| 01:41 | <boblet> | is there a phrase for someone who’s all like “your site doesn’t validate so you’re obviously an idiot”? |
| 01:43 | <ako> | depends on the severity of the errors |
| 01:43 | <boblet> | I think this was what I was getting at: http://meiert.com/en/blog/20081120/pseudo-standardistas/ |
| 01:44 | <ako> | e.g. if it's something that will most likely cause different DOM trees in different browsers, then he's right :> |
| 01:44 | <boblet> | ako: 6, of which 2 are legit (1 in a comment, 3 from plugins) |
| 01:44 | <ako> | if it's an unescaped & in some url... well... :> |
| 01:44 | <boblet> | and none of the errors cause different DOM trees |
| 01:45 | <ako> | in a... comment? how does that work? |
| 01:46 | <boblet> | apparently allowed HTML subset rules don’t apply to authors 9.9 |
| 01:46 | <boblet> | (fixed :) |
| 01:46 | <deltab> | ako: unescaped ampersands do cause real problems |
| 01:47 | <ako> | like? |
| 01:48 | <ako> | &s inside urls aren't escaped on most sites (or at least many sites) |
| 01:48 | <deltab> | I know of one popular forum package that turns &curr into ¤, which breaks urls containing "foo=123123¤t=3234" |
| 01:48 | <ako> | heh |
| 01:48 | <deltab> | which is a pattern used by one of the larger photo hosting sites |
| 01:50 | <ako> | well, it's still less severe than a garbled layout |
| 01:50 | <ako> | well, sort of |
| 01:51 | <deltab> | it's caused thousands of broken links on one site alone |
| 01:51 | <ako> | ok here is another exampe... bullcrap attributes |
| 01:51 | <deltab> | likewise &lang can be turned into a left angle bracket |
| 01:52 | <deltab> | in comments though, it doesn't matter; only -- matters there |
| 01:53 | <ako> | e.g. at work i can't really use a regular validator because that .net webforms stuff together with those f-ing retarded 3rd party controls generates pure junk markup |
| 01:53 | <ako> | i can only check if it's at least sorta well formed |
| 01:54 | <ako> | (my "validator" uses very sloppy rules) |
| 01:54 | <deltab> | the questions that really matter are "Have I made a mistake?" and "Will browsers understand this?" |
| 01:55 | <ako> | yes, hence the sloppy validator it can only catch things which will most likely break things in horrible ways |
| 01:55 | <deltab> | an SGML validator such as the one at validator.w3.org is poor at answering either question |
| 01:55 | <deltab> | validator.nu is better |
| 01:56 | <ako> | e.g. it will complain about <a><b></a></b> or if a tag isn't closed... but it uses sloppy rules for <br/>, <img/> or <p></p> for example |
| 01:57 | <ako> | well, in theory one could run automated tests... where the generated dom is sent back to the server |
| 01:58 | <ako> | but that would be quite a lot of work |
| 01:58 | <deltab> | you could set that up with Selenium |
| 01:59 | <ako> | sill... you'd need a bunch of machines for that crap |
| 02:00 | <ako> | oh well... nn |
| 02:00 | <ako> | <; |
| 02:02 | <AryehGregor> | I haven't played NetHack in months, and only beat it like three times, but once in a while I still see something like "nn" and my first reaction is "two nymphs standing next to each other, watch out!" |
| 02:20 | <estellevw> | Can anyone tell me if the spellcheck attribute i supported in any browser and how i can test it? |
| 02:21 | <estellevw> | i supported = is supported |
| 02:26 | <Aleoss> | estellevw: It should be supported in Chrome and FF for sure. |
| 02:27 | <estellevw> | other than the spec, i am not finding any documentation |
| 02:27 | <Aleoss> | estellevw: The only way to check it is test it out. Or unless someone already did that and did a report on a blog that can be found via Google. |
| 02:28 | <Aleoss> | estellevw: http://en.wikipedia.org/wiki/Comparison_of_layout_engines_%28HTML5%29#Global |
| 02:28 | <Aleoss> | estellevw: Learn to Google. |
| 02:28 | <estellevw> | I do google |
| 02:28 | <estellevw> | learn to be less arrogant |
| 02:29 | <Aleoss> | estellevw: You said all you could find is the spec. |
| 02:30 | <Aleoss> | estellevw: As I hypothesised. FF supports it and Chrome supports it but not to the spec standards. |
| 02:30 | <estellevw> | right. |
| 05:35 | MikeSmith | reads http://weblogs.mozillazine.org/roc/archives/2010/05/nero_vs_mpegla.html |
| 05:43 | <MikeSmith> | jwm: are you a developer? |
| 05:44 | <jwm> | more of a follower of technology |
| 05:44 | <jwm> | I procrastinate too much to get things done hehe |
| 05:44 | <jwm> | been working 13 years almost to get a network platform going |
| 05:44 | <jwm> | I guess I've been waiting for the right platform :) |
| 05:45 | <jwm> | didn't imagine the web would end up to potentially be that |
| 05:46 | <MikeSmith> | the perfect storm |
| 05:47 | <jwm> | yeah |
| 05:47 | <jwm> | now I am excited about the peer to peer provisions in the new spec |
| 05:47 | <jwm> | I'd like to do a social network and also distributed.net/folding@home esque payload system |
| 05:48 | <jwm> | and even silly file sharing network |
| 05:48 | <jwm> | whatever p2p can happen :) |
| 05:48 | MikeSmith | takes a look at http://dev.dist.us/ |
| 05:48 | <jwm> | hah |
| 05:48 | <jwm> | yeah |
| 05:48 | <jwm> | it's sad in 13 years I haven't gotten anything productive done |
| 05:49 | <jwm> | http://dev.dist.us/net/docs/ |
| 05:49 | <jwm> | you can see where I stole the cool physics code from opera project you worked at |
| 05:49 | <jwm> | hehe |
| 05:49 | <jwm> | their 10.50 launch page |
| 05:49 | <jwm> | it works faster in chrome than opera :) |
| 06:32 | <MikeSmith> | boblet: sent you a reply |
| 06:32 | <MikeSmith> | lemme know if it doesn't make any sense |
| 06:32 | <MikeSmith> | jwm: so what the hell have you been doing instead for the last 13 years? |
| 06:32 | <MikeSmith> | picking daisies? |
| 06:32 | <jwm> | might as well say that |
| 06:33 | <jwm> | been reading news, thinking up ideas |
| 06:33 | <jwm> | keeping my mind broad heh |
| 06:33 | <jwm> | redesigning my site every other week |
| 06:33 | <jwm> | working at a stupid telecom company |
| 06:33 | <jwm> | I'm one of the mom's basement non-success stories |
| 06:33 | <jwm> | the only thing I have going for me is I have cool 4 letter domain names |
| 06:34 | <jwm> | because I managed to register them before school zones did (dist domains were used for distRICT) heh |
| 06:34 | <jwm> | except a domain squatter did beat me to dist.com |
| 06:34 | <jwm> | by 4 days |
| 06:35 | <jwm> | I've followed AI developments also |
| 06:36 | <jwm> | I'm a big fan and big proponent of google at the same time hehe |
| 06:36 | <jwm> | er opponent |
| 06:37 | <MikeSmith> | nessy: http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2010-May/026456.html is well put |
| 06:38 | <MikeSmith> | maybe even FAQ-worthy |
| 06:38 | <MikeSmith> | though I don't know which FAQ would be the right place for it |
| 06:39 | <MikeSmith> | nessy: in particular, what you say about lessons learned from the last 10 years |
| 06:40 | <MikeSmith> | jwm: well, the next 13 years are going to make the last 13 look boring |
| 06:40 | <jwm> | I know MikeSmith |
| 06:40 | <jwm> | especially if peer to peer unfolds |
| 06:40 | <jwm> | I think that has the potential to unseat google |
| 06:40 | <jwm> | heck any ISP |
| 06:40 | <jwm> | heh |
| 06:41 | <roc> | if software patents and locked-down platforms continue their meteoric rise, the next 13 year are going to be very boring indeed |
| 06:41 | <jwm> | if the stupid ISPs would stop throttling outbound speeds |
| 06:41 | <jwm> | that's why peer to peer is so nice roc |
| 06:41 | <roc> | P2P doesn't help with those issues at all |
| 06:41 | <jwm> | create your own protected from government space |
| 06:41 | <MikeSmith> | roc: I think we need for things to get worse before they get better |
| 06:41 | <MikeSmith> | (as far as patents go) |
| 06:41 | <jwm> | have you looked at freenet? |
| 06:41 | <jwm> | it helps |
| 06:42 | <jwm> | there are a lot of projects to help people living behind firewalls heh |
| 06:42 | <jwm> | even 'civilized countries' are creating bad information policies |
