| 00:09 | <Hixie> | we get some very odd bug reports through the whatwg form |
| 00:15 | <Lachy> | yes, we know. They were mentioned here earlier |
| 00:15 | <Lachy> | you may have to raise the bar a little for posting through that form to help prevent spam |
| 00:15 | <Lachy> | like requiring people to be logged into the spec annotation system |
| 00:15 | <Lachy> | or forwarding them to bugzilla to complete the bug report |
| 00:17 | <Lachy> | Yay, I hope this now means we can move onto discussing something else for a while |
| 00:17 | <Dashiva> | Yeah, like @alt |
| 00:17 | <hober> | Lachy: exactly the thought I had |
| 00:17 | <Lachy> | no |
| 00:18 | <Lachy> | I expect we'll get a discussion about longdesc again shortly |
| 00:19 | <hober> | longdesc="" is so bad it almost makes summary="" look useful |
| 00:20 | <Dashiva> | Want to make a bet, Lachy? |
| 00:20 | <Dashiva> | First thread with at least 10 posts, alt or longdesc |
| 00:21 | <Lachy> | Dashiva, as long as you promise not to cheat by sending 9 responses to the first mention of alt :-) |
| 00:21 | <Dashiva> | My reputation would only harm the cause if I did, so no |
| 00:22 | <Hixie> | Lachy: the spec annotation system requires them to have sent feedback already, so that somewhat defeats the point of getting people who haven't sent feedback |
| 00:22 | <Hixie> | Lachy: and i can't redirect them to bugzilla if they haven't got an account yet, which seems likely |
| 00:22 | <Lachy> | Hixie, is the summary attribute explicitly excluded from the table elements' Content Attributes? |
| 00:23 | <Hixie> | yeah, because if we include it it'd have to say something about being discouraged, and we're avoiding saying anything bad about summary="" lest the attribute or its friends be offended |
| 00:24 | <Lachy> | what about letting logged in users have their bugs filed immediately, and others held for moderation (perhaps posted into IRC by a new IRC bot first, and allowing anyone in here to moderate it |
| 00:24 | <othermaciej> | I think you could list it and just link to the section about it at the bottom |
| 00:24 | <othermaciej> | but I don't think it's a big deal either way (at least to me) |
| 00:24 | <Hixie> | Lachy: write me a moderation system and i'll get right on that |
| 00:25 | <Lachy> | but that requires me to do work |
| 00:25 | <Lachy> | but I'll think about it. I'd like to learn how to write an IRC bot one day, and this could give me a good excuse |
| 00:26 | <Hixie> | if you have somewhere to host it, mozbot is a great bot infrastructure that lets you write new modules easily. |
| 00:26 | <Hixie> | and i don't say that just cos i wrote it :-P |
| 00:27 | <Lachy> | what's it written in? |
| 00:27 | <Lachy> | or what does it require the modules to be written in? |
| 00:27 | <Hixie> | perl and perl |
| 00:27 | <Lachy> | damn. |
| 00:28 | <Lachy> | I will find somewhere to host it |
| 00:29 | <Lachy> | I wonder if my other web host, A Small Orange, will allow me to run IRC bot from there? |
| 00:29 | <Lachy> | IIRC, Dreamhost doesn't allow that |
| 00:32 | <Philip`> | Do it as a mIRC script |
| 00:32 | <Lachy> | does that mean I would have to be running mIRC? |
| 00:34 | <Lachy> | Hixie, I'm wondering if it might be useful mentioning in the note about summary other kinds of disabilities other than blindness that may affect someone's ability to easily comprehend the table, and for whom a visible summary would help |
| 00:35 | <Lachy> | like people with cognitive disabilities |
| 00:36 | <Lachy> | I'm not sure what to say exactly. It might be better to get advice about this from someone who knows more about different types of disabilities than I do |
| 00:36 | <Hixie> | i'm not touching that text again |
| 00:38 | <Hixie> | sigh, i really feel we're short-changing AT users here |
| 00:38 | <Hixie> | and for such stupid reasons |
| 00:42 | <beowulf> | i agree |
| 00:50 | <GPHemsley> | Hixie, Lachy: Why not look into the MozBot (or whatever the software's name is)? |
| 00:50 | <GPHemsley> | oh |
| 00:50 | <GPHemsley> | should've finished reading |
| 00:57 | <Dashiva> | wtf: http://www.w3.org/Bugs/Public/show_bug.cgi?id=7224 |
| 00:57 | <Hixie> | ok gotta go |
| 00:57 | <Hixie> | bbl |
| 00:58 | <beowulf> | Lachy: you can run bots on dreamhost ps, i run a supybot from one |
| 01:28 | <Lachy> | beowulf, http://wiki.dreamhost.com/KB_/_Unix_/_Cron_Jobs_and_Persistent_Processes#What_is_your_persistent_.28background.29_process_policy.3F |
| 01:29 | <Lachy> | beowulf, I don't pay for dreamhost ps, shared hosting is sufficient for my needs |
| 01:31 | <othermaciej> | I need to write test cases for my <nav> implementation now |
| 07:41 | <hsivonen> | sigh. Larry poked in-band versioning again. |
| 07:56 | <GPHemsley> | How come image width/height only allows integer values? |
| 07:57 | <Dashiva> | Because pixels like to pretend they're atomic |
| 07:59 | <GPHemsley> | heh |
| 08:01 | <annevk2> | Hixie, iput.files needs to be readonly |
