| 00:00 | <jacobolus> | bga_: how do you do infinite bounds? |
| 00:00 | <jacobolus> | does that work? |
| 00:00 | <jacobolus> | as in context.clearRect(-Infinity, -Infinity, Infinity, Infinity) ?? |
| 00:00 | <bga_> | yeah |
| 00:01 | <jacobolus> | that works everywhere? |
| 00:01 | <jacobolus> | seems like an edge case that could easily blow up someplace |
| 00:05 | <jcranmer> | presumably most implementations would work by comparison |
| 00:06 | <jcranmer> | if they convert to internal integer coordinates, then it should map to max/min values |
| 00:06 | <jacobolus> | this seems to work just fine: http://pastie.textmate.org/1805281 |
| 00:12 | <bga_> | 'infinite' trick doesnt works in chrome and opera. fail #2 :) |
| 00:22 | <Philip`> | Non-finite or very large or (in some cases) quite large coordinates aren't at all interoperable |
| 00:23 | <Philip`> | e.g. some implementations seem to convert to some fixed-point data type |
| 03:44 | <jacobolus> | kind of unfortunate that canvas doesn't support dotted/dashed lines |
| 03:45 | <jacobolus> | I can manage in the straight line and arc cases easily enough, but manually doing dotted bezier curves is tricky enough to not be worth the effort |
| 04:26 | <jacobolus> | zewt, Philip`: so the "save context, set identity transformation, stroke, restore context" pattern is pretty great; too bad there's no similarly easy way to do text by first storing an "anchor" position and then setting a new transformation to apply to the actual text filling |
| 04:30 | <jacobolus> | if I could pull the matrix out, I could maybe do something like translate so the origin is where I want my anchor, then keep the translation part of the matrix constant while resetting the rest to the identity, and then restore afterward |
| 13:25 | <hsivonen> | the 2022 meme has made it to xkcd |
| 14:04 | <MikeSmith> | hsivonen: cool |
| 14:04 | <MikeSmith> | and only a few years after Jesus returns to earth |
| 14:05 | <jgraham> | Actually we're relying on him to finish off the testsuite |
| 14:07 | MikeSmith | puts Jesus in critical path on his testsuite Microsoft Project plan |
| 14:10 | <MikeSmith> | I just hope he's hip to IRC so that I can ping him and discuss problems in real time |
| 14:35 | <zcorpan> | MikeSmith: you won't need to discuss problems with him |
| 14:35 | <MikeSmith> | well, I mean personal problems |
| 14:35 | <MikeSmith> | like, when I need some life advice |
| 14:36 | <MikeSmith> | currently when I have a problem, I just ask myself, What would Jesus do? |
| 14:36 | <zcorpan> | ah |
| 14:36 | <MikeSmith> | but after he gets back, I can just ask him directly |
| 14:36 | <MikeSmith> | save myself a lot of thinking time |
| 14:37 | <webr3> | "why does jesus test us so much?" - btw I think it said he comes back twice, again in 2025 iirc |
| 14:37 | <zewt> | well, it may be of little help when his answer is invariably "change the gravitational constant of the universe" |
| 14:37 | <wilhelm> | I don't think I want to take life advice from a guy who ended up crucified at the will of an angry mob. |
| 14:39 | <MikeSmith> | he didn't really get crucified |
| 14:39 | <MikeSmith> | they actually crucified Judas (thinking he was Jesus) |
| 14:40 | <wilhelm> | Oh. Clever. |
| 14:48 | <jgraham> | webr3: Isn't coming back multiple times his best-known party trick? |
| 14:50 | <webr3> | jgraham: either that or the wine one |
| 14:56 | <zcorpan> | i read somewhere he wasn't crucified but just nailed to a pole |
| 14:56 | <jgraham> | webr3: I guess the wine one is better for picking up girls. It's a tough pickup tactic that needs a three day hiatus in a cave. |
| 14:57 | <jgraham> | so, erm, blas-for-me, blas-for-you |
| 17:20 | <AryehGregor> | Question to world: how do I get added as a member of the Web Apps WG? Do I contact the Google AC rep to add me, T. V. Raman? |
| 17:20 | <jgraham> | AryehGregor: Yes |
| 17:20 | <AryehGregor> | jgraham, thanks. |
| 17:28 | <TabAtkins> | AryehGregor: And I can save you some time by sending you the link for the webapps signup. TV will ask for it otherwise. ^_^ |
| 17:28 | <AryehGregor> | TabAtkins, heh, yeah, he asked me last time and I had no idea what he wanted. |
| 17:28 | <AryehGregor> | Forgot about that. |
| 17:28 | <AryehGregor> | What's the link? |
| 17:29 | <TabAtkins> | http://www.w3.org/Member/Mail/, find the group you want and click the "join" button. Send the url for that page to TV. |
| 17:32 | <AryehGregor> | Wait, I think I'm already in it. |
| 17:32 | AryehGregor | investigates |
| 17:32 | <AryehGregor> | No, I'm not. |
| 17:32 | <AryehGregor> | But the page is titled "Leaving the Web Applications Working Group". |
| 17:33 | <AryehGregor> | Maybe the participant list doesn't update right away? |
| 17:33 | <karlcow> | You are already part of it |
| 17:33 | <jgraham> | Maybe they're giving you a hint? |
| 17:33 | <AryehGregor> | Okay, so he must have added me already. |