| 06:46 | <roc> | a future where innovation is restricted to a few outlaws on darknets isn't a very innovative one |
| 06:46 | <jwm> | that's if the peer to peer is done using freenet's ways |
| 06:47 | <jwm> | not web tech which is more expansive and will be harder to block without blocking legitimate web traffic |
| 08:19 | <MikeSmith> | me just now finds http://www.robglidden.com/2010/05/how-googles-open-sourcing-of-vp8-harms-the-open-web/ |
| 08:19 | <jwm> | ehe |
| 08:19 | <jwm> | how opening a protocol harms a set of open protocol :) |
| 08:19 | <jwm> | s |
| 08:21 | <MikeSmith> | "leaving VP8 code out in the open with nothing but a mutual non-assert license leaves the patent issue not only unaddressed, but up for capture by those with uncharitable agendas, and on their turf and time frame" |
| 08:22 | <MikeSmith> | "Contributing VP8 to a standards group with a strong patent disclosure policy would be a good corrective move; it would force lurking patent holders to come fully into the public" |
| 08:23 | <jwm> | yeah |
| 08:23 | <jwm> | I completely agree |
| 08:23 | <jwm> | it has come about very weirdly |
| 08:23 | <jwm> | and I do think maybe google is still trying to figure it out |
| 08:24 | <othermaciej> | I did suggest to Google that VP8 should be submitted to a standards group with disclosure rules and relevant patent holders present |
| 08:24 | <jwm> | I mean vp8 seemed to come from left field when nobody was watching |
| 08:24 | <othermaciej> | they seemed surprisingly uninterested in taking it to a standards body |
| 08:25 | <jwm> | well |
| 08:25 | <othermaciej> | given the spec they have now, a second independent implementation would be almost impossible |
| 08:25 | <jwm> | there may be a better reason for that than they are afraid of it being vulnerable to other patents |
| 08:25 | <jwm> | it may be because h264 has so much steam they want to get out the door real fast |
| 08:27 | <MikeSmith> | othermaciej: have they actually made the spec (such as it is) publicly available yet? |
| 08:28 | <MikeSmith> | nessy: have you read http://www.robglidden.com/2010/05/how-googles-open-sourcing-of-vp8-harms-the-open-web/ yet? |
| 08:28 | <nessy> | nope, doing now |
| 08:29 | <othermaciej> | MikeSmith: yes |
| 08:29 | <othermaciej> | MikeSmith: http://static.googleusercontent.com/external_content/untrusted_dlcp/www.webmproject.org/en/us/media/pdf/vp8_bitstream.pdf |
| 08:29 | <MikeSmith> | ah yeah |
| 08:29 | <othermaciej> | many things are spec'd only as C code though |
| 08:30 | <nessy> | MikeSmith: Rob has never understood that open source communities build de-facto standards, too |
| 08:31 | <jwm> | well he mentioned that he thinks it should be open for the patent holders to look at too |
| 08:31 | <jwm> | like lawyer friendly docs |
| 08:31 | <jwm> | :) |
| 08:31 | <nessy> | he was trying to develop the open video standard format while in Sun |
| 08:31 | <nessy> | trying to go down the same path that Google has taken a lot more successfully now |
| 08:32 | <nessy> | it's amazing what you can do with the right funding ;) |
| 08:32 | <jwm> | hehe |
| 08:32 | <jwm> | problem is now google is getting so big that it can't keep track of what it is doing |
| 08:33 | <jwm> | I read an article about some engineers that work at google that didn't know 2.2 android was being released |
| 08:33 | <jwm> | kind of shocked me |
| 08:33 | <MikeSmith> | othermaciej: the "spec'd only as C code" thing seems to have generated quite a bit of ridicule already |
| 08:34 | <nessy> | even submitting a codec to a standards group doesn't get the lurking patent holders out of the thicked |
| 08:34 | <MikeSmith> | jwm: android engineers? |
| 08:34 | <MikeSmith> | nessy: true |
| 08:34 | <jwm> | MikeSmith: not quite android engineers but not too outside the mobile division |
| 08:34 | <MikeSmith> | but it certainly raises the bar some |
| 08:35 | <jwm> | you'd think maybe they'd of gotten a hey froto is released email or something :) |
| 08:35 | <othermaciej> | I think it should be taken to a standards group not just because of IPR issues, but because, if it's supposed to become a critical part of the open web, it should be specified by a group where all stakeholders can participate, and specified in a way that multiple independent implementations are possible |
| 08:35 | <othermaciej> | anything that has only a single viable implementation is not an open standard |
| 08:36 | <MikeSmith> | jwm: not sure what "not too outside the mobile division" means, but it doesn't seem terrifically surprising that engineers working on other projects didn't all know about the release |
| 08:36 | <jwm> | they should of developed vp8 in amaya first :) |
| 08:36 | <MikeSmith> | othermaciej: if it ends up being that it has only a single viable implementation, that would seem to be true |
| 08:38 | <MikeSmith> | difficult to see how it's going to get to multiple implementations without actually speccing it properly |
| 08:38 | <nessy> | othermaciej: are you quoting or is that your statement? |
| 08:38 | <MikeSmith> | I wonder if multiple implementations is an actual goal or not |
| 08:38 | <othermaciej> | nessy: that's my statement |
| 08:39 | <nessy> | I agree with the need for an independent implementation |
| 08:39 | <nessy> | I would think Google would encourage that, too |
| 08:39 | <othermaciej> | (and not necessarily the opinion of my employer, any open source projects I am involved with, or any standards bodies I am involved with) |
| 08:39 | <MikeSmith> | boblet: I think this guy needs some assistance: http://stackoverflow.com/questions/2895214/what-is-the-semantically-correct-way-to-use-the-article-tag-in-html-5-with |
| 08:40 | <boblet> | MikeSmith: thanks for the email. will check SF out |
| 08:40 | <boblet> | woops SO |
| 08:41 | <MikeSmith> | nessy: I guess it's one thing to encourage in principle, another thing to actually do the real work in practice to make sure it happens |
| 08:41 | <nessy> | I don't think though that a standard necessarily needs to be defined by a group of stakeholders - in particular something as complex as a video codec will take ages to create in a group - MPEG seems to be working on one such, so maybe that can become the successor to WebMM |
| 08:42 | <nessy> | I think for now, WebM can fill the need for a baseline codec and is independent enough from Google to be acceptable |
| 08:43 | <othermaciej> | what's the advantage of not taking WebM to a standards body? |
| 08:43 | <othermaciej> | ok, there is the resource cost, but clearly for Google that would be pocket change |
| 08:44 | <jwm> | isn't webm just for the container which is matroska? |
| 08:44 | <nessy> | the advantage that WebM has on its side compared to Theora (apart from better quality) is that there is a patent gorilla behind it to defend it *should* a patent suit happen - plus they threw their engineering force at it to build the software patches to provide application support - plus they threw their relationships at it to get other commercial entities behind it, too |
| 08:44 | <nessy> | jwm: if so, s/WebM/VP8/ :-) |
| 08:45 | <nessy> | othermaciej: I am all for taking it to a standards body, too |
| 08:45 | <annevk> | from what I heard Google was interested in taking it to a standards body at some point |
| 08:45 | <othermaciej> | webm is both the container and (I think) the collective name for the container/codec suite |
| 08:45 | <jwm> | I think vp8 is just an interrim project |
| 08:46 | <nessy> | I was probably one of the first to say that is a good idea even before it all happened - but it will take a long time and in the meantime VP8 can grow in support around an open source community, like many other technologies |
| 08:46 | <jwm> | http://www.matroska.org/news/webm-matroska.html |
| 08:47 | <nessy> | othermaciej: what other chance do we have for a baseline codec, realistically? |
| 08:47 | <othermaciej> | I think they are interested in it the same way I am interested in exercising more at some point - vaguely in favor, but no concrete plans that would actually lead to that result |
