| 00:32 | <BenMillard> | krijn, I've refined and added to my IRC log ideas: http://projectcerbera.com/!dev/irc-logs/ |
| 00:33 | <BenMillard> | krijn, I've only tested those demos in Firefox 2, but I can do cross-browser testing if you like |
| 00:35 | <BenMillard> | krijn, the "day" demo is not finished...I plan to make status messages more consistent with normal message and identify them using class to avoid lots of string checking in the JS |
| 00:58 | <billyjackass> | hsivonen: thanks for pointer regarding date/time microsyntaxes |
| 02:49 | <Hixie> | hsivonen: if you have questions re mutation events that you can't determine an answer to, let me know and i'll give you an answer for now -- but in general i'm blocked on mutation events and svg getting their specs fixed to be more clearly defined |
| 02:53 | <ajnewbold> | this html5, it's lucious |
| 03:03 | <Hixie> | i am not familiar with that word |
| 03:03 | <Hixie> | unless you're referring to the stargate character |
| 03:06 | <ajnewbold> | hmm, well, it helps if you spell it correctly |
| 03:06 | <billyjackass> | ludicrous? |
| 03:06 | <ajnewbold> | http://www.merriam-webster.com/dictionary/luscious |
| 03:07 | <Hixie> | aah, luscious |
| 03:07 | <Hixie> | glad you think so :-) |
| 03:08 | <ajnewbold> | :) |
| 03:50 | <ajnewbold> | erm, I don't understand this error. http://validator.w3.org/check?uri=http%3A%2F%2Fneatnik.net%2F&charset=%28detect+automatically%29&doctype=Inline&ss=1&group=0&user-agent=W3C_Validator%2F1.606#line-6 |
| 03:50 | <ajnewbold> | I'm not providing any attributes on <body>, so either I'm overlooking something simple or the validator's having a bad day |
| 05:47 | MikeSmith | finds himself wishing he had the equivalent of a JS minifier for dealing with messages from certain recent posters to the whatg list |
| 08:17 | Hixie | sets up his network so that the neighbour's machine that keeps trying to do bittorrents has all HTTP traffic redirected to a simple page that tells the guy that his torrenting is bottlenecking the connection |
| 08:18 | <Hixie> | (i am also running a program that kills any non-port-80 tcp/ip traffic from that host) |
| 08:33 | <jwalden> | why not just set up wpa? |
| 08:39 | <Hixie> | i want an open network |
| 08:39 | <Hixie> | i use open networks all over the place |
| 08:39 | <Hixie> | i feel i should give back |
| 08:39 | <Hixie> | wpa also makes it a pain to connect the devices to the network (like the wii, the ipod, the ps3) and makes it harder for guests to connect |
| 08:40 | <jwalden> | suit yourself :-) |
| 09:24 | <zcorpan> | ajnewbold: http://html5.validator.nu/?doc=http%3A%2F%2Fneatnik.net%2F gives no errors so it has to be something they screwed up in the integration |
| 11:50 | <hsivonen> | http://hsivonen.iki.fi/html5-gecko-build/ |
| 11:57 | <Lachy> | hsivonen, cool. I'll try it out later |
| 12:12 | <jgraham> | hsivonen: Cool |
| 12:12 | jgraham | wishes he had a convenient way to try it out |
| 13:11 | <rubys> | hsivonen: ping? |
| 13:18 | <annevk3> | congrats hsivonen! |
| 13:54 | <tthorsen> | looks like livedom.validator.nu does not like the following markup: <!DOCTYPE html><html><head></br> |
| 15:13 | <zcorpan> | http://parsetree.validator.nu/?parser=html5&content=%3C%21DOCTYPE+html%3E%3Chtml%3E%3Chead%3E%3C%2Fbr%3E&submit=Print+Tree |
| 15:15 | <zcorpan> | hsivonen: ^ doesn't give expected result |
| 15:40 | <rubys> | annevk3: http://intertwingly.net/blog/2008/12/03/HTML5-in-Gecko?#c1228319123 |
| 15:42 | <annevk3> | rubys, since we do not implement SVG or MathML inside text/html there is no such thing as a "foreign element" for Opera |
| 15:43 | <annevk3> | rubys, just like in <div><foo><bar/></foo></div> <bar/> is not actually self-closed per HTML5 |
| 15:49 | <annevk3> | rubys, also, yeah, WebKit does it incorrectly afaict |
| 15:50 | <zcorpan> | annevk3: </foo> should close the bar, though, shouldn't it? |
| 15:51 | <annevk3> | yeah |
| 15:51 | <annevk3> | not the best example |
| 15:51 | <zcorpan> | annevk3: how is it not a good example? sam has <svg><circle></svg> |
| 15:52 | <zcorpan> | we put everything after </svg> in the circle element |
| 15:52 | <annevk3> | oh really? |
| 15:52 | <zcorpan> | afaict |
| 15:52 | <annevk3> | mea culpa |
| 15:58 | <rubys> | it is a pity that HTML5 doesn't treat non-recognized HTML elements as foreign. |
| 16:00 | <rubys> | actually, isn't the spec ambiguous? <p> is an allowed HTML element. <math> is an element from the MathML namespace. What does this make <svg>? |
| 16:00 | <annevk3> | it's a pity we're stuck with HTML :) |
| 16:00 | <annevk3> | <svg> is currently not allowed |
| 16:01 | <rubys> | ok, so it is not normal, and not foreign, and not RCDATA, and not CDATA, and not Void. Is it a sixth kind of element? |