| 17:33 | <mpilgrim> | it's a zen koan |
| 17:33 | <AryehGregor> | But I don't show up here: http://www.w3.org/2000/09/dbwg/details?group=42538&public=1&gs=1&; |
| 17:34 | <karlcow> | you are the shadow DOM of WebApps |
| 17:34 | <mpilgrim> | the shadow DOM liveth forever and ever |
| 17:35 | <mpilgrim> | by cloning itself and appending itself to its parent shortly before removing itself from this world |
| 17:41 | <AryehGregor> | Deep clone or shallow clone? |
| 17:43 | <AryehGregor> | Ooh, I like this definition: http://dvcs.w3.org/hg/domcore/raw-file/tip/Overview.html#concept-clone-ext |
| 17:43 | <AryehGregor> | I want some definitions like that for DOM Range, for mutations. |
| 17:43 | AryehGregor | makes a mental note to write some up later. |
| 17:44 | <AryehGregor> | Another random comment: I love how all these Japanese people are always so incredibly polite. I wish Westerners were more polite. |
| 17:46 | <karlcow> | hmmm it is not really polite. It is the codification of the communication which is different. Difficult to explain. From a western point of view it *seems* polite. |
| 17:46 | <zewt> | seems more "distant" than "polite" to me |
| 17:47 | <zewt> | sort of like retail-employee "politeness" |
| 17:47 | <AryehGregor> | To me, "polite" means things like saying please and thank you, trying to phrase things non-critically, things like that. I'd say it's politeness, by that definition. |
| 17:47 | <AryehGregor> | Of course, it has downsides too. |
| 17:47 | <AryehGregor> | Bluntness can be useful. Israelis tend to be very blunt. |
| 17:48 | <AryehGregor> | Which is to say, not polite. |
| 17:48 | <AryehGregor> | There's a joke about that. |
| 17:49 | <karlcow> | communication protocols. :) |
| 17:49 | <AryehGregor> | A journalist goes up to four people -- an American, a Russian, a Chinese person, and an Israeli. The journalist asks them all, "What's your opinion on the recent meat shortage?" |
| 17:49 | <AryehGregor> | The American says "What's a 'shortage'?" |
| 17:49 | <AryehGregor> | The Russian says, "What's 'meat'?" |
| 17:49 | <AryehGregor> | The Chinese person says, "What's an 'opinion'?" |
| 17:49 | <AryehGregor> | The Israeli says, "What's 'excuse me'?" |
| 17:49 | <AryehGregor> | (maybe the joke was originally told before the Soviet Union's collapse, the Russian part makes more sense then) |
| 17:50 | <TabAtkins> | The joke would work better if the reporter had said "excuse me" at some point. ^_^ |
| 17:50 | <AryehGregor> | Dangit. |
| 17:50 | <AryehGregor> | It was supposed to say "Excuse, what's your opinion on the recent meat shortage?" |
| 17:50 | <AryehGregor> | I always ruin jokes that are more than about one line long. |
| 17:50 | AryehGregor | hangs his head in shame |
| 17:50 | <TabAtkins> | Boooo |
| 17:51 | <AryehGregor> | My father is an excellent joke-teller, but somehow I don't seem to have inherited it.] |
| 17:51 | <AryehGregor> | it. |
| 17:51 | <Philip`> | You could tell the jokes backwards, starting at the punchline, so you can be sure you'll set them up properly |
| 17:51 | <AryehGregor> | That's an excellent idea. I wonder why no one has ever thought of it before. |
| 17:52 | <TabAtkins> | AryehGregor: I like your much-more-polite versions of my responses to those bugs. |
| 17:52 | <AryehGregor> | TabAtkins, see, that's what I mean about being polite. |
| 17:52 | <AryehGregor> | It's not nice to say to some confused newbie "Why don't you read beyond the table of contents?" |
| 17:52 | <karlcow> | The Japanese answer could be something along: "We understand the issue. Thanks." |
| 17:53 | <AryehGregor> | The spec is several hundred pages long and incomprehensible to normal human beings. |
| 17:53 | <AryehGregor> | It's only polite to take thirty seconds to explain properly when someone has taken the effort to give feedback, even if the feedback is misguided. |
| 17:53 | <TabAtkins> | AryehGregor: I'm always somewhat torn. Yes, that's a rude thing to say. It's also rude to even *ask* that question without doing a little bit of research yourself, at which point the answer becomes obvious. |
| 17:54 | <TabAtkins> | Jeezus, Kyoto is a million miles away from Tokyo. |
| 17:54 | <AryehGregor> | Someone without web standards background is likely to be completely confused by the ambiguous terminology, and will have a lot of trouble successfully researching it themselves. |
| 17:54 | <karlcow> | The French answer might be "French meat is delicious" (completely unrelated and pulling it to French culture) or if the journalist is cute "Are you free tonight?" |
| 17:55 | <AryehGregor> | I'd agree with you somewhat more if it were a question like "What's the details element supposed to do?" where just Ctrl-Fing through the ToC would give you the answer easily. |
| 17:55 | <karlcow> | TabAtkins: not by shinkansen |