| 08:47 | <jwm> | the only thing nice about software patents is when they expire |
| 08:47 | <jwm> | so we just need to wait another 10 years and we can use h264 |
| 08:47 | <jwm> | just like gif :) |
| 08:48 | <othermaciej> | nessy: I'm not aware of a currently viable solution, but it's possible that WebM/VP8 could be one in the future |
| 08:48 | <nessy> | why is it not viable now? |
| 08:49 | <othermaciej> | the IP situation is unknown, it's not an open standard, and it does not yet have any hardware support (and is apparently hard to implement in hardware) |
| 08:50 | <nessy> | we know that Google has patents on it and would be one of the first targets if something went wrong with IP - that's not an unknown patent situation IMHO |
| 08:51 | <nessy> | what consists an "open standard" is a matter of discussion - is RSS an open standard? is SRT? |
| 08:52 | <othermaciej> | maybe after some time has passed it will be more clear whether anyone wants to make an IP claim (or if it went through a standards process with disclosure requirements, it would at least get IP claims of the usual suspects out in the open) |
| 08:52 | <othermaciej> | RSS - yes, it has a published spec, and multiple independent implementations |
| 08:52 | <othermaciej> | SRT - no adequate spec, but Hixie is trying to fix that |
| 08:53 | <nessy> | is that your criteria for an open standard? interesting... |
| 08:53 | <nessy> | so, theora is an open standard? |
| 08:54 | <othermaciej> | I don't know enough about theora to say |
| 08:54 | <nessy> | I guess on the HW front, we should see soon... |
| 08:55 | <nessy> | btw: I don't see a need to write WebM into HTML5 as the baseline codec now - I think the market will decide more than anything else |
| 08:55 | <othermaciej> | are there any theora decoders that are not derivative works of the original VP3 code drop? |
| 08:55 | <othermaciej> | I agree, it is likely the market will decide |
| 08:56 | <nessy> | sometimes a de-facto standard can be achieved quicker than a standard that is created through lengthy discussions between many parties |
| 08:57 | <nessy> | the theora codebase now has a couple of serious forks by different groups, but I actually don't think anyone cared to write a completely new implementation |
| 08:58 | <nessy> | they are all still format-conformant, so not spec forks, but only forks for different platforms |
| 08:59 | <nessy> | but anyone could build an independent re-implementation if they wanted to - strangely nobody has cared to yet |
| 09:00 | <nessy> | maybe what needs to be done to encourage independent implementations is to make sure that the reference implementation is really bad ;) |
| 09:01 | <othermaciej> | now that Google owns On2's patents I guess they could also clarify that the patent rights are not tied to the specific implementation |
| 09:03 | <annevk> | othermaciej, from what I heard it sounded like there were plans, but I could be mistaken I suppose |
| 09:03 | <annevk> | (re standards body) |
| 09:10 | <othermaciej> | yeah I have no idea really |
| 09:11 | <jwm> | anyone know if the peer to peer provisions in html5 will be brought back to life? |
| 09:11 | <jwm> | maybe in a websocket form? |
| 09:11 | <jwm> | or if I could help in doing that? |
| 09:12 | <annevk> | if someone figures out the communication protocols that would go a long way |
| 09:12 | <annevk> | so if you can do that :) |
| 09:12 | <jwm> | well if you do it over websocket |
| 09:12 | <jwm> | that wouldn't be a problem would it? |
| 09:12 | <jwm> | you just specify a socket receive functionality |
| 09:12 | <annevk> | that only solves client <> server |
| 09:12 | <annevk> | not client <> client |
| 09:13 | <jwm> | I'm going to be starting a web based p2p network project |
| 09:13 | <jwm> | and I'd like to use web standards completely |
| 09:14 | <annevk> | no need for motivation, I'm sure we'd all love to have it :) |
| 09:14 | <jwm> | I think that would change the face of the web completely |
| 09:14 | <jwm> | or not the face but the landscape :) |
| 09:15 | <othermaciej> | MikeSmith: here is an article on a similar theme to what you posted: http://www.computerworlduk.com/community/blogs/index.cfm?entryid=2973&blogid=41 |
| 09:15 | MikeSmith | reads now |
| 09:16 | <boblet> | MikeSmith: posted http://stackoverflow.com/questions/2895214/what-is-the-semantically-correct-way-to-use-the-article-tag-in-html-5-with/2895665#2895665 |
| 09:16 | <boblet> | now wondering if I should change my templates based on my own advice :S |
| 09:17 | <MikeSmith> | heh |
| 09:19 | <MikeSmith> | hmm, "the new license Google is using for the project is one that's not been submitted to the Open Source Initiative for approval. As it stands it possibly can't be approved due to Google's ironic inclusion of a "field of use" restriction in the patent grant" |
| 09:20 | <annevk> | grmbl |
| 09:20 | <othermaciej> | that article is inaccurate on one point (by implication) |
| 09:20 | <othermaciej> | MPEG-LA does not offer any indemnification |
| 09:21 | <MikeSmith> | who does offer indemnification? |
| 09:22 | <MikeSmith> | what instances are there of companies/organizations actually offering indemnification for a particular technology? |
| 09:22 | <jwm> | why don't we just use gstreamer or something else to do video |
| 09:22 | <jwm> | why put the codec in the browser |
| 09:23 | <othermaciej> | MikeSmith: it apparently does happen sometimes but I don't know much about this area |
| 09:24 | <annevk> | jwm, we don't want to end up with a plethora of formats everyone has to support on whatever device imaginable |
| 09:25 | <jwm> | but that won't happen, the market will decide |
| 09:25 | <jwm> | I'd rather it left be open so that the market can be open to new formats |
| 09:26 | <jwm> | video/audio is something that is always developing |
| 09:26 | <jwm> | some new tech seems to come out all the new time |
| 09:26 | <jwm> | -new |
| 09:28 | <boblet> | new video tech!? quick, someone start a patent pool |
| 09:28 | <boblet> | it’s getting hot, and we’ll need something to drink martinis by pretty soon… |
| 09:30 | <ment> | what does patent pool mean? |
| 09:30 | <boblet> | that’s what the lawyers can afford to buy after litigation :| |
| 09:31 | <ment> | lol |
| 09:31 | <boblet> | I _think_ it means making a list of all the patents you have that you think you can sue the competing guys with |
| 09:31 | <boblet> | (but my first definition seems more accurate to me) |
| 09:32 | <boblet> | MikeSmith: added <p> to the flowchart of doom, but not as a new diamond: http://oli-studio.com/temp/sectioning-flowchart3.png |
| 09:32 | <boblet> | should anyone have flowchart feedback it would be gratefully received. especially if you aren’t too sure about the whole section/article thing |
| 09:33 | <MikeSmith> | boblet: ah, that looks good |
| 09:34 | <boblet> | MikeSmith: i kinda liked the linearity of the last one, but Dashiva’s reordering suggestion is prolly better for ppl who aren’t familiar |
| 09:34 | <boblet> | it’s hard to know if this will still make sense to ‘normal’ people |
| 09:35 | <jwm> | yeah not enough nudity |
| 09:35 | <MikeSmith> | normal people don't choose elements manually |
| 09:35 | <MikeSmith> | they push buttons |
| 09:35 | <MikeSmith> | and turn knobs |
| 09:36 | <boblet> | jwm: I took my t-shirt off, any improvement? |
| 09:36 | <boblet> | MikeSmith: well, ‘normal’ webmonkeys then |
| 09:36 | <jwm> | these are standards we're dealing with, you need to document your taking of your shirt off next time |
| 09:36 | <boblet> | the ones that are all section/article confused |
| 09:37 | <boblet> | jwm: sheesh |
| 09:37 | <jwm> | why not just steal visios graphics? |
| 09:37 | boblet | searches for the official shirt-removal form, and finds it’s in a members-only section of w3.org |
| 09:37 | <boblet> | foiled again! |
| 09:40 | <jgraham> | Those Members-only no-shirts parties are wild I tell you |
| 09:41 | <MikeSmith> | heh |
| 09:41 | <jwm> | until RMS joins in |
| 09:41 | <MikeSmith> | lol |