| 16:02 | <zcorpan> | rubys: i guess the secion you're reading only deals with elements that are allowed in html5 |
| 16:02 | <annevk3> | the syntax and language are intertwined |
| 16:02 | <zcorpan> | section |
| 16:03 | <rubys> | so, what does the spec say about how to handle <foo/>? |
| 16:03 | <zcorpan> | rubys: in the parsing section it would go under "Any other start-tag token" or what it says |
| 16:49 | <zcorpan> | aha... PFWG's approach to get past CR is to remove 'must's from the spec |
| 16:49 | <zcorpan> | that's interesting |
| 16:49 | <zcorpan> | (actually a whole conformance class) |
| 16:54 | <jgraham> | I didn't think they had any MUSTs to begin with... |
| 16:55 | <zcorpan> | they had some |
| 16:58 | <jgraham> | Oh well |
| 17:00 | <jgraham> | I guess eventually they will have none at all any any behaviour will be considered both conforming and therefore accesible |
| 19:31 | <gsnedders> | ECMA has no semantic index of standards |
| 19:35 | <gsnedders> | Nor does the ISO |
| 19:36 | <jcranmer> | IETF? |
| 19:36 | <jcranmer> | ANSI? |
| 19:38 | <gsnedders> | jcranmer: IETF does of RFCs, dunno about other docs |
| 19:38 | <gsnedders> | s/docs/series/ |
| 19:38 | <gsnedders> | And ANSI I don't really care about :) |
| 19:38 | <jcranmer> | IANA? |
| 19:38 | <jcranmer> | ICANN? |
| 19:39 | <gsnedders> | how often do people want to reference them? :) |
| 19:39 | <jcranmer> | IANA protocol registries are actually fairly useful |
| 19:39 | <jcranmer> | for example, MIME content types |
| 19:39 | <jcranmer> | official extensions for protocols with extensioning mechanisms |
| 19:39 | <jcranmer> | default TCP ports |
| 19:40 | <gsnedders> | To my knowledge not really |
| 19:41 | <gsnedders> | (have an index) |
| 19:41 | <jcranmer> | IANA has a master list, alphabetacized by protocol name that I usually just Ctrl-F |
| 19:41 | <gsnedders> | jcranmer: That isn't very semantic though |
| 19:41 | <gsnedders> | jcranmer: And can't be easily scrapped |
| 19:42 | <gsnedders> | (well, it can be easily scrapped, but not in a generic way) |
| 19:42 | <jcranmer> | I know it has a master list, I was wondering if it had a semantic one |
| 19:42 | <gsnedders> | It doesn't |
| 19:42 | <jcranmer> | :-( |
| 19:42 | <gsnedders> | RFCs and W3C TRs are all that do of major tech. publications as far as I can see |
| 19:43 | <jcranmer> | you know what sucks... just how much one has to search to get around documentation |
| 19:43 | <jcranmer> | trying to navigate to the DOM or CSS specs from www.w3.org is a bit frustrating |
| 19:43 | <gsnedders> | Just guess that it is w3.org/TR/CSS/ :P |
| 19:44 | <jcranmer> | I normally stick with CSS21 |
| 19:44 | <gsnedders> | I just guess shortname |
| 19:44 | <gsnedders> | *the shortname |
| 19:44 | <jcranmer> | some of the DOM specs are annoying |
| 19:44 | <gsnedders> | Yeah, they are. |
| 19:44 | <jcranmer> | even worse is trying to get to the Windows SDK reference from msdn.microsoft.com |
| 19:45 | <jcranmer> | you pretty much have to either search or be psychic |
| 21:45 | <gsnedders> | Philip`: Where should we meet up on Monday? |
| 21:59 | <yecril71> | A valid URL with a valid fragment identifier constitutes a broken link only if the resource referenced cannot be found. |
| 22:00 | <yecril71> | Referring to a fragment identifier that is not defined in the resource does not make it broken. |
| 22:01 | <yecril71> | href="#foo bar" does not match id="foo%20bar" but you can have it vice versa. |
| 22:02 | <yecril71> | That is because an identifier is not an URL and it is not subject to decoding. |
| 22:04 | <yecril71> | An identifier containing spaces makes the document invalid, and the document |
| 22:04 | <yecril71> | validator should behave accordingly. |
| 22:05 | <yecril71> | A disconnected subtree need not be a whole document. |
| 22:06 | <yecril71> | I do not believe DOM Core should ever reinvent what a subtree is. |
| 22:09 | <yecril71> | Since the identifier is guaranteed to be the same in the document and in a disconnected subtree, there is really no room for any ambiguity. |
| 22:11 | <yecril71> | The DOM mechanisms and the CSS mechanisms, if the latter are ever defined for scripting, have different purposes. |
| 22:11 | <yecril71> | I do not think a full consistency is required here; otherwise one of them seems to be redundant. |
| 22:32 | <Philip`> | gsnedders: I don't have any especially good ideas on novel locations |
| 22:47 | <gsnedders> | Philip`: Should we say the same as before, then? |
| 22:54 | <Philip`> | gsnedders: I suppose that would be reasonable |
| 22:55 | <Philip`> | gsnedders: I hope you'll be on IRC on Monday to remind me :-) |