| 03:33 | <karlcow> | by curiosity I tried to render the same origin algorithm http://www.la-grange.net/2010/01/22/same-origin-algorithm.png |
| 03:33 | <karlcow> | might be wrong |
| 04:18 | <TabAtkins> | karlcow: Yeah, it's wrong. The opaque-identifier portion should completely separate from the same-components section. If they're opaque, they don't *have* components. |
| 04:19 | <karlcow> | TabAtkins: ah. I'll modify. It is not clear in the prose of the spec I have the feeling. |
| 04:19 | karlcow | likes to visualize to understand |
| 04:19 | <TabAtkins> | If you translate it from prose directly to if/then code in your favorite programming language, it becomes very clear. ^_^ |
| 04:19 | <karlcow> | (python) |
| 04:21 | <karlcow> | but my favorite languages are 1. fr:poetry 2. graphic_representation |
| 04:46 | <karlcow> | TabAtkins: thanks |
| 09:42 | <Hixie> | huh, brief power black out in mountain view |
| 09:42 | <Hixie> | i wonder what that was about |
| 09:45 | <erlehmann> | my bet: the PROJECT has escaped ;) |
| 09:45 | <erlehmann> | cue x-files melody |
| 09:47 | <Hixie> | nobody twittered about it that i can see, so... maybe just the complex i'm in |
| 09:58 | <Hixie> | i should get a UPS |
| 09:58 | <Hixie> | that power outage ruined by 250+ day uptime |
| 09:58 | <Hixie> | (on the box linux box routing all my network traffic, which i installed about 250 days ago) |
| 09:59 | <Philip`> | I wonder if development of the 'uptime' program was secretly supported by the people who make UPS devices |
| 10:00 | <Hixie> | hah |
| 10:01 | <Hixie> | are UPSes any good these days? 15 years ago they could barely do 7 minutes for your typical load |
| 10:02 | <Hixie> | wow, apparently not |
| 10:02 | <Hixie> | 3 minutes?! |
| 10:02 | <Hixie> | 5 minutes?! |
| 10:03 | <Hixie> | $378 for less than 7 minutes at barely 1kW? |
| 10:06 | <Hixie> | my _iPod_ can last a whole day, and it fits in my pocket |
| 10:07 | <erlehmann> | my g1 lasts a whole hour, and it fits in my - OH WAIT :D |
| 10:07 | <Hixie> | surely something that can be put under a desk, that doesn't need to handle continuous drain-charge cycles, and that is plugged into the mains most of the day, can do better |
| 10:08 | <Hixie> | i wonder how much petrol would be needed for a generator to support a 1kW load for an hour |
| 10:09 | <Hixie> | surely we have rechargeable materials with more efficient power densities than petrol by now |
| 10:09 | Hixie | grumbles at hardware people :-P |
| 10:09 | <erlehmann> | as if |
| 10:10 | <Hixie> | i was buying a printer recently and as far as i could tell, they essentially haven't improved in 10 years |
| 10:11 | <Hixie> | and now i find UPSes haven't improved in like 15 years |
| 10:11 | <Hixie> | i know chips improve, and laptops improve, and software sure has improved, and network connectivity has improved |
| 10:11 | <Hixie> | so wtf are the printer and UPS people doing?! |
| 10:14 | <erlehmann> | working on their uranium-powered printers in their secret lair ? ;) |
| 10:14 | <Hixie> | i hope so |
| 10:14 | <Hixie> | when they finally release something new, it had better be effing AWESOME. |
| 10:15 | <erlehmann> | like bre-pettis-AWESOME :D |
| 10:15 | <erlehmann> | thats it. printer people worked on adding more dimensions to their plotters ! |
| 10:16 | <Hixie> | i expect hte next gen printers to be able to print onto a whole ream of paper at once without moving any of the sheets, so that you just stick a ream in, and then you pick it up three seconds later and the whole ream is fully printed |
| 10:16 | <Hixie> | DOUBLE-SIDED. |
| 10:17 | <Hixie> | and the next gen UPS had better just be a cable you can replace your current cable with |
| 10:17 | <erlehmann> | nono, they print the paper as well |
| 10:17 | <Hixie> | which lasts 72 hours on one charge, charges in 5 seconds, and can support 10MW loads. |
| 10:18 | <erlehmann> | looks like you're going for an Uranium Power Supply there … |
| 10:31 | <Lachy> | wow, the comments from some of the CSSWG memebers regarding selectors api on public-webapps/www-style seem surprisingly confused about how it works. |
| 10:32 | <Lachy> | also about the use cases that are being addressed, but that's slightly more understandable if they haven't actively followed its development |
| 10:34 | <Lachy> | I suppose I should clearly document all of those somewhere, outside of the mailing list discussion, for easy reference. |
| 10:34 | <Lachy> | I'll put it on the whatwg wiki |
| 10:40 | <Lachy> | Hixie, build your own UPS out of an array of car batteries. They should last quite a while, if you have enough. |
| 10:41 | <Philip`> | You could just use a laptop |
| 10:44 | <Lachy> | http://www.dansdata.com/upsupgrade.htm or http://www.dansdata.com/diyups.htm |
| 10:47 | <Hixie> | Philip`: the laptops didn't die, but it turns out they're useless without the wireless router and the cable modem |
| 10:48 | <Hixie> | Philip`: and the mac mini stops being entertaining when it, its media disk, its speakers, and its screen all go off at once |
| 10:52 | <Lachy> | Hixie, is a mac mini particularly entertaining for some reason when it's on? |
| 10:53 | <Lachy> | perhaps if you use it as a media centre |
| 10:53 | <Hixie> | it's my tv replacement |
| 10:53 | <Lachy> | ok |