| 09:42 | <jgraham> | Now I will have nightmares |
| 09:42 | <jgraham> | Thanks |
| 09:43 | <MikeSmith> | well, if I were half the man that RMS is |
| 09:43 | <MikeSmith> | wait, I think I actually am |
| 09:43 | <jwm> | you know I never hear much about what he thinks of w3c |
| 09:43 | <jwm> | or web in general |
| 09:44 | <boblet> | annevk said the Keio no-shirt parties are really steamy |
| 09:44 | <jgraham> | The man uses email to get web pages |
| 09:44 | <jwm> | I know he hates how closed off the code/systems are that run servers |
| 09:44 | <jgraham> | That should tell you everything |
| 09:44 | <jwm> | I wonder if he'd appreciate a p2p web |
| 09:45 | <jwm> | where the software was all in the browser |
| 10:13 | <Lachy> | jwm, tor and freenet have P2P web solutions, and they're slow and painful to use. |
| 10:14 | <jwm> | they are slow anyway |
| 10:14 | <jwm> | the way the network coding is done |
| 11:34 | <Dashiva> | boblet: Your images seem to have disappeared |
| 11:55 | <boblet> | Dashiva: try again |
| 12:03 | <erlehmann_> | anyone knows how to coerce youporn into giving html5 video ? |
| 12:03 | <erlehmann_> | i heard it had this, but i only see flash. |
| 12:04 | <erlehmann_> | aside, when porn picks up html5, it is ready ;) |
| 12:06 | <ment> | so "html5 readiness research" is in fact looking at porn all day? :) |
| 12:10 | <Dashiva> | erlehmann_: ycombinator? |
| 12:11 | <erlehmann_> | Dashiva, wtf are you talking about. |
| 12:11 | <erlehmann_> | i have no idea what ycombinator is. |
| 12:11 | <jgraham> | ment: Only if you are Hyatt, probably |
| 12:12 | <Dashiva> | erlehmann_: I heard the claim that youporn had html5 video earlier, but I could only source it to a single post from there |
| 12:12 | <jgraham> | erlehmann_: It's a venture capital company for tech startups. But, more relevantly, it is the home of Hacker News |
| 12:12 | <erlehmann_> | Dashiva, i read it on netzpolitik.org, but they didn't name any sources. |
| 12:13 | <jgraham> | erlehmann_: Which is like reddit but without many of the things that make reddit crap |
| 12:13 | <Dashiva> | The actual youporn site just links to .3gp file |
| 12:13 | <jgraham> | erlehmann_: and some unique things that make it crap |
| 12:13 | <Dashiva> | Which, true enough, isn't flash |
| 12:18 | <nielsle> | If you launch a html5 based youporn competitor now, then you will be able to target a large group of linux-using virgins. |
| 12:19 | <Dashiva> | They would never use closed-source porn |
| 12:23 | <boblet> | Dashiva: oh, any feedback? I’m getting “not understandable for normal webmonkeys” from someone, and am worried it’s true :) |
| 12:24 | <Dashiva> | I would agree, but at the same time normal webmonkeys would probably never think to use such a chart anyway |
| 12:24 | <boblet> | Dashiva: ohrly? |
| 12:24 | <Dashiva> | The users would have to be semantic-conscious enough to actually care about the right element |
| 12:37 | <slartsa> | apparently youporn is a necessary element of html development :) |
| 12:39 | <zcorpan_> | we should make youporn use websockets with draft76 to settle the handshake |
| 12:43 | <jgraham> | ${purile_porn_handshake_joke} |
| 12:45 | <gsnedders> | jgraham: Is that meant to be Perl? |
| 12:46 | <jgraham> | gsnedders: Genshi |
| 12:47 | <slartsa> | this channel is kinky, I like it here |
| 12:47 | <gsnedders> | slartsa: Wait, what sort of porn do you think we're talking about? |
| 12:48 | <slartsa> | any you could think of ;) |
| 12:48 | <jgraham> | |------------------------------| Channel kinkiness |
| 12:48 | <jgraham> | ^ you are here |
| 12:48 | <jgraham> | This is quite far above the average for the channel as you see |
| 12:49 | <slartsa> | Awesomeness. |
| 12:56 | <nielsle> | erlehmann_: Install the firefox plugin named user agent switcher. If you set your use agent to iphone then you can watch some of the videos as mp4. |
| 13:02 | <erlehmann_> | >firefox >mp4 |
| 13:02 | <erlehmann_> | do you realize the error of your ways, nielsle |
| 13:02 | <erlehmann_> | ??? |
| 13:02 | <erlehmann_> | besides, vp8 has double the version number |
| 13:03 | <erlehmann_> | so it must be TWICE AS GOOD |
| 13:03 | <boblet> | Dashiva: I’m hoping to make something that the ppl who use http://www.flickr.com/photos/antoniolupetti/3894233282/sizes/l/ or http://media.smashingmagazine.com/cdn_smash/wp-content/uploads/images/html5-cheat-sheet/preview.gif would find useful |
| 13:14 | <boblet> | would you say that the figure element is tangentially related, or part of the page’s main content? |
| 13:18 | <Dashiva> | It's part of the main content, but doesn't need to be inside the main content |
| 13:23 | <boblet> | Dashiva: thanks, that’s what I thought |
| 13:24 | <Dashiva> | At least that's how I understand it. If it wasn't part of the main content it'd be an <aside> |
| 13:26 | <boblet> | I’m actually wondering if the presentation (something with caption) will end up trumping the “could be moved to a different place, such as an appendix” part |
| 13:27 | <boblet> | am thinking that I’ve used figure for content that if moved would kill the article :/ |
| 13:36 | <Dashiva> | I think it's a common guideline to require all figures to be referenced explicitly in the main text |
| 13:36 | <Dashiva> | Serving as the "entry point" where it's natural to look at the figure |
| 14:26 | <nessy> | wow - nothing makes my browsers steam up my laptop as much as loading the html5 spec! |
| 14:27 | <slartsa> | http://hixie.ch/tests/evil/html/parsing/compat/001.html try that |
| 14:28 | <Philip`> | nessy: I bet the Web Applications 1.0 spec steams it up more |
| 14:29 | <nessy> | you mean: http://www.whatwg.org/specs/web-apps/current-work/ ? |
| 14:30 | <nessy> | that's the one I meant ;-) |
| 14:30 | <nessy> | maybe I shouldn't use "html5 spec" for it, but "html5" has become a placeholder for all the modern web app specs, I guess |
| 14:31 | <nessy> | haha - only seconds after loading that page, Chrome goes haywire and the same for firefox |
| 14:31 | <slartsa> | :P |
| 14:31 | <Philip`> | nessy: No, I mean http://www.whatwg.org/specs/web-apps/current-work/complete.html |
| 14:32 | <nessy> | Philip`: you're right - that page makes it take off even before the page is loaded |
| 14:34 | <slartsa> | took 10 seconds to load but works |
| 14:34 | <slartsa> | HUGE amount of data |
| 14:40 | <doublec> | nessy, try doing an incremental search in firefox with the complete whatwg spec loaded |
| 14:40 | <doublec> | nessy, definitely painful |
| 14:41 | nessy | shudders |
| 14:41 | <nessy> | the html5 spec is definitely the challenge of the future for web browsers ;) |
| 14:41 | <doublec> | browsing the spec with curl http://... |less however is fast |
| 14:43 | <gsnedders> | What you really want is http://gsnedders.html5.org/html5.txt |
| 14:44 | <Philip`> | You want the PDF, since that (I assume?) has working hyperlinks |
| 14:45 | <doublec> | gsnedders, you are right, that txt file is nice |
| 14:46 | <doublec> | 27.584 references, wow! |
| 14:52 | <nielsle> | I am toying with a canvas based painting program, and I want to make an eraser. In other words I want make the canvas transparent again. Can I define a strokeStyle that does this? |
| 14:53 | <nielsle> | (I hope that I am in the right channel.. I tried to judge from previous conversation :) ) |
| 14:54 | <Philip`> | nielsle: You should be able to do it by setting globalCompositeOperation |
| 14:55 | <Philip`> | to e.g. destination-out (I think) |
| 14:55 | <Philip`> | and then draw a solid line |
| 14:55 | <nielsle> | Cool thanks. |
| 14:55 | <Philip`> | (Maybe it's more like source-out or something) |
| 15:07 | <Dashiva> | Hum, http://www.osnews.com/story/23346/Nero_Files_Antitrust_Case_Against_MPEG-LA |
| 15:07 | <Dashiva> | It seems there are lawsuits all over lately |
| 15:25 | <nessy> | Dashiva: that's a good thing - stuff gets clarified in lawsuites |
| 15:25 | <nessy> | stops FUD from happening |
| 15:29 | <boblet> | and lawyers get new pools |