| 17:55 | <AryehGregor> | But over here you have "HTML" being used in two conflicting ways, which is confusing. |
| 17:55 | <AryehGregor> | Maybe we should officially start calling the syntax only "text/html" or something. |
| 17:55 | <AryehGregor> | Although that's also confusing, since then you're calling it by a MIME type. |
| 17:56 | <TabAtkins> | karlcow: I can't tell if http://goo.gl/maps/Ff8F is via shinkansen or not, because there's too much japanese. |
| 17:57 | <karlcow> | 140 min by Shinkansen from Tokyo to Kyoto |
| 17:57 | <TabAtkins> | Ooh, what's the price on that sort of trip (if you know off the top of your head) |
| 17:57 | AryehGregor | always sort of wonders if someone is trying to track him, when he's given a goo.gl URL |
| 17:58 | <karlcow> | 10,000 yens |
| 17:58 | <karlcow> | around |
| 17:58 | <karlcow> | depends on the time |
| 17:58 | <TabAtkins> | Nah, that's just what Maps popped out when I asked it for a link. |
| 17:58 | <AryehGregor> | Ah, I see. |
| 17:58 | <AryehGregor> | When I was considering applying to Google for a full-time job, I of course made sure that the URLs in my resume were shortened using goo.gl instead of tinyurl. |
| 17:59 | <TabAtkins> | ...Why were you shortening your urls at all in your resume? |
| 17:59 | <AryehGregor> | I thought it was funny that I'd be using goo.gl to spy on how many people were reading the resume I submitted to Google, where and when. |
| 17:59 | <AryehGregor> | (I never actually submitted it, though.) |
| 17:59 | <AryehGregor> | Um, I dunno. Let me look. |
| 17:59 | <karlcow> | Nozomi is the fastest bullet train between Tokyo and Kyoto |
| 18:00 | <AryehGregor> | Wow, I so totally never use word processing software. |
| 18:00 | <karlcow> | http://en.wikipedia.org/wiki/Nozomi_(Shinkansen) |
| 18:00 | <AryehGregor> | This is like six months old and it was #8 on recent documents. |
| 18:00 | <AryehGregor> | Anyway, it turns out it was a link to my reflection tests, which were relevant since I was going to apply for web standards work. |
| 18:01 | <TabAtkins> | AryehGregor: Why shorten the link, though? Shortened links are for twitter only, mostly. |
| 18:01 | <AryehGregor> | They're for print also. |
| 18:01 | <karlcow> | The train is leaving from Shinagawa station |
| 18:01 | <karlcow> | so you will have to go from Haneda to Shinagawa. |
| 18:02 | <AryehGregor> | I was asked for a PDF resume, IIRC, so I assumed that someone might be printing it somewhere along the line. |
| 18:02 | <TabAtkins> | karlcow: Apparently Google is sending us on JR新幹線のぞみ99号 for the long part of the trip. |
| 18:02 | <TabAtkins> | But for some reason it's got a 5-minute section before that for 7k yen. ?_? |
| 18:02 | <karlcow> | JR is the company |
| 18:02 | <AryehGregor> | I'm proud to say I recognize the character の from that line. It's hiragana for "no", usually meaning "of". |
| 18:02 | <AryehGregor> | I think. |
| 18:02 | <TabAtkins> | Yeah, that's the shinkansen. |
| 18:02 | <AryehGregor> | That's like the only character I ever recognize when I read Japanese. |
| 18:02 | <AryehGregor> | Sometimes the characters for "Japan" too. |
| 18:02 | <AryehGregor> | Or "person", or like three other words. |
| 18:02 | <AryehGregor> | But mostly "of". |
| 18:03 | <karlcow> | or more exactly one of the companies of Japan for train |
| 18:04 | <AryehGregor> | Logogramies make me sad. I like trying to read foreign languages even if I don't understand them, but with CJK I have no chance. |
| 18:05 | <AryehGregor> | I know the Greek alphabet, and can do okay on Cyrillic, plus of course I can read Hebrew. I should learn the Arabic alphabet. |
| 18:05 | <karlcow> | TabAtkins: this might help you http://www.shinkansen.co.jp/jikoku_hyo/en/ |
| 18:06 | <AryehGregor> | Alternatively, I should go find something useful to do instead of spamming IRC chats. |
| 18:06 | <karlcow> | for the timetable |
| 18:07 | <karlcow> | and also the very practical http://www.hyperdia.com/ |
| 18:08 | <karlcow> | example http://www.hyperdia.com/cgi/en/search.html?dep_node=HANEDA%20AIRPORT%20INTERNATIONAL%20BUILDING&arv_node=KYOTO&via_node01=&via_node02=&via_node03=&year=2011&month=04&day=18&hour=13&minute=12&search_type=0&search_way=&transtime=undefined&sort=0&max_route=5&ship=off&lmlimit=null&search_target=route&facility=reserved&sum_target=7 |
| 18:10 | <TabAtkins> | Ooh, that one's very useful, karlcow. |
| 20:02 | <Hixie> | wow, http://www.google.com/green/ uses a lot of new html features |
| 20:02 | <Hixie> | <video>, <nav>, <footer>, all kinds of stuff |
| 20:09 | <AryehGregor> | <div id="learn-space" class="floatleft width-230"> |
| 20:09 | <AryehGregor> | <video> with no fallback, at that. |
| 20:10 | <AryehGregor> | Only WebM and MP4. |
| 20:10 | <AryehGregor> | No Flash or anything. |
| 20:10 | <AryehGregor> | Unless JS adds fallback. |
| 20:11 | <Hixie> | haven't tested it in IE |
| 20:11 | <Hixie> | the page certainly still has the influence of media-specific presentation :-( |