| 08:03 | <annevk2> | Hixie, also, do we really want input.files != input.files? |
| 08:03 | <zcorpan_> | interesting how many on twitter are amazed by the audio&canvas demo |
| 08:03 | <zcorpan_> | i guess we should have made such a demo when opera 9.0 was released |
| 08:04 | <GPHemsley> | it's a good sign, at least |
| 08:09 | <hsivonen> | do I understand correctly that JF wants the spec to prescribe a UI string for validators? |
| 08:10 | annevk2 | wonders if and if, how, MS Paint is accessible |
| 08:13 | <zcorpan_> | for a painting app, you probably want to activate the platform's move-cursor-with-keyboard accessibility feature |
| 08:15 | <zcorpan_> | but to be honest, i don't know why anyone would want to paint with the keyboard |
| 08:16 | <zcorpan_> | maybe hardware will solve the painting thing by moving towards touch screens |
| 08:17 | <Hixie> | annevk2: a live input.files seems like it'd get confusing, i thought we were avoiding live apis where we made new ones up |
| 08:22 | <annevk2> | I'd imagine that once the user selects a bunch of files the UA creates a FileList object |
| 08:22 | <annevk2> | then everytime input.files is accessed that is returned until the user makes a new selection |
| 08:23 | <annevk2> | at which point a new object is created |
| 08:24 | hsivonen | wonders what the normal "Hardware ECC recovered" rates for disks are |
| 08:24 | hsivonen | also wonders if a ReadyNAS prefers to do RAID reads from disk one |
| 08:24 | <hsivonen> | I see radically different error recovery counts on the two disks |
| 08:28 | <Hixie> | annevk2: i guess we could do that |
| 08:29 | <Hixie> | annevk2: file a bug or send mail or something? |
| 08:29 | <hsivonen> | https://twitter.com/myakura/statuses/3157898371 |
| 08:30 | <Hixie> | let them know that the mailing list is the right place to send feedback... |
| 08:33 | <annevk2> | Hixie, also, DataTransfer.files? |
| 08:34 | <hsivonen> | looks like techcrunch missed that Firefox 3.5 is no longer in beta... |
| 08:37 | <Hixie> | annevk2: i'm going down the list, one at a time |
| 08:37 | <annevk2> | techcrunch also called H.264 proprietary |
| 08:38 | <annevk2> | it's not really informed journalism afaict |
| 08:38 | <annevk2> | for the admittedly few articles I read there |
| 08:40 | <MikeSmith> | hsivonen, Hixie: the "them" in the case of that tweet is my friend Masataka Yakura |
| 08:40 | <MikeSmith> | I will encourage him to post to the list about it |
| 08:45 | <hsivonen> | annevk2: surely it's informed to call H.264 proprietary in the FSF sense even if not in the ISO sense |
| 08:48 | <hsivonen> | what's the problem with simply having <rp> as conforming? |
| 08:48 | <annevk2> | it is conforming |
| 08:49 | <MikeSmith> | the tweet was about <rb> |
| 08:49 | <hsivonen> | oh |
| 08:49 | <annevk2> | I'm thinking it maybe should not be so you do not need to bother about fallback |
| 08:49 | <annevk2> | <ruby> support is mandatory after all |
| 08:50 | <MikeSmith> | "proprietary" seems like a good word to avoid using at all |
| 08:50 | <hsivonen> | annevk2: video is mandatory, too, but we still have fallback for legacy browsers |
| 08:51 | <MikeSmith> | Alan Ginsberg used to make a point of using alternative terms like "unlegal drugs" instead of "illegal drugs" |
| 08:52 | <MikeSmith> | maybe people should just say "unfree" instead of "proprietary" |
| 08:53 | <MadAtWork> | video is something a browser could be unable to support due to platform restraints. |
| 08:53 | <MadAtWork> | ruby sounds like something that even lynx/links can support in a way or another. |
| 08:53 | <annevk2> | hsivonen, ah yeah, I forgot to mention that the other data point I had was that myakura suggested ruby fallback was very poor |
| 08:59 | <hsivonen> | MadAtWork: we have a blanket exception for platform limitations anyway |
| 09:00 | <sayrer> | annevk2: you have to pay to use h264 |
| 09:01 | <sayrer> | annevk2: if they didn't own it, presumably you wouldn't have to pay them |
| 09:01 | <sayrer> | annevk2: so, proprietary |
| 09:01 | <MadAtWork> | My point was that for video, or image, or audio, a fallback is necessary, for ruby or rp it's not that useful. |
| 09:02 | <annevk2> | sayrer, I always thought proprietary/standards-based was separate from RF |
| 09:02 | <sayrer> | annevk2: proprietary is an english word |
| 09:02 | <sayrer> | annevk2: I'm sure SDOs have definitions that suit them |
| 09:03 | <MadAtWork> | proprietary software can be free. |
| 09:03 | <annevk2> | seems I was misinformed |
| 09:04 | <sayrer> | annevk2: well, I'm interested in non-proprietary codecs that one would have to pay for |
| 09:05 | <sayrer> | but this is just grade school level debate. probably above your average HTML5 argument. |
| 09:05 | <annevk2> | did I do anything wrong? |
| 09:05 | <sayrer> | not lately :) |
| 09:06 | <sayrer> | afaik! |
| 09:06 | <annevk2> | no need to be so rude then |
| 09:06 | <sayrer> | your average --> does not mean Anne van Kesteren's arguments |
| 09:06 | <sayrer> | in case it sounded too personal |