| 15:39 | <nessy> | othermaciej: just got my suspicions of earlier on confirmed about theora - there are indeed two independent implementations - the one from xiph and the one from ffmpeg - they both originated from vp3 code, but both have changed so much that there is assumed there is no code overlap any longer |
| 17:12 | <AryehGregor> | So, suppose I'm a web developer and give users the ability to create images somehow. Suppose the user provides no alt text, for whatever reason. Should I put no alt attribute, or empty alt text, or something else? Is there any guidance about this? |
| 17:13 | <Dashiva> | That's one of the controversial areas |
| 17:13 | <AryehGregor> | Yeah, I figured. |
| 17:14 | <AryehGregor> | It seems to me that logically, you'd want to either have no alt attribute or a magic value. Having alt="" or the filename or whatever strikes me as bad from everyone's perspective. |
| 17:14 | <Dashiva> | Yes |
| 17:15 | <AryehGregor> | But it also strikes me that if you allow this, then suddenly zillions of sites will have to switch from alt="" to your magic value and everyone will be forced to admit that few sites have ever used decent alt text. |
| 17:15 | <AryehGregor> | As opposed to calling required alt text a success, as a Mozilla developer recently did on public-html. |
| 17:15 | <Dashiva> | Heh |
| 17:15 | <jgraham> | You just use missing alt as your magic value |
| 17:15 | <AryehGregor> | Yes, that seems to make the most sense. |
| 17:15 | <Dashiva> | Well, the controversy isn't just the syntax, but also whether it should be conforming or not |
| 17:15 | <jgraham> | Not to the Experts |
| 17:16 | <AryehGregor> | But then you get people complaining that your app isn't conforming, unless missing alt text is conforming. |
| 17:16 | <jgraham> | unless they changed their minds |
| 17:16 | <AryehGregor> | So you're pressured into missing alt text. |
| 17:16 | <Dashiva> | That's how it ends up, yes |
| 17:16 | <AryehGregor> | I mean, you're pressured into adding empty or useless alt text. |
| 17:16 | <Dashiva> | If missing alt is the solution, and missing alt is nonconforming, you get bogus alt |
| 17:17 | <Dashiva> | If magic value is the solution, you get bogus magic value |
| 17:17 | <AryehGregor> | What I'd like is some guidance that actually tells me what the best course of action is for screen readers. |
| 17:17 | <Philip`> | Current or future or science-fiction screen readers? |
| 17:17 | <Dashiva> | If missing alt is the solution, and missing alt is conforming, you get "but that's not machine checkable" |
| 17:17 | <AryehGregor> | Surely screen readers need to know the difference between "don't know what the alt should be" and "alt should be empty, this is decorative". |
| 17:17 | <AryehGregor> | Current or near future. |
| 17:17 | <Dashiva> | Screen readers want missing alt |
| 17:17 | <Dashiva> | They have to handle those anyway |
| 17:18 | <AryehGregor> | Dashiva, do they want it more than filler like the filename? |
| 17:18 | <AryehGregor> | I assume they could fill in the filename themselves. |
| 17:18 | <AryehGregor> | Or other useful info. |
| 17:18 | <Philip`> | Decorative is easy, that's just alt="" |
| 17:18 | <Dashiva> | I'm sure you can find enough screen readers to justify either case |
| 17:18 | <Dashiva> | Because apparently we have to consider every version of every product ever released |
| 17:19 | <Dashiva> | (JAWS is currently version 11, but version 4 is brought up in discussions...) |
| 17:19 | <Philip`> | And every user's configuration of said products |
| 17:22 | <AryehGregor> | Maybe nobody upgrades JAWS because of how ludicrously expensive it is? |
| 17:23 | <Philip`> | That's the common argument |
| 17:24 | <AryehGregor> | A blind person once told me Mac's built-in screen reader is the best. |
| 17:25 | <Philip`> | Other blind people may say very different things |
| 17:26 | Philip` | is unaware of objective comparisons between the various products |
| 17:26 | Philip` | guesses people learn to use one and then want to stick with it for a long time, because it's so hard to learn another even if it's supposedly better |
| 17:26 | <AryehGregor> | What he said completely met my expectations of the different platforms, though, so I'm inclined to believe him. |
| 17:27 | <AryehGregor> | He said on Windows, the software is lousy, made by third parties, and unreasonably expensive. |
| 17:27 | <AryehGregor> | On Linux, it's still lousy, but at least it's free. |
| 17:27 | <AryehGregor> | On Mac, it comes with the OS and is excellent. |
| 17:27 | <AryehGregor> | I'd expect Apple to make a good screen reader. It's the sort of thing they do well. UI, basically. |
| 17:28 | <AryehGregor> | Whereas open-source stuff, and random special-purpose Windows apps, tend to have lousy UI. |
| 17:30 | <Philip`> | Looks like JAWS is about $900, but presumably you get significant discounts via employers or other groups |
| 17:30 | <Philip`> | so maybe JAWS + a cheap Windows PC/laptop is still cheaper than a Mac |
| 17:31 | <AryehGregor> | You can surely get a Mac of some kind for $900. |
| 17:32 | <Philip`> | You can, but you can get a Windows one for much less, and the difference may be greater than a discounted JAWS |
| 17:33 | <Philip`> | Seems like it won't be a huge difference, though |
| 17:33 | <AryehGregor> | Why do you assume that you can get a discounted JAWS so easily? |
| 17:38 | <Dashiva> | Isn't that the norm? |
| 17:38 | <Dashiva> | Bulk discounts for organizations |
| 17:38 | <Philip`> | AryehGregor: Because they talk about multi-user pricing on their site, and because I vaguely remember some mentioning how people often get that kind of thing through an employer rather than buying it personally |
| 17:38 | <Philip`> | though I could be entirely misremembering and/or wrong |
| 17:38 | <Philip`> | s/some/someone/ |
| 17:38 | <AryehGregor> | Dashiva, if you have an entire organization of blind people, yes. |
| 17:39 | <othermaciej> | I assume they can charge so much for JAWS because users can get some kind of subsidy |
| 17:39 | <Philip`> | AryehGregor: Any reasonably large organisation will probably have enough to make it worthwhile |
| 17:39 | <AryehGregor> | On the other hand, maybe blind people do tend to work for companies full of blind people. I suppose they'd have some trouble at a random employer, even in a carefully-selected field. |
| 17:39 | AryehGregor | wonders about that. |
| 17:40 | <Dashiva> | I would assume most blind people are members of an interest organization for blind (or somewhat more general) |
| 17:40 | <TabAtkins> | The first time I ever saw a blind person actually at work was here at Google. |
| 17:40 | <TabAtkins> | Dude works in one of the nearby buildings. |
| 17:40 | <AryehGregor> | Dashiva, *most*? Really? |
| 17:41 | <AryehGregor> | I can't imagine it's most. |
| 17:41 | <AryehGregor> | TabAtkins, I'd be totally unsurprised if Google made a point of employing more blind and deaf people than normal. |
| 17:41 | <TabAtkins> | Indeed. |
| 17:41 | <Dashiva> | AryehGregor: Why not? |
| 17:41 | <AryehGregor> | Do you know what he does? |
| 17:41 | <TabAtkins> | AryehGregor: No clue. |
| 17:41 | <TabAtkins> | I just see him at lunch. |
| 17:42 | <AryehGregor> | Dashiva, because . . . there are surely way more blind people than employees of blind-person advocacy groups? |
| 17:42 | <jgraham> | So if I have a document with some timeouts and document.write in it in such a way that the original document is blown away, should the timeouts still fire? |
| 17:43 | <TabAtkins> | jgraham: first thought is no. That is not based on any evidence or even knowledge, though. |
| 17:43 | <jgraham> | Yeah that is my first thought, based on the same |
| 17:43 | <Philip`> | TabAtkins: Are you attempting to apply logic to determine the expected outcome? |
| 17:43 | <othermaciej> | wikipedia says: http://en.wikipedia.org/wiki/Blindness#Epidemiology |
| 17:43 | <TabAtkins> | Philip`: Yes. In absence of reality, logic is all we have. |
| 17:44 | <Philip`> | TabAtkins: I'm afraid you are violating this channel's rules |