| 20:11 | <Hixie> | and suffers from a case of divitis |
| 20:13 | <zcorpan> | the fallback (visually at least) is <img> |
| 20:14 | <zcorpan> | seems like that page calls for background-image:url(webm) |
| 20:15 | <AryehGregor> | <div class="blue data-center-content width-230 colorblock"> |
| 20:15 | <AryehGregor> | Terrible alt text, too: <img src="img/how-can-we-help-in-the-face-of-disasters.jpg" width="230" height="161" alt="How Can We Help In The Face Of Disasters" class="block-bg"> |
| 20:15 | <AryehGregor> | Oh, no, it isn't. |
| 20:15 | <AryehGregor> | It's good alt text. |
| 20:15 | <AryehGregor> | Since the image is just a picture of text. |
| 20:15 | <AryehGregor> | I think. |
| 20:15 | <zcorpan> | no it isn't |
| 20:15 | <zcorpan> | it repeats the heading |
| 20:16 | <AryehGregor> | No, it's not. |
| 20:16 | <AryehGregor> | Yeah, it repeats the heading. |
| 20:17 | <zcorpan> | i think i'd just do <p><a href>Can competition become cooperation?</a></p> |
| 20:18 | <zcorpan> | and the background video should be done in css (except css doesn't support that) |
| 20:19 | <TabAtkins> | It totally should. |
| 20:19 | <zcorpan> | TabAtkins: fix css |
| 20:20 | <TabAtkins> | Suggest to me background-image:url(movie) should work, and I will. |
| 20:21 | <TabAtkins> | Right now, you can do background-image: element(#the-video); with an out-of-document video. |
| 20:21 | <zcorpan> | i guess we should implement that first |
| 20:21 | <TabAtkins> | (For some definition of "can do".) |
| 20:22 | <zcorpan> | background-image:url(movie) wouldn't work so well with the current codec situation in practice |
| 20:22 | <TabAtkins> | True. |
| 20:22 | <TabAtkins> | background-image: image("movie.webm", "movie.mp4"); |
| 20:23 | <zcorpan> | image("movie.webm") format("video/webm"), ... |
| 20:24 | <TabAtkins> | I need a word for "according to specs you can, but nobody implements it yet so it's currently theoretical only". |
| 20:24 | <TabAtkins> | spec-can |
| 20:24 | <zcorpan> | "I'm high as a kite" |
| 20:24 | <TabAtkins> | "Right now, you speccan do ..." |
| 20:30 | <webr3> | "soon" |
| 20:56 | <jgraham> | TabAtkins: If you weren't going to Japan on Google's dollar it might be worth your while to know that you can (or could 6 years ago) get some kind of foreigners-only tourist train ticket that makes travelling about more affordable, at least if you travel a lot |
| 20:56 | <TabAtkins> | Interesting. |
| 20:57 | <TabAtkins> | My wife and friend are coming with me and not being paid for by Google, so it might still be good to look into. |
| 20:57 | <jgraham> | http://www.japanrailpass.net/ |
| 20:59 | <jgraham> | Seems to be not valid on at least some Shinkansen |
| 20:59 | <jgraham> | I am a bit hazy on what we did because we went with a japanese friend so mostly we just followed her instructions |
| 21:00 | <TabAtkins> | Seems like it wouldn't actually save me anything if we're just doing a single roundtrip between tokyo and kyoto. |
| 21:01 | <jgraham> | No, I doubt it in that case. We travelled quite a bit more than that (although we did do that as part of the trip) |
| 21:01 | <TabAtkins> | Yeah, seems like a great deal if you'll be taking the train a few times. |
| 21:02 | jgraham | wonders if anyone noticed the irony of using relatively-cpu-intensive video on a page about saving energy |
| 21:03 | <AryehGregor> | It's amazing how ironic environmentalism can often be. |
| 22:11 | <AryehGregor> | Hmm, ISSUE-80 decision is in. |
| 22:11 | AryehGregor | reads |
| 22:38 | <Hixie> | oh hey |
| 22:38 | <Hixie> | we're on xkcd today! |
| 22:43 | <aho> | oh... boy :> |
| 22:45 | <mpilgrim> | i see a new XKCD poster in my future |
| 22:49 | <tndH> | all that work, just to get to that last line... |
| 22:52 | <TabAtkins> | I like reading it as an actual future history. |
| 22:53 | <Hixie> | the temperature gradient in 2090 sounds painful |
| 22:53 | <Hixie> | also i'm taking my money out of dollars if that history is accurate |
| 22:54 | <TabAtkins> | What, the debt being 700% of GDP scares you? |
| 22:54 | <Hixie> | TabAtkins: no, the debt going up and down like a yoyo |
| 22:54 | <TabAtkins> | Heh, yeah. |
| 22:55 | <AryehGregor> | Is there actually such a thing as a blind photographer? Really? |
| 22:55 | <AryehGregor> | I mean . . . how? |
| 22:55 | <Hixie> | TV is blind and is a photographer |
| 22:55 | <Hixie> | TV Raman, i mean |
| 22:55 | <AryehGregor> | TV? |
| 22:55 | <AryehGregor> | Oh. |
| 22:55 | <AryehGregor> | Interesting. |
| 22:55 | <Hixie> | as i recall, he likes taking photos of things so his friends can describe where he was |
| 22:55 | <Hixie> | tohim |
| 22:56 | <Hixie> | later |
| 22:56 | <AryehGregor> | Makes sense. |
| 22:56 | <Hixie> | which seems like a pretty reasonable thing to me |
| 22:56 | <AryehGregor> | "Photographer" makes it sound like it's a professional thing. |
