-tutorial.html
# [23:13] * jgraham was sure that got moved to the Mozilla wiki somewhere
# [23:20] * Joins: Sander (svl@71.57.109.108)
# [23:20] <ddailey> thanks
# [23:22] <ddailey> I'll give it a read. I got some stuff working in FF and Opera, and got some other stuff working in IE. It all seemed completely different, but then I didn't have the faintest idea what I was doing.
# [23:23] <jgraham> I have no idea what IE does. If it doesn't follow the spec, that document won't help
# [23:26] <ddailey> as I recall the expressions resolved similarly, but it was figuring out how to invoke them that was quirky
# [23:39] * Joins: AGraf_ (Ashe@213.47.199.86)
# [23:41] * Quits: AGraf (Ashe@213.47.199.86) (Ping timeout)
# [23:45] * Joins: asbjornu (asbjorn@84.48.116.134)
# [23:45] <ddailey> a problem I've had in trying to understand the discussions is this: where to draw a boundary on what is HTML -- where are its boundaries?
# [23:46] <ddailey> we've got script with document.write, we've got some DOM, CSS is sorta there in a presentational sense, sorta in a semantic sense
# [23:47] <ddailey> we've got events, we've got XML and object, and all the foreign content
# [23:49] <ddailey> for reading and writing files we've got webforms (and various x-things) and I can't remember what the RFC was that covered file output
# [23:50] <ddailey> when I find issues concerning say a weird document.write script that seems to change styles in one browser but not in another I have no way
# [23:51] <ddailey> to figure out even which spec to begin looking in to know if it's a bug or my misunderstanding of how it's supposed to be -- the specs have gotten a bit spread out
# Session Close: Tue May 08 00:00:00 2007

The end :)