| 17:45 | <AryehGregor> | othermaciej, I guess when it comes to blindness you have to remember that not everyone who's legally blind is really fully blind. |
| 17:45 | <othermaciej> | "do the timeouts still fire" is probably a more relevant question |
| 17:45 | <Dashiva> | AryehGregor: How does employees figure into it? |
| 17:45 | <TabAtkins> | Philip`: Damn, you're right. |
| 17:45 | <AryehGregor> | My mother once taught an art course with a blind art major in it, who was an excellent student and got an A. |
| 17:45 | <Dashiva> | I said member, not employee |
| 17:45 | <TabAtkins> | I need to base more opinions on knee-jerk opinions, then. |
| 17:45 | <AryehGregor> | She could still see a little. |
| 17:45 | <othermaciej> | AryehGregor: that article distinguishes those of the legally blind who are not totally blind as "low vision" |
| 17:46 | <othermaciej> | although many low vision persons have low enough vision that they would in practice need a screen reader to use a compuer |
| 17:46 | <AryehGregor> | Dashiva, oh, I misread you. Then you might be right. |
| 17:46 | <othermaciej> | "Approximately ten percent of those deemed legally blind, by any measure, have no vision. The rest have some vision, from light perception alone to relatively good acuity. http://en.wikipedia.org/wiki/Low_vision is sometimes used to describe visual acuities from 20/70 to 20/200." |
| 17:47 | <TabAtkins> | Damn, really? Without contacts I'm legally blind, then. |
| 17:47 | <Philip`> | TabAtkins: Opinions are fine because they're easy to detect and ignore; logic is dangerous because it tricks you into believing its conclusions are true |
| 17:47 | <TabAtkins> | Makes sense. There's no way I can drive without contacts or glasses. I'd kill myself and others. |
| 17:47 | <AryehGregor> | TabAtkins, it all says "with best correction possible". |
| 17:48 | <AryehGregor> | So it's measuring how good your vision is *with* contacts or glasses or whatever. |
| 17:48 | <Philip`> | so we should stick to opinions and facts (and ignore the former) |
| 17:48 | <TabAtkins> | AryehGregor: Ah, kk. Well, still, my statement stands. Without contacts I'm blind. ^_^ |
| 17:48 | <TabAtkins> | Philip`: That's disturbingly convincing. |
| 17:48 | <Philip`> | AryehGregor: Are cyborg eyes possible? |
| 17:48 | <AryehGregor> | Philip`, surely you mean we should ignore the latter. |
| 17:49 | <Philip`> | Oh, it says "spectacle or contact lens correction", which excludes that |
| 17:49 | <TabAtkins> | AryehGregor: You and I have nearly identical commenting patterns. (Based on gavinsharp's room stats.) |
| 17:50 | <Philip`> | (Well, one of the pages says that) |
| 17:51 | <AryehGregor> | TabAtkins, oh, you mean like time zones. Odd. You're on the West Coast, right? |
| 17:51 | <AryehGregor> | *And* I tend to go to bed early, which should exacerbate the difference. |
| 17:51 | <TabAtkins> | AryehGregor: Now I am, yeah. But I was Central for most of my commenting history. |
| 17:51 | <AryehGregor> | Hmm, that makes more sense. |
| 17:52 | <TabAtkins> | AryehGregor: Also, I'm rarely in the room in the evening, so you going to bed early actually helps align us. |
| 17:52 | <AryehGregor> | Blue isn't even visible for either of us. |
| 17:52 | <TabAtkins> | I'm pretty sure I've never commented in a blue time. |
| 17:53 | <AryehGregor> | I'm sure I have occasionally, although it doesn't say the time zone. |
| 17:55 | <jgraham> | Isn't gavin in Mountain View? Although his server might not be of course |
| 17:55 | <TabAtkins> | It has to be an american time zone, or else I'd be in blue at some point. |
| 17:56 | <AryehGregor> | What if it's a Canadian time zone?! |
| 17:56 | <TabAtkins> | We own those time zones too. |
| 17:56 | <TabAtkins> | alternately: I meant american supercontinent. |
| 17:56 | <AryehGregor> | No one in America follows Newfoundland time. |
| 17:56 | <AryehGregor> | It's not aligned on one-hour boundaries. |
| 17:56 | <TabAtkins> | Nobody cares about newfie time, though, so it's all right. |
| 17:56 | <AryehGregor> | Also too far east. |
| 17:56 | <AryehGregor> | Granted. |
| 17:57 | <AryehGregor> | Typical Wikipedia moment: "Oh, so is that what it's called? *reads*" http://en.wikipedia.org/wiki/Tactile_paving |
| 17:57 | <TabAtkins> | Huh. I always thought that was just for grip or something. |
| 17:58 | <AryehGregor> | Yeah, I never realized it was for blind people, although it makes sense. |
| 18:00 | <TabAtkins> | boblet's in Japan, right? |
| 18:02 | <Dashiva> | His 404 page is Japanese, at least |
| 18:02 | Philip` | always knew they were for blind people, but assumed they were to tell blind people where it was safe to cross the road, rather than to tell them where it's *not* safe to walk without care |
| 18:03 | <TabAtkins> | Philip`: Good accessiblity is multi-purpose! |
| 18:09 | <AryehGregor> | Philip`, that wouldn't explain why they're on the edges of train platforms. |
| 18:12 | <Philip`> | AryehGregor: Indeed, my assumption of the explanation did not fit perfectly with reality |
| 19:25 | <jwm> | http://x264dev.multimedia.cx/?p=377 |
| 19:51 | gsnedders | wonders what self-raising flour is in Swedish, and what grease-proof paper is |
| 19:51 | gsnedders | also wonders whether to cook today or tomorrow |
| 20:14 | <TabAtkins> | gsnedders: grease-proof paper: wax paper? or parchment paper? |
| 20:15 | <gsnedders> | Um, http://en.wikipedia.org/wiki/Greaseproof_paper |
| 20:15 | <TabAtkins> | Ah, kk. |
| 20:24 | <jgraham> | gsnedders: self-rasing lour doesn't exist |
| 20:24 | <gsnedders> | jgraham: gah, |
| 20:24 | <jgraham> | you just have to use ordinary flour + baking powder |
| 20:24 | <gsnedders> | *gah |
| 20:24 | <gsnedders> | jgraham: What's baking powder, then? |
| 20:24 | <jgraham> | Something quite obviosu |
| 20:25 | <jgraham> | Bakpulver |
| 20:25 | <gsnedders> | ah, ok |
| 20:25 | <jgraham> | Greaseproof paper is something like Bakpaper |
| 20:26 | gsnedders | is proving how much baking he's done in the past year |
| 20:27 | <jgraham> | I don't know the Swedish for "my oven's on fire" so if you reach that stage you're on your own |
| 20:27 | gsnedders | has concluded he'll just bake after lunch tomorrow |
| 20:27 | <jgraham> | What? |
| 20:28 | <gsnedders> | Well, it's a bit on the late side already, and it won't take long |
| 20:28 | <gsnedders> | (See, having lunch at $ridiculousTime has advantages |
| 20:30 | <Dashiva> | http://www.robglidden.com/2010/05/how-googles-open-sourcing-of-vp8-harms-the-open-web/ |
| 20:31 | <Dashiva> | So is the argument here that we should put web video on pause for 10 years while stuff is developed? |
| 20:32 | <gsnedders> | jgraham: And now you're going to call me insane :P |
| 20:40 | <TabAtkins> | Dashiva: I'd like to believe that the dude recognizes that every solution is bad, and that he's merely pointing out the badness of the particular solution that Google chose. |
| 20:42 | <Dashiva> | It doesn't seem that way. "It is well known that many experts consider it now feasible to standardize serviceable royalty-free codecs." and "Google should get on board on this important trend, not undermine it" |
| 20:43 | <Dashiva> | This also seems to show a misunderstanding of how patent trolls work: "Contributing VP8 to a standards group with a strong patent disclosure policy would be a good corrective move; it would force lurking patent holders to come fully into the public." |
| 20:43 | <TabAtkins> | Yeah, I'd just like to believe that. |
| 20:44 | <TabAtkins> | Truthfully he seems to be painfully naive. I noticed that last one you quoted in particular. |
| 20:51 | <AryehGregor> | Dashiva, a strong patent disclosure policy like ISO, which produced H.264? |
| 20:51 | <AryehGregor> | Disclosure isn't the same as licensing. |
| 20:51 | <TabAtkins> | Yup, I think he has a mistaken idea of what it means to "disclose" a patent. |
| 20:51 | <AryehGregor> | Patent trolls don't join groups with patent licensing policies. |
| 20:51 | <TabAtkins> | It can mean "Yup, we have a patent. Have fun licensing it!" |