| 22:56 | <Hixie> | (and i just love the concept that he'd be required to include alt text to save himself from himself...) |
| 22:56 | <TabAtkins> | I think my favorite year is 2068. |
| 22:57 | <AryehGregor> | 2068 is pretty good, yeah. |
| 22:57 | <AryehGregor> | Lord Jesus ruling an entirely gay Earth. |
| 22:57 | <TabAtkins> | And then, apparently, legalizing public masturbation the next year. |
| 22:57 | <Hixie> | also the rfc number for http :-P |
| 22:57 | <AryehGregor> | Is it just me, or do climatologists really like the year 2100? |
| 22:58 | <Hixie> | round number |
| 22:58 | <Hixie> | everyone likes 2100 |
| 22:58 | <TabAtkins> | It's a good "medium future" number. |
| 22:58 | <Hixie> | it's like 2000 was a few decades ago |
| 22:58 | <aho> | it's a good "i'll be dead then LOL" date |
| 22:58 | <AryehGregor> | 2100 is "medium" future? |
| 22:58 | <AryehGregor> | I guess you're a sci-fi fan. |
| 22:58 | <Hixie> | i was just gonna say, 2100 sounds near-future to me :-P |
| 22:59 | <AryehGregor> | To me, 2100 is "so far in the future that I'm not really confident making any kind of prediction about it whatsoever" future. |
| 22:59 | <TabAtkins> | Depends on how singulatarian I'm feeling. |
| 22:59 | <tw2113> | AryehGregor in 2100, IE6 will still be installed on at least 1 computer |
| 23:00 | <othermaciej> | by 2100, most of us will be dead, unless there are major unforeseen advances in life extension technology |
| 23:00 | <TabAtkins> | I'm betting on mind virtualization by then. |
| 23:00 | <tw2113> | now to get better minds |
| 23:00 | <othermaciej> | a future with that level of advancement is almost by definition not predictable |
| 23:00 | <AryehGregor> | I'd be 112. That doesn't seem so unattainable, given the advances in biotech we've had in the last few decades. |
| 23:01 | <othermaciej> | I mean, megascale geo-engineering is probably easier than either major life extension or mind uploads |
| 23:01 | <AryehGregor> | I mean, open-heart surgery was first performed in 1952. |
| 23:01 | <TabAtkins> | othermaciej: Indeed, thus my comment. If I pretend that the singularity won't occur sometime in the next century, then 2100 is medium-future. |
| 23:01 | <Hixie> | AryehGregor: i that _successful_ open heart surgery, or any? ;-) |
| 23:01 | <AryehGregor> | Successful. |
| 23:01 | <tw2113> | too much of me fears repeating of the movie Idiocracy |
| 23:01 | <othermaciej> | 2100 is either far-future or you won't live to see your predictions about it |
| 23:01 | <TabAtkins> | The first reasonably successful one. Dude lived 3 days with his new heart. |
| 23:02 | <AryehGregor> | othermaciej, insightful point. |
| 23:02 | <TabAtkins> | othermaciej: That seems incorrect. Reasonable projections of biological life-extension can let you reach 2100 without the world changing to an unpredictable degree. |
| 23:02 | <AryehGregor> | I doubt we'll see any "singularity" in the sense of asymptotically rapid progress. |
| 23:03 | <AryehGregor> | It's possible, but I don't see any reason to believe it. |
| 23:03 | <TabAtkins> | Agreed; I'm referring more to virtualization becoming practical. |
| 23:03 | <AryehGregor> | It's not clear to me how likely that is. |
| 23:03 | <AryehGregor> | Because we'd have extremely high standards to consider it successful. |
| 23:04 | <TabAtkins> | I think it's an engineering problem at this point, and a solveable one. |
| 23:04 | <AryehGregor> | We'd consider an artificial heart successful as long as it can replicate the essential function of a heart, even if it does so very differently in practice. |
| 23:04 | <TabAtkins> | Just need a few more moore's law iterations, plus sufficient scanning-tech advances. |
| 23:04 | <othermaciej> | TabAtkins: life expectancy is not rising at better than a year per yer |
| 23:05 | <Hixie> | AryehGregor: hey you just described a "successful" heart operation as something that extends life by three days, according to Tab, so why would you have a higher bar for this? ;-) |
| 23:05 | <AryehGregor> | But if electronic copies of human brains have even moderately large personality changes or similar, we'd be unlikely to consider the subject really "the same person". |
| 23:05 | <othermaciej> | (I don't know if it is rising much at all in developed countries any more) |
| 23:05 | <TabAtkins> | Hixie: Once you've established that 3 days is possible, the rest is engineering. |
| 23:05 | <Hixie> | TabAtkins: sounds like the same applies here :-) |
| 23:05 | <TabAtkins> | othermaciej: I suspect we're on the rising cusp of biotech advancement, though. |
| 23:05 | <TabAtkins> | We'll get a nice sigmoid going on in the near future. |
| 23:06 | <Hixie> | watching TED talks, we're clearly on the rising cusp of biotech advancement |