| 09:06 | <sayrer> | apologies |
| 09:07 | <sayrer> | "your average" means something like "the typical" in american english |
| 09:11 | <Philip`> | http://www.on2.com/index.php?594 - "Avoid potential costly royalties associated with open standards." |
| 09:11 | <Philip`> | http://www.on2.com/index.php?308 - "our proprietary On2 Video compression technology" |
| 09:12 | <jgraham> | Philip`: Just tell us what the OED has to say about the matter :) |
| 09:12 | <Philip`> | They seem to use proprietary/open independently of RF status |
| 09:12 | <sayrer> | Philip`: that is funny! |
| 09:12 | <MadAtWork> | sayrer: funny, "your average" doesn't actually refer to the statistic average, more like to the statistic mode :P |
| 09:12 | <MadAtWork> | AH the quirks of the language. |
| 09:13 | <Philip`> | jgraham: Dictionary definitions are boring and pointless :-p |
| 09:13 | <MadAtWork> | Philip`: that "potential costly royalties" puzzled me as well yesterday. |
| 09:13 | <sayrer> | MadAtWork: I am very bad at avoiding idioms of the USA. I get burned quite often! |
| 09:13 | <jgraham> | Philip`: Often they are interesting and useful :) |
| 09:14 | <sayrer> | HTML5 is an open standard |
| 09:14 | <MadAtWork> | sayrer: idioms are part of a human conversation, no need to apologize, but it's still funny how technical term and idiomatic expression mismatch. |
| 09:14 | <sayrer> | "HTML5 is an open standard" |
| 09:14 | <sayrer> | epither or complitment |
| 09:15 | <sayrer> | epithet, even |
| 09:15 | Philip` | hates it when essays/articles talking about X start with giving a dictionary definition of X, because it just seems incredibly lazy |
| 09:15 | <MadAtWork> | jgraham: I guess people enjoys spinnging the meaning of words. |
| 09:16 | <sayrer> | HTML5 is an open standard. I will charge everyone the same amount as I license my IPR. |
| 09:16 | <Philip`> | By the way, someone should tell the On2 site designers about the value of meaningful URLs |
| 09:16 | <sayrer> | URLs don't have meaning |
| 09:16 | <Philip`> | They do to humans |
| 09:16 | <MadAtWork> | They don't on their own, but they gain meaning. |
| 09:17 | <sayrer> | maybe to humans who can parse the bizaree bidirectional URL syntax |
| 09:17 | <sayrer> | nobody would wish it on their worst enemy. it is an abomination. |
| 09:18 | <MadAtWork> | Hm, to put in image macro terms: "pretty url is pretty". |
| 09:18 | <MadAtWork> | Which stands as a value on its own, of course by pretty I mean readable. |
| 09:18 | <Philip`> | sayrer: You don't need to be an expert to derive more meaning from a hypothetical "http://www.on2.com/on2-video-advantages"; than from "http://www.on2.com/index.php?594"; |
| 09:19 | <sayrer> | Philip`: dude, people don't even know the difference between the browser UI and the web page |
| 09:19 | <Philip`> | and it would make me much less likely to get mixed up when pasting links into IRC |
| 09:19 | <hsivonen> | index.php?124 is *so* 2002 |
| 09:20 | <sayrer> | Philip`: do you get that? people don't understand the different trust one should give the web page and the browser chrome? |
| 09:20 | <sayrer> | it makes most browser nerd arguments seem pretty stupid |
| 09:20 | <hsivonen> | sayrer: yay for EV certs, then |
| 09:21 | <Philip`> | sayrer: I don't care about those people, I only care about me |
| 09:21 | <sayrer> | Philip`: that's what the loser of the argument always says |
| 09:21 | <sayrer> | hsivonen: exactly |
| 09:21 | <MadAtWork> | sayrer: I trust people will eventually grow up being more and more knowledgeable about web stuff? |
| 09:21 | <sayrer> | nope |
| 09:22 | <hsivonen> | MadAtWork: haha |
| 09:22 | <sayrer> | user education doesn't work |
| 09:22 | <MadAtWork> | You have a crystal ball I don't? |
| 09:22 | <sayrer> | it's a hard problem |
| 09:22 | <MadAtWork> | Um, no it's not. |
| 09:22 | <jgraham> | User education is somewhat different from people growing up with someing as a fundamental part of their culture |
| 09:22 | <sayrer> | MadAtWork: show me a study that praises user education |
| 09:22 | <MadAtWork> | It's a hardly addressed problem. |
| 09:23 | <MadAtWork> | sayrer: My 14 yo sister knows more about pcs than my father does. |
| 09:23 | <Philip`> | We could rank people by order of knowledgability of web stuff, and then kill the bottom half, and that'd greatly improve the average |
| 09:23 | <sayrer> | MadAtWork: not sure you have a strong argument here |
| 09:24 | <sayrer> | Philip`: I've been phished |
| 09:24 | <Philip`> | which is probably a similar effect to waiting 40 years |
| 09:24 | <sayrer> | you? |
| 09:24 | <jgraham> | (I don't know if people growing up with the web also grow up understaning the difference between the chrome and the web page though) |
| 09:24 | <sayrer> | jgraham: it is really a very subtle concept |
| 09:24 | <MadAtWork> | Philip`: and somehow at the same time it might relieve some pension fund costs... hmmm... |