| 10:53 | <Hixie> | and music player |
| 10:54 | <Hixie> | and all around entertainment system |
| 10:58 | <Philip`> | You could use laptops with one laptop acting as a wireless router, connected to the internet via the mobile phone network |
| 10:58 | <Philip`> | Also, you could entertain yourself by reading a book |
| 10:58 | <Philip`> | So there's really no reason to worry about power cuts at all |
| 10:59 | <Philip`> | (Well, for the book you might need some candles too) |
| 11:00 | <Huvet> | where can I find the documentation for html5lib? |
| 11:02 | <Philip`> | Huvet: import html5lib; help(html5lib) |
| 11:02 | <Philip`> | (from Python interpreter) |
| 11:02 | <Philip`> | seems to be about the only documentation, given that the wiki seems to have lost its pages |
| 11:03 | <Huvet> | ok, thanks |
| 11:05 | <Hixie> | Philip`: what are these "candles" you speak of |
| 11:05 | <Hixie> | isn't that a unit of brightness |
| 11:08 | <Lachy> | candles are a fire hazard, especially because you need a few hundred of them to replicate the amount of light obtained from a decent 100W incandescent bulb (or equivalent). |
| 11:14 | <Philip`> | It's possible to see quite adequately without needing the equivalent of a 100W incandescent bulb |
| 11:15 | Philip` | isn't suggesting using them as an identical replacement for current electrical lighting systems |
| 11:22 | <jgraham> | Huvet: There is some documentation somewhere but it got lost at some point |
| 11:24 | <jgraham> | http://html5lib.googlecode.com/svn/wiki/UserDocumentation.wiki |
| 11:24 | <jgraham> | (it may not be entirely up to date) |
| 11:25 | <Huvet> | ah, thanks |
| 11:31 | <Philip`> | jgraham: Do you know why that doesn't show up on the web page? |
| 11:32 | <Philip`> | (http://code.google.com/p/html5lib/w/list links to UserDocumentation but says page not found) |
| 11:32 | <Philip`> | Maybe it's a SVN vs Hg thing? |
| 11:39 | <Lachy> | Philip`, trying to read in rooms that have less light than that from a 100W bulb, or better yet, long fluroescent tube lighting, can get quite tiring for my eyes. |
| 11:41 | <Dashiva> | Well, you could use the light to illuminate the book and not the entire room |
| 11:45 | Philip` | has a 20W desk lamp that's easily more than adequate |
| 11:51 | <Lachy> | I know what everyone finds comfortable is different, but I can't work with desk lamps. They give an uncomfortable bright spot while the rest of the room is comparitively dark. |
| 11:59 | <jgraham> | Philip`: I guess it must be but I think it worked for a while after we switched (but I could be wrong about that) |
| 11:59 | <jgraham> | (however I blame Google anyway) |
| 11:59 | <Philip`> | Lachy: You need smaller rooms |
| 16:06 | <Dashiva> | Is JF saying that if we can't be sure about author intent, we should abandon all attempts at doing anything useful? |
| 16:35 | <TabAtkins> | Dashiva: It's strange, but that's the vibe I get too. |
| 16:37 | <TabAtkins> | I'm guessing it's a particular manifestation of "perfect is the enemy of good". |
| 16:42 | <webben> | Dashiva: I don't think so. |
| 16:43 | <Dashiva> | I can't imagine he would, but I'm having trouble reading it any other way |
| 16:43 | <webben> | I think he's just saying the spec should not imply feasibility of "non-existent techniques". |
| 16:46 | <webben> | and taking the phrase "image analysis heuristics" to imply technology that can determine author intent. |
| 16:47 | <webben> | I think that's a stretch, but meh. |
| 16:47 | <Dashiva> | Yes, that |
| 16:47 | <Dashiva> | is reasonable, until someone tells you that's not what it means :) |
| 16:52 | <webben> | Dashiva: Yeah, but even from that point, you can still make assertions that the spec's language implies something different that what is meant (or is ambiguous). |
| 16:53 | <webben> | Dashiva: hence his citing of a dictionary definition of heuristics, I guess. |
| 16:55 | <TabAtkins> | Protip: As soon as you're citing dictionary definitions, you've lost the argument. |
| 16:55 | <webben> | Not really sure that's true. |
| 16:56 | <webben> | Not when you're discussing what does a blob of text imply, anyhow. |
| 16:56 | <webben> | That is, there's a difference between an argument about essences and an argument about phrasing. |
| 16:56 | <TabAtkins> | I find that even in that sort of dicussion, whipping out a dictionary means the argument is officially over and you're on the losing side. |
| 16:56 | <TabAtkins> | It's like godwining. |
| 16:57 | <webben> | A /better/ argument, I find is to point to examples of other readers reading it that way. |
| 16:57 | <webben> | But - that's more expensive. |
| 19:05 | <foolip> | microdata toy for all the children: http://foolip.org/microdatajs/demo/live.html |
| 19:06 | <paul_irish> | fun! |
| 19:11 | <Dashiva> | That provert is new to me |
| 19:12 | <Dashiva> | *proverb |
| 19:16 | <TabAtkins> | foolip: On the vcard and ical tabs, you mean "itemtype" |
| 19:18 | <foolip> | TabAtkins: thanks, typo fixed |
| 23:01 | <othermaciej> | Hixie: I sent a big list of sites that use cut/copy/paste events in email |
| 23:02 | <othermaciej> | Hixie: if you would like help figuring out exactly how they use those events, I can try to gather more data, but I think they all have obfuscated JavaScript so it's hard for me to tell directly |