| 23:06 | <AryehGregor> | othermaciej, current life expectancy rise is driven by incremental improvement in medicine. I'd say that if we see a really huge increase in lifespan within the next century, it will be because artificial organs become much cheaper and safer. |
| 23:06 | <othermaciej> | so you're postulating currently unknown dramatic technological improvement |
| 23:06 | <othermaciej> | we've been on the rising cusp of biotech advancement for a good 15 years now |
| 23:06 | <Hixie> | indeed |
| 23:06 | <TabAtkins> | Yes? |
| 23:06 | <AryehGregor> | At some point, it would be reasonable to do things like preemptively replace largely healthy hearts to avoid possible heart failure. |
| 23:06 | <TabAtkins> | And we've been 10 years away from AI for 50 years. ^_^ |
| 23:06 | <TabAtkins> | We're still closer to AI now than we were half a centure ago. |
| 23:06 | <AryehGregor> | And things like that. |
| 23:06 | <zcorpan> | maybe if i want to live longer, i should get some sleep |
| 23:06 | <zcorpan> | nn |
| 23:07 | <AryehGregor> | If you can replace failing organs cheaply and reliably, you could remove most common causes of natural death, other than brain failure. |
| 23:07 | <TabAtkins> | Yup. |
| 23:07 | <AryehGregor> | Alternatively, one can imagine a brain-in-a-vat scenario, where other organs become irrelevant. |
| 23:08 | <TabAtkins> | That seems less likely to me, personally, for social reasons if nothing else. |
| 23:08 | <AryehGregor> | Of course, brains fail too, in terms of senility and so on, but I could easily see a large lifespan increase along these lines within the next century. |
| 23:08 | <othermaciej> | you can imagine all sorts of things that are currently infeasible |
| 23:08 | <AryehGregor> | Sure. |
| 23:08 | <othermaciej> | if you assume such things, then claims of predicting the future beyond them are not credible |
| 23:09 | <AryehGregor> | Yes, that much I agree with. I don't trust any predictions beyond twenty years or so, unless they're very vague. |
| 23:09 | <AryehGregor> | Well, about society and such, I mean. |
| 23:09 | <AryehGregor> | I trust that Halley's comet will return in 2061. |
| 23:10 | <othermaciej> | unless grey goo nanobots convert the whole solar system to paper clips |
| 23:11 | <AryehGregor> | I'm pretty sure the worst-case scenario for nanobots doesn't include space travel. |
| 23:11 | <AryehGregor> | I do sort of wonder how likely it is that we'll destroy the world somehow. |
| 23:11 | <AryehGregor> | I'm betting not very, but who knows. |
| 23:12 | <TabAtkins> | Gray goo has fundamental problems with energy, I think. Molecular conversion is a pretty expensive process. |
| 23:12 | <AryehGregor> | Clearly, if there's any non-negligible chance of some endeavor destroying the world, we should avoid that endeavor at all costs. |
| 23:12 | <TabAtkins> | Maybe you can gray-goo a chip factory, but it would more or less stop when it hits concrete. |
| 23:12 | <AryehGregor> | Well, gray goo is a pretty unlikely scenario, yeah. |
| 23:13 | <AryehGregor> | Nanobots are also likely to be fragile. |
| 23:13 | <TabAtkins> | Yeah, one good EMP (or hell, a microwave burst) should fry them. |
| 23:13 | <AryehGregor> | Zap them with some EMP or a flamethrower or something, that should take care of them. |
| 23:13 | <AryehGregor> | They're a bit too small for shielding. |
| 23:13 | <AryehGregor> | So I don't think it's really worth worrying about. |
| 23:13 | <AryehGregor> | Strong AI is more worrying. |
| 23:13 | <jcranmer> | you assume they use electronic gates |
| 23:13 | <TabAtkins> | On the subject of runaway replicators, robo-miners are more credible. |
| 23:13 | <AryehGregor> | But we don't have to worry about that anytime soon. |
| 23:14 | <TabAtkins> | But that would just exhaust our asteroid belt or something. |
| 23:14 | <TabAtkins> | jcranmer: Flamethrower or microwave weaponry doesn't assume that. |
| 23:14 | <jcranmer> | EMP does |
| 23:15 | <jcranmer> | flamethrower assumes easy breakability |
| 23:15 | <TabAtkins> | No, it assumes that nanoscale machinery isn't immune to heat. |
| 23:15 | <jcranmer> | especially because they likely would design nanomachines to have good heat dissipation |
| 23:15 | <TabAtkins> | Yes, good heat dissipation for their own processes. |
| 23:15 | <TabAtkins> | The fact that I have a good heatsink on my CPU doesn't mean it can withstand a flamethrower. |
| 23:16 | <AryehGregor> | Nanomachines will automatically have a practically infinite ratio of surface area to volume. |
| 23:16 | <AryehGregor> | So they'll always be at the temperature of their ambient environment. |
| 23:16 | <AryehGregor> | But that doesn't help them if their ambient environment is 1000°F. |
| 23:16 | <AryehGregor> | TabAtkins, your CPU's heat dissipation is a joke compared to anything nanoscale. |
| 23:17 | <AryehGregor> | But it doesn't matter. |
| 23:17 | <AryehGregor> | Dissipation is only relevant for heat produced by the thing itself, not environmental heat. |