| 09:26 | <MadAtWork> | In every company I've been the young people knew the difference, the 30-something sometimes did, 40+ people didn't unless they had a background in cs/it. |
| 09:26 | <Philip`> | Someone should make a web browser which uses primary colours for its chrome and renders all web page content in monochrome, to make the difference clearer |
| 09:26 | <MadAtWork> | The stats are lower if you go around asking what a web browser is. |
| 09:26 | <sayrer> | Philip`: I guess you agree |
| 09:41 | zcorpan_ | notes that networkState might not be NETWORK_LOADING when inspecting it in the onprogress event handler |
| 09:42 | <zcorpan_> | 1. you receive a byte and decide to que a task to fire a progress event |
| 09:42 | <zcorpan_> | 2. you find that the video now has completed loading and set networkState to NETWORK_LOADED |
| 09:42 | <zcorpan_> | 3. the event loop spins and fires the progress event |
| 09:43 | <zcorpan_> | the non-normative event summary table is convenient for writing test cases but might result in inaccurate or wrong tests |
| 10:07 | <hsivonen> | I'm again annoyed by Google not providing obvious ways to file bugs against their services |
| 10:08 | <hsivonen> | finally found a bug report form |
| 10:12 | <Philip`> | I imagine an obvious easily-findable bug report form would not be very scalable when you have a billion users |
| 10:13 | <jgraham> | Oh I always thought you were supposed to use the HixieBot for bug reports |
| 10:13 | <hsivonen> | maybe having to go through multiple levels of searching in order to file a bug is a bozo filter |
| 10:13 | <Philip`> | I just type bugs into the http://google.com search box, and assume they'll find it when analysing the logs |
| 10:14 | <MadAtWork> | haha |
| 10:15 | <hsivonen> | It's so sad that with all the compute power and face recognition in Street View, YouTube still hasn't as simple image analysis as detecting and autocropping out letterboxing |
| 10:16 | <roc> | Philip`: send your complaints about the NSA the same way |
| 10:17 | <Philip`> | roc: I prefer to send my compaints about the NSA over HTTPS, to make them stand out more to the loggers |
| 10:17 | <Rik`> | isn't the NSA able to collect bugs from our thoughts yet ? |
| 10:18 | <roc> | not as far as you know |
| 10:18 | <Philip`> | Rik`: That's how I complain about Uri Geller |
| 10:19 | <Hixie> | we're definitely going to have to make dataTransfer.types visible during drags |
| 10:19 | <Hixie> | but what should we do about dataTransfer.files? |
| 10:20 | <roc> | I got to use the quadratic formula for work today |
| 10:20 | <Rik`> | (thx wikipedia for giving me a quick hint about uri geller) |
| 10:22 | <hsivonen> | (even more sad that YouTube has a "let's fix this in documentation" fix instead of a silent software fix) |
| 10:25 | <Philip`> | roc: Did you look on Wikipedia to double-check the equation? |
| 10:25 | Philip` | always does that for things he really ought to know :-( |
| 10:25 | <roc> | I am dreadfully embarrassed to say that I did |
| 10:25 | <roc> | but at least I had remembered it correctly |
| 10:25 | <Hixie> | i didn't remember that even when i was at uni |
| 10:26 | <Hixie> | i just wrote a program for it on my TI-83 |
| 10:26 | <Philip`> | (-b+/-sqrt(b^2-4ac))/2a or something? |
| 10:26 | <roc> | correct |
| 10:26 | <annevk2> | Hixie, just on dropping right? |
| 10:26 | <Philip`> | You ought to just work it out from first principles |
| 10:26 | <annevk2> | Hixie, isn't that the case for other data too? |
| 10:26 | <Hixie> | annevk2: for the other data, we can just expose .types |
| 10:31 | <jgraham> | Philip`: I am pleased to find I have not forgotten how to do that already |
| 10:32 | <Philip`> | Unfortunately I have :-/ |
| 10:34 | <jgraham> | (the trick is that ax^2 + bx + c = 0 === (x + b/2a)^2 + c/a - (b/2a)^2 = 0 |
| 10:34 | <jgraham> | ) |
| 10:35 | <Rik`> | jgraham: if a != 0 |
| 10:36 | <jgraham> | Rik`: Well yes, but it's not a quadratic if a = 0 |
| 10:36 | <Philip`> | You're also assuming 2 != 0 |
| 10:36 | <Philip`> | otherwise the answer will be undefined |
| 10:38 | <Hixie> | annevk2: but i don't know how to do that for .files |
| 10:40 | <MikeSmith> | I notice that http://wiki.whatwg.org/wiki/HTML_vs._XHTML claims that <iframe> is a CDATA element. Is that accurate or not? The current spec doesn't indicate it's a CDATA element. |
| 10:40 | <Philip`> | "Lachy [n=Lachlan⊙lpc] has quit" - it doesn't seem such perfect privacy when http://www.google.com/search?q=%22london.perfect-privacy.com points to Lachy as the fourth search result |
| 10:42 | <Lachy_> | LOL |
| 10:43 | <jgraham> | Given their inability to write english sentences, for example "With Perfect Privacy you can anonymize and encrypt your whole Internet", I'm not sure that's a great surprise |
| 10:43 | <Philip`> | MikeSmith: "A start tag whose tag name is "iframe" ... Follow the generic CDATA element parsing algorithm." |
| 10:43 | <Philip`> | says the tree constructor section |
| 10:44 | <MikeSmith> | Philip`: thanks |