| 20:52 | <AryehGregor> | Even in a group like the W3C, anyone with patents could always refuse to join the Working Group rather than disclose or license its patents. |
| 20:53 | <AryehGregor> | Assuming they're a member in the first place. |
| 20:54 | <Dashiva> | But I read another thing somewhere else. What if someone came up with a patent that turned out to be essential to h264, and refuse to license it. Could they shut down everyone using h264? |
| 20:55 | <AryehGregor> | Sure, in theory. You don't have to license your patents to anyone. |
| 20:55 | <AryehGregor> | It's unlikely that they wouldn't accept whatever payout they were offered, though. |
| 20:55 | <AryehGregor> | Also, we could just have H.264a released really quickly to work around the patent. |
| 20:55 | <AryehGregor> | Plus there's H.265 one of these days. So it wouldn't help much, I don't think. |
| 20:56 | <Dashiva> | Well, the concept was that someone would do it because they wanted to give open video a boost by shutting down the best alternative |
| 20:56 | <TabAtkins> | That would be somewhat hilarious, actually. |
| 21:07 | <TabAtkins> | I submitted a comment to glidden's post expanding on what we're saying here. |
| 22:08 | <annevk> | omg lost |
| 22:08 | <annevk> | still can't fathom it |
| 22:08 | <annevk> | "retardation" comes to mind |
| 22:09 | <ment> | lost what? |
| 22:09 | <jgraham> | The TV series? |
| 22:09 | <ment> | ah, Lost :) |
| 22:09 | <ment> | well now i'm glad i've stopped watching that after first season |
| 22:10 | <annevk> | LOST but I was going for no caps |
| 22:10 | <jgraham> | I have never seen it but I thought the general consensus was that the writers had gone "fuck we can do whatever we want and they will kep watching this shit" around season 2 |
| 22:12 | <Dashiva> | Yeah |
| 22:14 | <annevk> | it had a lot of highlights for me, but the wrapping up of it utterly failed |
| 22:15 | <volkmar> | Hixie: ping |
| 22:15 | <Dashiva> | Time travel is always a bad sign |
| 22:32 | <Philip`> | Dashiva: Not if it's in Doctor Who |
| 22:38 | <Dashiva> | Doctor Who doesn't have time travel, it just has travel that happens to use time as medium instead of space |
| 22:45 | <jwm> | aka time travel |
| 22:45 | <jwm> | :) |
| 22:45 | <gratz|home> | Doctor Who doesn't really exist |
| 22:45 | <AryehGregor> | He doesn't? |
| 22:45 | <AryehGregor> | Noooo! |
| 22:47 | <jwm> | any plans to add saving of data uris to html 5? |
| 22:48 | <AryehGregor> | "Saving of data uris"? |
| 22:48 | <jwm> | generated data of any kind, or data uris themselves to files |
| 22:48 | <Philip`> | Dashiva: They're not really the same - if it was equivalent to travel through space then you'd never be able to meet yourself |
| 22:48 | <jwm> | instead of having to have a server connection to download a file |
| 22:48 | <jwm> | just set a mimetype to any data in javascript |
| 22:48 | <TabAtkins> | jwm: You can just load a data uri and then Save As, no? |
| 22:48 | <jwm> | I mean an actual file dialog generated from javascript |
| 22:49 | <TabAtkins> | Oh, that. Yes, there is. Not through data urls, though. |
| 22:49 | <AryehGregor> | Philip`, I can meet myself anytime I like. |
| 22:49 | <AryehGregor> | Hello, AryehGregor. How are you? |
| 22:49 | <AryehGregor> | I'm okay, thanks for asking. |
| 22:49 | <Dashiva> | Philip`: Travel to a mirror |
| 22:50 | <jwm> | TabAtkins: cool, any info in the spec? |
| 22:50 | <jwm> | also does anyone know about the [dropped] peer to peer provisions in html5, I know websockets got moved out |
| 22:50 | <jwm> | but I'd like to help if I can to add p2p to websockets |
| 22:50 | <TabAtkins> | The work isn't in HTML, but rather in javascript. Check the webapps archives for File objects and similar. |
| 22:50 | <jwm> | a possible idea I was having were listen websockets |
| 22:51 | <jwm> | also adding upnp/dlna tech |
| 22:51 | <jwm> | but that would be browser specific |
| 22:51 | <jwm> | can't call it a browser anymore once we do p2p though :) |
| 22:51 | <jwm> | TabAtkins: cool yesterday I subscribed to the list |
| 22:51 | <Philip`> | Dashiva: That still won't let you violate causality |
| 22:52 | <Philip`> | whereas time travel does as long as you don't mind being attacked by special effects because of it |
| 22:52 | <Dashiva> | That's not really important, since none of the causality violations in Dr Who have any real consequences |
| 22:54 | <jwm> | webworkers + file api + websockets (with listening) = freenet web :) |
| 22:54 | <TabAtkins> | A "serverless" web? |
| 22:54 | <jwm> | yeah |
| 22:54 | <jwm> | that is what I would like to help create heh |
| 22:55 | <TabAtkins> | It's a cool goal. |
| 22:55 | <jwm> | real cloud/real web |
| 22:55 | <jwm> | if you want to use buzzwords :) |
| 22:55 | <Dashiva> | Unite not good enough for you? :) |
| 22:55 | <jwm> | nah |
| 22:55 | <jwm> | unite still uses opera's servers |
| 22:55 | <jwm> | it's nothing more than a proxy |
| 22:56 | <jwm> | I'd like to get the web to where it has the least reliance on client/server communication as possible |
| 22:56 | <Dashiva> | That's just for the hostname |
| 22:56 | <jwm> | you can't do peer to peer in unite |
| 22:56 | <jwm> | as far as I read |
| 22:56 | <AryehGregor> | It's a cool goal that's more or less worthless in practice, because a client-server model works much better unless you're trying to evade a concerted effort by powerful but not-too-powerful enemies who are trying to systematically silence you. |
| 22:56 | <jwm> | AryehGregor: ie governments? |
| 22:56 | <AryehGregor> | Pretty much. |
| 22:56 | <jwm> | it'd be harder than heck to filter web p2p traffic |
| 22:57 | <jwm> | you can't just block ports and call it a day |
| 22:57 | <AryehGregor> | Particularly civilized governments that actually believe in trials and due process, rather than breaking down your door and shooting you. |
| 22:57 | <jwm> | you'd block a lot of legitimate traffic |
| 22:57 | <AryehGregor> | You can't just block ports with BitTorrent either. |
| 22:57 | <jwm> | performance of p2p can be great though |
| 22:57 | <AryehGregor> | In some cases, yes. If it means that the server is on the same network as you. |
| 22:58 | <AryehGregor> | If you have to go over twice as much last-mile copper wire, though, it's probably slower. |
| 22:58 | <jwm> | I get higher throughput using torrent than I get from regular servers |
| 22:58 | <jwm> | yeah I'm just saying it would help relieve the reliance on central servers |
| 22:58 | <jwm> | not eliminate it |
| 22:58 | <jwm> | but that option would be there |
| 22:58 | <jwm> | it'd help a lot of projects that can't afford to pay for bandwidth |
| 22:59 | <jwm> | also I'd like to run payloads across the network |
| 22:59 | <jwm> | AI, scientific research stuff (kinda like folding@home or distributed.net) |
| 22:59 | <AryehGregor> | By letting them sneak the bandwidth away from their ISP because home connections are usually not metered. A technical solution to a social problem. |
| 22:59 | <jwm> | if you had 300 million people running a browser that would be willing to let you use 10% of their cpu power/network bandwidth that'd be a powerful network |
| 23:00 | <AryehGregor> | It'd also use up a ton of power for something you probably don't care about. |
| 23:00 | <jwm> | it'd cut down on all the advertising based revenue driven sites |
| 23:00 | <AryehGregor> | Witness the school administrator who got fired for installing Folding@Home on all his school's computers. |
| 23:00 | <AryehGregor> | Means they can't shut off and save power. |
| 23:01 | <jwm> | yeah but you are just bringing up negative aspects |
| 23:01 | <jwm> | which there are plenty |
| 23:01 | <jwm> | but there are already 80% of workers at most institutions looking at facebook |
| 23:01 | <AryehGregor> | You still need a central server that receives all requests if you want the site to be up-to-date. |
| 23:01 | <jwm> | using up a large majority of network bandwidth and processor power |
| 23:01 | <jwm> | watching youtube, etc |
| 23:01 | <jwm> | you can do a supernode hybrid p2p network |
| 23:01 | <jwm> | like gnutella |