| 23:17 | <AryehGregor> | Nothing's going to stand in the way of high environmental temperature. |
| 23:18 | <AryehGregor> | Unless someone invents a good thermal insulator. Now *that* would have major engineering applications. |
| 23:18 | <TabAtkins> | AryehGregor: Of course, but my point was that dissipation is irrelevant in the presence of extreme heat, as you point out. |
| 23:18 | <AryehGregor> | It's not a question of extreme or not extreme, it's internal versus external. |
| 23:18 | <AryehGregor> | If you have good enough heat dissipation, you can tolerate arbitrarily large amounts of internal heat, by distributing it to your environment. |
| 23:18 | <TabAtkins> | Extreme in a relative sense. "Much hotter than operating temperature bounds". |
| 23:19 | <AryehGregor> | If your CPU had perfect heat dissipation, it could produce a megawatt of heat indefinitely with no ill effects. |
| 23:19 | <AryehGregor> | But it would still melt if the environment around it were too hot. |
| 23:19 | <TabAtkins> | You're going out of your way to interpret my statements in isolation, rather than in the obvious context of a flamethrower being applied. |
| 23:20 | <AryehGregor> | I'm just saying, flamethrowers are an external heat source, so dissipation is irrelevant. The place that the heat is being dissipated to is also being heated up, so the dissipation doesn't work. |
| 23:20 | <AryehGregor> | You need a temperature difference to dissipate heat. |
| 23:21 | <Hixie> | if you're putting a megawatt of heat into your environment, your environment will quickly get "too hot" :-P |
| 23:21 | <TabAtkins> | I'm perfectly aware of this, and it's precisely what I said originally. ^_^ |
| 23:21 | <AryehGregor> | Hixie, yeah, that's a detail. |
| 23:21 | <Hixie> | just sayin' ;-) |
| 23:21 | <TabAtkins> | Hixie: Not if your heat dump is a laser and you're in vacuum, for example. |
| 23:21 | <AryehGregor> | Would that work, thermodynamically? |
| 23:21 | <Hixie> | i thought we were talking about robots in a factory |
| 23:21 | <Hixie> | sorry, not paying that close attention here |
| 23:21 | <AryehGregor> | Well, you could always just radiate the heat, if you were in a vacuum. |
| 23:22 | <TabAtkins> | I'm not sure how heat-dissipation lasers work, but I've seen them enough in scifi that I assume they're possible. |
| 23:22 | <AryehGregor> | We *were* talking about nanorobots, but that was a while ago. |
| 23:22 | <TabAtkins> | If you're just radiating, you're stuck with blackbody efficiency at best. |
| 23:22 | <AryehGregor> | Yes, that's true. |
| 23:22 | <AryehGregor> | But I'm pretty sure you can't just radiate heat away using a laser. |
| 23:23 | <AryehGregor> | I mean, how much entropy is in a laser beam? |
| 23:23 | <TabAtkins> | I'd have to actually do some research to see if it's plausible. Like I said, I've just seen it in enough scifi, particularly hard stuff, that I assume it works somehow. |
| 23:23 | <AryehGregor> | Where's the entropy going? |
| 23:23 | <jcranmer> | laser beams are pretty much low entropy by definition |
| 23:23 | <AryehGregor> | Laser beams can contain lots of energy, but I don't see where the entropy is. |
| 23:23 | <jcranmer> | they're light amplified by having all the waves in the same direction |
| 23:23 | <AryehGregor> | And you need to dump the entropy someplace. |
| 23:23 | <AryehGregor> | Right, that's what I'd think. |
| 23:24 | <AryehGregor> | You'll create entropy if the laser hits something and heats it up, but that doesn't help you. |
| 23:24 | <AryehGregor> | You need to create it in the process of generating the laser. |
| 23:24 | <jcranmer> | a laser beam you can see is bad |
| 23:24 | <jcranmer> | the laser equipment is what gets hot |
| 23:24 | <AryehGregor> | Yeah, creating the laser will create lots of extra entropy. Surely more than any you lose by it. |
| 23:25 | <AryehGregor> | You can sink as much heat as necessary by radiation if you have enough surface area. |
| 23:25 | Philip` | thinks we ought to design computers to be entropy-efficient rather than just energy-efficient, since delaying the death of the universe seems a pretty important long-term goal |
| 23:25 | <AryehGregor> | If there were such a thing as a good thermal conductor, that would be trivial. |
| 23:25 | <AryehGregor> | But there are no good thermal insulators or conductors. |
| 23:34 | <AryehGregor> | Wow, top-posting is especially obnoxious when a) the poster starts a new thread so it's not auto-collapsed in Gmail, and b) the quoted post is like ten pages long. |
| 23:34 | <TabAtkins> | Faulkner is basically horrible at writing emails in every way. |
| 23:34 | <TabAtkins> | I shudder to think how his emails sound in a screen reader. |
| 23:36 | <AryehGregor> | Finally! https://bugs.webkit.org/show_bug.cgi?id=57942 |
| 23:36 | <AryehGregor> | Some browser should have done this a couple of years ago, when minification was becoming common. |