| 10:45 | <Lachy> | Philip`, in this case, I still have significant plausible denyability since I'm not the only one using their service |
| 10:45 | <Lachy> | *deniability |
| 10:46 | <MikeSmith> | Hixie: seems like the CDATA list in http://dev.w3.org/html5/spec/syntax.html#elements-0 should have <iframe> included |
| 10:46 | <MikeSmith> | want me to file a bug on it? |
| 10:47 | <Philip`> | Lachy: Given the number of search results, there aren't a huge number of other people using their service |
| 10:50 | Philip` | wonders if you could look at the TCP packets and timings received from the anonymising proxy, and deduce the latency of the user (and thus get some insight into their geographic location) |
| 11:14 | <virtuelv> | encrypting the whole Internet |
| 11:14 | <virtuelv> | accidentally |
| 11:16 | gsnedders|work | often does that |
| 11:18 | <virtuelv> | http://www.reddit.com/r/funny/comments/6zk5r/i_accidently_a_fleshlight/ ← for those missing the reference |
| 11:21 | <Lachy> | Philip`, that's probably because most services people use don't publicly publish their access logs where they can get indexed by search engines, unlike IRC |
| 11:25 | <Lachy> | virtuelv, the images in the forum that reddit page links to seem to be unavailable. |
| 11:25 | zcorpan | notes that .duration might have interop problems especially for files where the audio component and video component have different lengths |
| 11:26 | <Lachy> | virtuelv, what does it mean to "accidentally a fleshlight"? (I know what a fleshlight is, I'm just curious what's being done with it) |
| 11:26 | <Lachy> | well, other than the usual stuff |
| 11:29 | <virtuelv> | Lachy: that's the point of the meme, there is something missing between accidentally and the (whole) fleshlight |
| 11:29 | <virtuelv> | in this case, there is definitely something missing between "your" and "whole internet" |
| 11:31 | <Lachy> | virtuelv, how is that a meme? |
| 11:32 | <Philip`> | virtuelv: Why is something missing? |
| 11:32 | <virtuelv> | http://www.google.com/search?client=opera&rls=en&q=%22accidentally+the+whole%22&sourceid=opera&num=100&ie=utf-8&oe=utf-8 |
| 11:32 | <virtuelv> | I'm not pointing directly to the top result there |
| 11:32 | <Philip`> | virtuelv: It seems to make grammatical sense to me |
| 11:33 | <annevk2> | it's tempting to write "Hi Sam, I agree that the wording on summary is sufficient for publication. However, like others I think it needs to be made non-conforming before Last Call." |
| 11:33 | <virtuelv> | Philip`: does the sentence «I accidentally the whole bottle» make sense to you? |
| 11:34 | <virtuelv> | you accidentally WHAT the whole bottle? |
| 11:34 | <Philip`> | virtuelv: anonymize and encrypt it |
| 11:34 | <jgraham> | annevk2: Isn't it just |
| 11:35 | <annevk2> | hm? |
| 11:36 | <virtuelv> | Philip`: I don't have a whole internet anywhere |
| 11:37 | <jgraham> | annevk2: Tempting to write that. It seems there is a tactic of drawing a line in the middle that people will converge on and then trying to slowly move the line toward the previous, unacceptable, position |
| 11:39 | <annevk2> | yeah, I'm also annoyed with all the overhead these meaningless emails generate |
| 11:39 | <Lachy> | http://www.encyclopediadramatica.com/I_accidentally_X |
| 11:39 | <annevk2> | we already know the summary issue is not closed |
| 11:39 | <annevk2> | no need to hear it again from all the naysayers |
| 11:40 | <jgraham> | This "whole internet" thing reminds me of http://www.youtube.com/watch?v=QAUyaELfwBo |
| 11:40 | <Philip`> | virtuelv: It doesn't make semantic sense (given that the Internet does not belong to anybody), but that's completely different to not making grammatical sense |
| 11:44 | <jgraham> | Philip`: It doesn't make grammatical sense either I think although I'm not sure I can explain the technical reason |
| 11:45 | <jgraham> | But grammer allows for things like countable vs uncountable nouns |
| 11:45 | <jgraham> | so "I want more waters" is grammatically wrong if you mean "I want more water" |
| 11:46 | <jgraham> | (whereas "I want more emails" is fine because email is countable) |
| 11:49 | <Lachy> | jgraham, that depends if you consider the plural of email to be "email" or "emails" |
| 11:50 | <Lachy> | hmm, I think there are cases where either can be used, and others where only one works. |
| 11:50 | <Philip`> | It could say "anonymize and encrypt your whole intranet", and that would be perfectly fine if e.g. you were a network administrator, and the only difference with Internet is that there's only one Internet |
| 11:51 | <hsivonen> | sigh. still stalled with updating and relauncihng the v.nu system, because I hod forgotten to press a button that makes my DNS changes actually have an effect |
| 11:51 | <hsivonen> | usability++ |
| 11:52 | <jgraham> | Philip`: That I think is the critical difference. But I wasn't sure if "singleton" is a concept in grammar |
| 11:52 | <Philip`> | jgraham: If someone made a second Internet, would the sentence magically change from grammatically incorrect to grammatically correct? |
