| 03:14 | <Hixie> | http://lists.w3.org/Archives/Public/public-html/2007Nov/0123.html |
| 03:14 | <Hixie> | http://lists.w3.org/Archives/Public/www-archive/2007Nov/0039.html |
| 03:53 | <fantasai> | Lachy: Can I copy parts of http://blog.whatwg.org/submit-article ? |
| 05:43 | <Lachy> | fantasai, of course you can |
| 05:45 | <othermaciej> | hey Lachy |
| 05:49 | <Lachy> | hi |
| 06:02 | <raspberry-lemon> | hello |
| 06:02 | <raspberry-lemon> | can html5 be used today? |
| 06:03 | <jwalden> | lots of it is, in existing browsers :-) |
| 06:03 | <jwalden> | be more specific about what you want to use |
| 06:03 | <raspberry-lemon> | well, it's for a web framework called merb |
| 06:04 | <raspberry-lemon> | right now we're discussing xhtml/html issues |
| 06:04 | <raspberry-lemon> | and trying to figure out what we should default to serving |
| 06:05 | <jwalden> | HTML is generally a safer choice, because then you don't have to do content negotiation or deal with two different DOMs |
| 06:05 | <raspberry-lemon> | yeah |
| 06:06 | <othermaciej> | assuming you mean serving xhtml as application/xhtml+xml |
| 06:06 | <othermaciej> | browser xhtml implementations also tend to be less thoroughly tested and optimized |
| 06:06 | <othermaciej> | since the web is pretty much html |
| 06:06 | <raspberry-lemon> | people are going to use it to serve xhtml as text/html, no matter what we do |
| 06:07 | <raspberry-lemon> | so this is more about reasonable defaults |
| 06:08 | <othermaciej> | serving xhtml as text/html is not so great either |
| 06:08 | <raspberry-lemon> | oh, i know |
| 06:08 | <othermaciej> | but will probably cause fewer practical problems than serving different MIME types to different browsers |
| 06:09 | <raspberry-lemon> | *nods* |
| 06:09 | <raspberry-lemon> | we also have a bunch of helpers for creating code snippets |
| 06:09 | <raspberry-lemon> | image tags, etc. |
| 06:10 | <raspberry-lemon> | at the moment, they're producing xhtml syntax (trailing slashes for empty elements) |
| 06:10 | <raspberry-lemon> | which was what brought me here, really |
| 06:11 | <webben> | raspberry-lemon: Can your framework give developers a choice? |
| 06:11 | <webben> | (perhaps even a choice of content negotiation) |
| 06:11 | <raspberry-lemon> | yes |
| 06:11 | <webben> | raspberry-lemon: So this is literally about what to use as a default serialization? |
| 06:11 | <raspberry-lemon> | yes |
| 06:11 | <raspberry-lemon> | we'd like to use html |
| 06:12 | <raspberry-lemon> | but we know that if we go for html4.01, we'll have lots of confused users |
| 06:12 | <webben> | confused by what? |
| 06:13 | <othermaciej> | HTML5 allows the trailing slashes |
| 06:14 | <webben> | If users are going to be confused by HTML 4.01 syntax, then they seem likely to be confused by a whole new version of HTML. |
| 06:14 | <webben> | Seems the only risk here is that users get confused and produce invalid documents. |
| 06:15 | <webben> | By using an XHTMLism. |
| 06:15 | <webben> | The documents will still "work" in browsers. |
| 06:15 | <raspberry-lemon> | right |
| 06:17 | <webben> | The risk of using HTML5 is greater, since the draft continues to change substantially. |
| 06:17 | <webben> | The potential benefits mainly revolve around features, but those come and go. |
| 06:18 | <webben> | I'd look at whether the more stable bits of HTML5 allows your developers to offer any benefits to /their/ end-users. |
| 18:53 | <Philip`> | http://philip.html5.org/demos/canvas/opera-3d/textured-cube.html |
| 18:54 | <Philip`> | Can't see any way to do lighting, so I can't even make a nice-looking cube :-( |
| 19:18 | <Hixie> | hmm |
| 19:18 | <virtuelv> | Hixie: hm what? |
| 19:18 | <Hixie> | so the next issue i have to reply to is for an element in html5 for trees |
| 19:19 | <Hixie> | as in directory trees, family trees, that kind of thing |
| 19:19 | <othermaciej> | like graph-style trees? |
| 19:19 | <othermaciej> | or tree view type trees? |
| 19:19 | <MikeSmith> | Hixie - I remember aaron leventhal mentioning that as a deficiency in HTML ... |
| 19:20 | <othermaciej> | I think datagrid provides the right semantics for n-ary trees |
| 19:20 | <othermaciej> | but unclear if it is desirable presentation for all cases |
| 19:33 | <Hixie> | well datagrid is more of a control, i'm looking for a way to mark up the data |
| 19:33 | <Hixie> | e.g. a family tree |
| 20:12 | Hixie | wonders how to do it |
| 20:12 | <othermaciej> | how to do what? |
| 20:14 | <Hixie> | the family tree thing |
| 20:15 | <othermaciej> | RDF? |
| 20:15 | othermaciej | ducks |
| 20:15 | <Hixie> | hmm, good idea |
| 20:15 | Hixie | embeds RDF into html5 |
| 20:15 | <othermaciej> | family trees are not actually trees |
| 20:15 | <Hixie> | yeah |
| 20:15 | <othermaciej> | in the general case |
| 20:15 | <Hixie> | they're graphs |
| 20:15 | <Hixie> | complicated ones in many cases |
| 20:15 | <othermaciej> | they are directed graphs with labelled edges |
| 20:16 | <Hixie> | yeah |
| 20:20 | <annevk> | hmm, RDF |
| 20:20 | <annevk> | nice |
| 21:18 | <piratepenguin> | does either of firefox's video patch or Opera's video build support <audio>? |
| 21:21 | <othermaciej> | WebKit nightlies support it I believe |
| 21:23 | <piratepenguin> | hmm cool but how would I test that on GNU/Linux, what's the simplest way? |