| 23:36 | <TabAtkins> | Yay! |
| 23:36 | <TabAtkins> | <3 Inspector Team. |
| 23:39 | <AryehGregor> | Asa Dotzler is . . . pretty concerned about privacy, isn't he? |
| 23:40 | <othermaciej> | Asa Dotzler is concerned about anything relating to non-Firefox browsers |
| 23:43 | <AryehGregor> | Currently he seems to be concerned mostly about search engines. |
| 23:43 | <AryehGregor> | At least judging by his blog. |
| 23:43 | <AryehGregor> | To the extent of repeatedly and enthusiastically endorsing Bing over Google, because of their allegedly superior privacy policy. |
| 23:45 | <AryehGregor> | Am I the only one who thinks Do-Not-Track is a complete waste of time? Advertisers know that the majority of their users don't want to be tracked if you ask them in the abstract without specifying any consequences for saying no, so what does DNT actually change? |
| 23:45 | <AryehGregor> | Of course if you ask people whether they'd like to be tracked, all else being equal, they'll say no. |
| 23:46 | <zewt> | i thought the main point was to act as an explicit "opt out" for purposes of opt-out legislation, but given that the spec allows browsers to send it by default, that seems out the window |
| 23:46 | <AryehGregor> | Yes, that's the killer problem. The site doesn't know if the user has actually opted out, or if something has opted out for them. |
| 23:46 | <AryehGregor> | If it's actually honored, every random piece of software out there is going to helpfully set it off. Why not? |
| 23:47 | <zewt> | well, it's not even *trying* since the spec explicitly allows UAs to send DNT: 1 as the default |
| 23:47 | <AryehGregor> | User tracking is a tradeoff of privacy versus utility, it can't be addressed by posing a one-sided question. |
| 23:47 | <zewt> | which is basically expending bits in every HTTP request to say "people don't like being spied on", which is just wasted bytes |
| 23:47 | <AryehGregor> | Pretty much. |
| 23:48 | <zewt> | may as well put a post-it on your car window, "do not smash" |
| 23:48 | <zewt> | DNS: 1 |
| 23:49 | <AryehGregor> | The real questions you have to ask are: Are you, a user, willing to use this site even though it might track you? And are you, as a website provider, willing to provide service to these users even though you can't track them? |
| 23:49 | <AryehGregor> | Those are the tradeoffs people face. |
| 23:50 | <AryehGregor> | In practice, websites use tracking that's hard to evade because it uses the same mechanism as things that users actually want, namely cookies. |
| 23:50 | <AryehGregor> | And similar things. |
| 23:50 | <AryehGregor> | Not to mention entirely server-side tracking by IP address or other fingerprinting. |
| 23:50 | <kbrosnan> | if people knew what kind of data ad/marketing agencies had on them i doubt many would be happy with the picture it paints of them |
| 23:51 | <AryehGregor> | kbrosnan, of course not. But sites track users because they need to generate revenue to provide a free or lower-cost service, and tracking helps generate revenue. |
| 23:52 | <AryehGregor> | If sites have less revenue, users suffer as a result. |
| 23:52 | <AryehGregor> | You could even go so far to say that users who deliberately try to generate less revenue, such as by opting out of tracking, are leeching off other users. |
| 23:53 | <AryehGregor> | They're refusing to compromise on their personal preferences while profiting from the fact that others are willing to compromise. |
| 23:54 | <AryehGregor> | I wouldn't go that far, but portraying tracking as evil certainly makes no sense to me. |
| 23:54 | <AryehGregor> | Or arguing that users should have some right to be exempt from it, while they're using services funded partly by tracking. |
| 23:55 | <zewt> | tracking information about me, without telling me you're doing it, what you're storing or giving me any way to erase your data--that's a blatant violation of privacy by a whole lot of people's measures |
| 23:56 | <kbrosnan> | maybe it finaly forces a pay for content model. guess we'll see where nytimes is in several months / couple years |
| 23:56 | <AryehGregor> | Not by mine. If I walk into a store, I expect that I might be videotaped and that information might be saved indefinitely. |
| 23:56 | <AryehGregor> | Anytime I provide information to someone else, I have no particular expectation that they won't save it and use it somehow. |
| 23:56 | <AryehGregor> | If I don't want them to, I should refrain from providing the information. |
| 23:57 | <zewt> | by wearing a bag over your head when you walk into the store? |
| 23:57 | <AryehGregor> | If I really wanted to, I'd have to wear a mask or something, yeah. Fortunately, I don't care. |
| 23:57 | <kbrosnan> | currently your only option is to stay out of the web completely and live a cash lifestyle |
| 23:57 | <AryehGregor> | I also have to go now. |
| 23:57 | <AryehGregor> | Bye. |