| 11:54 | <jgraham> | Philip`: Maybe. I'm also unclear if it is considered a syntactic error to place a non-owned-item after a possesive or if that is a sematic error |
| 11:54 | <Lachy> | I wonder if that could be considered false advertising, since the IRC messages I posted while I was connected to the VPN are no longer encrypted when others receive them |
| 12:01 | Philip` | wonders how many 'anonymising' services are secretly run by government departments |
| 12:18 | <Lachy> | Philip`, http://www.perfect-privacy.com/faq.html#faq6 |
| 12:22 | <jgraham> | Lachy: It seems like a good way of targetting people with something to hide |
| 12:23 | <Lachy> | yeah, probably. |
| 12:23 | <Lachy> | But I don't use it for illegal activities. I only access legal services through it |
| 12:23 | <Lachy> | like BBC iPlayer and Hulu |
| 12:24 | <jgraham> | That is probably illegal if you are not a UK license payer |
| 12:24 | <jgraham> | (iPlayer) |
| 12:25 | <hsivonen> | I think territorial restrictions on Web sites are antithetical to the Web |
| 12:26 | hsivonen | is still bitter that Daily Show now blocks visitors by country |
| 12:26 | <Lachy> | that's their fault for not offering an alternative without territorial restrictions, even if users outside the UK had to pay a small fee |
| 12:26 | <Philip`> | UK residents only need a TV license for live video, not prerecorded stuff that's on iPlayer |
| 12:26 | <hsivonen> | also, IMO, everyone anywhere should be able to pay and subscribe to the BBC |
| 12:27 | <Philip`> | I guess that if you're not in the UK then it's just plain copyright infringement, independent of any TV licensing |
| 12:27 | <Lachy> | Philip`, the UK should just make the BBC tax payer funded instead of requring TV licences |
| 12:27 | <Philip`> | That seems very unfair on people who don't watch TV at all |
| 12:28 | <hsivonen> | new legislation in Finland will consider TVless people such oddballs that they can be charged as collateral damage |
| 12:28 | <jgraham> | I would for sure subscribe to the BBC if it were possible. But that would require the BBC to forfeit their ability to sell their shows abroad |
| 12:29 | <jgraham> | Which I assume they don't want to do |
| 12:29 | <Lachy> | jgraham, what's the difference? Either way, theyr'e getting money. What does it matter if it's thorugh subscription fees or DVD sales? |
| 12:29 | <hsivonen> | jgraham: If I don't need to pay for subtitling, why shouldn't I be able to purchase the shows from the BBC directly? |
| 12:29 | <Philip`> | I believe the BBC says the reason it doesn't have (advertisting-supported) non-UK iPlayer access is that it doesn't have the rights for most programmes, and negotiating the rights is a long and complex and ongoing process |
| 12:29 | <jgraham> | Lachy: Not sell as DVDs but sell to other networks |
| 12:29 | <jgraham> | or subsiduries such as BBC America |
| 12:30 | <hsivonen> | jgraham: the concept of regional networks sucks |
| 12:30 | <Lachy> | jgraham, you're making the false assumption that internet distribution would kill TV distribution, which it won't. |
| 12:31 | <Lachy> | that's like saying just because films are available on DVD before they air on free-to-air TV, then TV networks wouldn't buy those films and show them, but that is clearly not the case |
| 12:32 | <jgraham> | Lachy: I'm making the reasonable assumption that offering the shows online reduces their resale value by reducing the market |
| 12:32 | Philip` | has noticed the thumbnails on iPlayer for movies are almost always photos of actors in interviews, not images from the film itself, and wonders if that's rights-related |
| 12:33 | <Lachy> | they just need to adapt their business model to the times, rather than simply fight to maintain an old business model designed for last century |
| 12:33 | <jgraham> | hsivonen: Indeed, but it's what we're stuck with for now |
| 12:33 | <hsivonen> | in the case of The Daily Show and Colbert Report, no regional network carries them |
| 12:34 | <hsivonen> | and they don't even sell them to me by episode with DRM infestation on iTunes |
| 12:34 | <hsivonen> | they don't monetize my willingness to pay in any way |
| 12:34 | <hsivonen> | but I guess I'm preaching to the choir here |
| 12:36 | <Lachy> | fwiw, I would quite happily spend the $100/month or so I spend on services that allow me to access content that isn't otherwise available legally in my area on legitimate services if they fixed their business models |
| 12:38 | <Lachy> | actually, it's only about $70 |
| 13:09 | <hsivonen> | I wonder if https://bugzilla.mozilla.org/show_bug.cgi?id=508075 has ever been a practical problem for Opera |
| 13:11 | <gsnedders|work> | hsivonen: Yes |
| 13:13 | <zcorpan> | closing on <body> would be the wrong fix |
| 13:13 | <zcorpan> | the right fix is reparsing |
| 13:14 | <zcorpan> | which is a wontfix :) |
| 13:16 | <gsnedders|work> | hsivonen: (Very very very rare, though) |
| 13:23 | <gsnedders|work> | Anyone know about FF plans for outerHTML support? |
| 13:25 | <hsivonen> | gsnedders|work, zcorpan: ok. thanks |
| 13:43 | <hsivonen> | hmm. I wonder if Parallels hung randomly or if Parallels doesn't work properly when the VM hard disk image is on an AFP volume |
| 19:47 | <ojan> | Hixie: you there? |
| 19:48 | <ojan> | annevk2: or you? i bet you'd know the answer to my question. |
| 19:50 | <paul_irish> | is there a more appropriate IRC channel for the webfont conversation? |
| 19:51 | <othermaciej> | people have discussed webfonts here |
| 19:51 | <othermaciej> | this isn't necessarily the home base of all people with an interest |
| 19:53 | <annevk2> | ojan, somewhat |
| 19:53 | <annevk2> | are webfonts even discussed over IRC? |
| 19:54 | <Lachy_> | annevk2, not frequently |
| 19:54 | <ojan> | annevk2: bit of a n00b question...i want to propose an extension to the CSS position property. is www-style the right mailing list? |
| 19:54 | <annevk2> | yeah |
| 19:54 | <ojan> | annevk2: k. thanks. |
| 20:00 | Lachy | ponders whether it's worth responding to the summary issue again explaining why I'm not very happy about the compromise, but can accept it for now, at least until I get more evidence |
| 20:02 | <Lachy> | ... probably not, but I should at least write my thoughts down for future reference |
| 20:23 | <othermaciej> | Lachy: I don't think it would be disruptive to do that |
| 20:23 | <othermaciej> | Lachy: I think it could be constructive |
| 20:24 | <othermaciej> | to show that there are real concessions being made by both sides |
| 20:36 | <Dashiva> | It's good to see that we're making progress on listening to each other's arguments: http://lists.w3.org/Archives/Public/public-html/2009Aug/0341.html |
| 20:40 | <othermaciej> | Dashiva: yeah... |
| 20:44 | <annevk2> | it somewhat worries me that Manu's email does not mention the xmlns namespace issue at all |
| 20:44 | <annevk2> | it has been repeatedly pointed out as at least a potential issue to the RDFa guys |
| 20:45 | <Philip`> | As long as you can hack it with namespace-unaware APIs to work in simple cases, surely it's good enough to rely on as a fundamental part of a technology intended to become ubiquitous and widely deployed |
| 20:45 | <Dashiva> | It's a feature, not an issue |
| 20:46 | <annevk2> | I'm always met with sarscasm here, damn you guys! |
| 20:48 | <othermaciej> | annevk2: what's the xmlns namespace issue? |
| 20:48 | <annevk2> | in XML xmlns is put in a namespace |
| 20:48 | <annevk2> | a special "XMLNS Namespace" |
| 20:49 | <annevk2> | this does not happen in HTML |
| 20:49 | <Philip`> | (except in SVG/MathML in HTML, I think) |
| 20:49 | <othermaciej> | Philip`: my vague recollection is that namespace-unaware APIs are not entirely interoperable for attributes with colons in them in current browsers |
| 20:50 | <othermaciej> | annevk2: I wasn't really clear on why Manu was investigating what happens with xmlns attributes |
| 20:50 | <othermaciej> | annevk2: maybe he misunderstood the real issue |
| 20:50 | <Philip`> | othermaciej: But parsing and getAttribute seems to work, which is apparently good enough for the RDFa people |
| 20:51 | <othermaciej> | you know, it wouldn't be so hard for them to design something that is essentially like RDFa but avoids namespaces so that weirdness in text/html can be avoided |
| 20:51 | <othermaciej> | they would get a lot more traction that way, I think |
| 20:51 | <Dashiva> | Can I call it irony that a technology from XHTML doesn't use the very methods added for it? |
| 20:52 | <annevk2> | othermaciej, yeah I suppose |
| 20:52 | <annevk2> | othermaciej, I mean if what he just reported was news to him o_O |
| 20:52 | <Philip`> | othermaciej: That'd break compatibility with legacy content that uses xmlns:foo in text/html, though |
| 20:52 | <Philip`> | (and legacy parsers and legacy tutorials etc) |
| 20:53 | <Dashiva> | What's the phrase again? Delay, divert, too late? |
| 20:54 | <gsnedders> | Philip`: Note that we tried to support @xmlns in text/html, IIRC, and had it break tons |
| 20:55 | <Hixie> | othermaciej: microdata is "essentially like rdfa" but avoids the problems with rdfa |
| 20:55 | <Philip`> | gsnedders: You might have tried to support it in a stupid way, so that's not evidence that it's impossible to do anything differently to what's done today |
| 20:58 | <gsnedders> | Philip`: Well I didn't work then |
| 20:58 | <Philip`> | gsnedders: (I'm using "you" in the same sense that you used "we") |
| 20:59 | <gsnedders> | Philip`: Yes, I know… But it means I know nothing |
| 20:59 | <Philip`> | gsnedders: I've not seen anyone describing what Opera actually tried to implement, and "support @xmlns in text/html" is extremely vague, so there doesn't seem to be enough information to conclude anything |
| 21:00 | <othermaciej> | annevk2: I think it was news to him |
| 21:00 | <Dashiva> | I think they made xmlns:foo be (xmlns, foo) instead of (null, xmlns:foo) |
| 21:00 | <othermaciej> | annevk2: based on earlier private communication with him, he thought there had to be a new spec requirement to require preserving xmlns nodes in the DOM as-is |
| 21:01 | <annevk2> | then he misunderstood the issue |
| 21:01 | <annevk2> | the issue was that xmlns was not put in the correct namespace when it went into the DOM |
| 21:01 | <annevk2> | the other issue of course is that xmlns would mean something completely different in HTML (it would only apply to RDFa) |
| 21:01 | <Philip`> | That's only a problem if there are any practical consequences to it not being put in the correct namespace |
| 21:01 | <othermaciej> | Hixie: I'm thinking maybe they could come up with a version that solves the namespace issues, but still have it be close enough to "RDFa in XHTML" that they would be happy to still call it RDFa |
| 21:02 | <othermaciej> | Hixie: maybe that exercise would lead them to conclude that microdata meets their needs, I dunno |
| 21:02 | <Hixie> | seems unlikely |
| 21:02 | <Philip`> | (so I suppose it would be good to demonstrate those practical consequences) |
| 21:02 | <Hixie> | they're never going to make a version that doesn't have prefixes at all |
| 21:02 | <othermaciej> | Hixie: in any case it seems better than completely rejecting all that feedback |
| 21:03 | <othermaciej> | annevk2: maybe you could help him see that he may have misunderstood the issue |
| 21:05 | <annevk2> | mah, you're too good |
| 21:06 | <annevk2> | I just like to bitch about these things and let someone else care, oh well, I'll write it |
| 21:09 | <Dashiva> | annevk2: You lost that chance when you were the serious one vs Philip earlier |
| 21:11 | <annevk2> | pretty hard to get a break in this channel too :p |
| 21:13 | <Dashiva> | It's okay, we still love you |
| 21:25 | <Hixie> | i hope we have a vote on this summary stuff, i really don't like what's in the spec now |
| 21:26 | <gsnedders> | Hixie: Can you fix the the issue you added when fixing the infinite loop? |
| 21:27 | <Hixie> | it's on the pile |
| 21:27 | <Hixie> | is it urgent? |
| 21:28 | <Hixie> | i'm trying to do the file api stuff first |
| 21:28 | <gsnedders> | Hixie: Not really |
| 21:28 | <Hixie> | btw can anyone reproduce http://stanford.edu/~jlebar/moz/screen1.png ? |
| 21:31 | <Philip`> | Hixie: Yes |
| 21:31 | <Philip`> | Hixie: e.g. set minimum font size to 16 |
| 21:31 | <Philip`> | (in Firefox) |
| 21:34 | <othermaciej> | Hixie: it looks like it may be heading in a direction where it will only be settled by vote |
| 21:34 | <othermaciej> | I'm ok with that as long as Sam actually calls a vote to decide the issue at some point, instead of letting the issue fester forever |
| 22:05 | <gsnedders> | Hixie: But it'd be nice to have a parsing algorithm that worked :P |
| 22:05 | <Hixie> | gsnedders: i'll get there |
| 22:05 | <Hixie> | Philip`: aah, interesting |
| 22:05 | <Hixie> | Philip`: i wonder how to work around that |
| 23:15 | jgraham | mutters something |
| 23:18 | <gsnedders> | Aahhh! My ears! |
| 23:18 | <gsnedders> | The noise! |
| 23:31 | <Hixie> | sicking: yt? |
| 23:31 | <sicking> | Hixie, pong |
| 23:32 | <Hixie> | when i first specced localStorage, it supported arbitrary structured data |
| 23:32 | <Hixie> | but mozilla wanted it to be only strings instead |
| 23:32 | <Hixie> | now that Files are being introduced, i've got a request from mozilla to allow localStorage to include File or DOMString |
| 23:32 | <Hixie> | would it make sense to just allow any structured data again? |
| 23:32 | <sicking> | bet you didn't see that one coming, did'ya ;) |
| 23:33 | <Hixie> | maybe using the "structured clone" concept? http://www.whatwg.org/specs/web-apps/current-work/#safe-passing-of-structured-data |
| 23:33 | <sicking> | we aim to be unpredictable, never know where you have us. Like Ninjas we are |
| 23:33 | <Hixie> | hah |
| 23:33 | <sicking> | Hixie, hmm.. i would be for that yeah |
| 23:34 | <sicking> | Hixie, i'll have to check with the people that worked with localStorage, but IMHO it makes a lot of sense |
| 23:34 | <Hixie> | k |
| 23:34 | <Hixie> | that means being able to store File, FileList, FileData, RegExps, ImageData, etc |
| 23:35 | <Hixie> | as well as any combination of Object or Array that doesn't have loops in it |
| 23:51 | <sicking> | Hixie, sounds good, though a lot of work |
| 23:51 | <sicking> | Hixie, But I think it's the right thing to do |
| 23:52 | <Hixie> | k |
| 23:52 | <sicking> | RegExps was actually the only thing that stuck out there when i first looked at the HTML spec |
| 23:52 | <sicking> | but i guess they can be serialized |
| 23:53 | <Hixie> | if you can serialise ImageData and FileList, serialising RegExp will be a cake walk |
| 23:53 | <sicking> | yeah, i guess |
| 23:53 | <sicking> | well.. except if they have callback functions |
| 23:54 | <Hixie> | it only clones the pattern and flags |
| 23:54 | <Hixie> | "Return a newly constructed RegExp object with the same pattern and flags as input." |
| 23:54 | <Hixie> | even lastIndex isn't copied |