| 23:01 | <AryehGregor> | I'm being negative because peer-to-peer stuff has been tried before and has never really succeeded except for illegal activity. |
| 23:02 | <AryehGregor> | The vast majority of things are client-server for a reason. |
| 23:02 | <jwm> | well I'd like to change that |
| 23:02 | <jwm> | I use p2p to download large legal files too |
| 23:02 | <jwm> | works nice |
| 23:02 | <AryehGregor> | You can't change the reasons for why most things are client-server, because they're good reasons. |
| 23:02 | <jwm> | why not argue the other side if you want to be unbiased |
| 23:02 | <jwm> | client server doesn't get rid of illegal behavior either |
| 23:03 | <jwm> | heck a lot of piracy happens on irc |
| 23:03 | <AryehGregor> | I don't believe I'm especially biased (or rather, I can't tell which direction my bias lies in). I've reached a conclusion based on my analysis of the evidence. |
| 23:03 | <jwm> | or usenet |
| 23:03 | <AryehGregor> | I said P2P is mostly useful for illegal activity, not that client-server isn't also useful for that. |
| 23:03 | <jwm> | if you could create a web that had the capability of running your favorite app without advertising if you just let it use some of your resources |
| 23:04 | <Philip`> | Few web sites are static, so they need to do lots of processing per request, and distributed processing is really hard, and very widely distributed processing across unreliable heterogeneous undebuggable hardware is much harder |
| 23:04 | <jwm> | it'd make us less reliant on information hungry companies like google |
| 23:04 | <jwm> | and rather on each other |
| 23:04 | <AryehGregor> | Distributed computing power isn't worth much. |
| 23:04 | <jwm> | tell that to botnets :) |
| 23:04 | <AryehGregor> | It's hard to use for general-purpose problems because of how unreliable it is. |
| 23:04 | <AryehGregor> | Again, illegal activities. P2P is good if you want to evade authorities, because it's hard to stamp out. |
| 23:05 | <AryehGregor> | Also, botnets are usually controlled centrally. |
| 23:05 | <AryehGregor> | With a client-server model. |
| 23:05 | <jwm> | so you wouldn't want to see any p2p in your browser at all? |
| 23:05 | <jwm> | not even for running a webapp that could use it for benefits outside illegal activity? |
| 23:05 | <jwm> | I say it's worth it to let it happen |
| 23:05 | <AryehGregor> | I don't care about P2P per se. I care about concrete useful features. I don't foresee P2P providing any for me. |
| 23:05 | <jwm> | let researchers give it a try |
| 23:05 | <jwm> | well I do |
| 23:06 | <AryehGregor> | You could say that about anything. Ultimately it's about what browser implementers think they should prioritize. |
| 23:06 | <jwm> | I'd love to not rely on google but instead have an army of opensource engineers making apps that have tons of bandwidth or cpu activity |
| 23:06 | <daedb> | I already have p2p in my web browser :p |
| 23:06 | <AryehGregor> | Problem is, that doesn't work so well. Coordination is essential for large-scale activities. |
| 23:06 | <Philip`> | Bandwidth and CPU are cheap, armies of developers are expensive |
| 23:06 | <AryehGregor> | Even something like Folding@Home isn't peer-to-peer, really. Distributed computing is different from P2P. |
| 23:07 | <jwm> | opensource projects are sort of loosely coupled |
| 23:07 | <AryehGregor> | And yeah, usually the easy part is getting hosting. The hard part is the actual development. |
| 23:07 | <AryehGregor> | You can get hosting from ads, or donated mirrors, or just personal funding unless your app is huge. |
| 23:07 | <jwm> | yes I know, distributed computing can be distributed in many different ways, p2p just being one of them |
| 23:08 | <jwm> | yeah but barring money, ads, and personal funding, and you know the first two are hard to come by typically |
| 23:08 | <jwm> | what if the only reason people avoid an idea is that it would cost money |
| 23:08 | <jwm> | to get the idea out there |
| 23:09 | <jwm> | Philip`: the army of developers could be the people using the network itself |
| 23:09 | <jwm> | just like google |
| 23:09 | <jwm> | google's developers aren't just the people that sit in an office all day coding |
| 23:09 | <AryehGregor> | Ads are very easy to come by. They're very efficient, and will only become moreso, I suspect. |
| 23:09 | <jwm> | they are also the customers that visit and share their information |
| 23:10 | <jwm> | I think it's a really interesting distribution model that is being left out of browsers |
| 23:10 | <Philip`> | jwm: Armies of people writing code and debugging it and administering the systems are expensive, then |
| 23:10 | <jwm> | it'll probably be here in the future |
| 23:10 | <Philip`> | and users aren't part of that |
| 23:11 | <jwm> | networks are getting faster, computers are getting faster |
| 23:11 | <jwm> | it's a waste not to have access to all those resources |
| 23:11 | gsnedders | notes that the majority of countries where internet access is becoming commonplace don't have very quick internet access |
| 23:12 | <jwm> | yes but there are still 100s of millions of people with fast internet |
| 23:12 | <AryehGregor> | You're conflating distributed computing and peer-to-peer networking again. |
| 23:12 | <jwm> | it'll just get better |
| 23:12 | <AryehGregor> | Distributed computing isn't very useful for most tasks, because it's unreliable and high-latency. |
| 23:12 | <jwm> | I don't have the two confused, I am just exposing what I am interested in |
| 23:12 | <jwm> | I'm interested in p2p and distributed computing |
| 23:13 | <jwm> | ibm's grid computing |
| 23:14 | <jwm> | I'm in the communications field and go into a lot of homes.. tons of people use p2p |
| 23:14 | <Philip`> | jwm: Networks and computers are getting faster meaning they're going to no longer be the bottlenecks in developing a system, except in special cases like large file transfers or large numerical computations |
| 23:14 | <jwm> | that's more reason to incorporate it into the web |
| 23:15 | <jwm> | even more people would be using it if it was as simple as a javascript call in some library set |
| 23:16 | <AryehGregor> | Yes, well, good luck. But it's not going to happen. It makes much more organizational sense to have computation centralized. |
| 23:16 | <AryehGregor> | If anything, the trend du jour is the opposite -- cloud computing. |
| 23:16 | <jwm> | for some things it does |
| 23:16 | <AryehGregor> | Few things. Mostly, as I said and as simple observation confirms, illegal things. |
| 23:17 | <AryehGregor> | (which isn't necessarily a criticism, since there are unreasonable laws, particularly in some countries) |
| 23:17 | <jwm> | if you don't have a revenue model I don't think you should be without full access to the web |
| 23:17 | <jwm> | well I would like to actually bring non illegal activity to p2p heh |
| 23:17 | <jwm> | but ok, you're against p2p |
| 23:17 | <AryehGregor> | If you don't have money, there are a lot of things you can't do. No way around that. |
| 23:18 | <AryehGregor> | Not against it, just skeptical of the claims that it's broadly useful or likely to take off. |
| 23:18 | <AryehGregor> | It's had decades to take off, and hasn't. |
| 23:18 | <jwm> | well we don't know until we try on the web |
| 23:18 | <jwm> | yeah it has, but then again, for illegal activity |
| 23:18 | <AryehGregor> | Why would the web be qualitatively different from all the other places P2P has been tried? |
| 23:18 | <gsnedders> | It could've worked fine on a larger scale on the web by just, e.g., integrating BitTorrent clients into web browsers |
| 23:18 | <jwm> | because the web doesn't require you to install additional applications or setup ports |
| 23:18 | <jwm> | it's more accessible |
| 23:19 | <gsnedders> | (Which Opera does, on desktop.) |
| 23:19 | <jwm> | imagine if youtube was just an app you downloaded |
| 23:19 | <jwm> | it'd be used by far less people |
| 23:20 | <jwm> | I work in communications and nearly half of my clients run p2p software heh |
| 23:20 | <Philip`> | It'd be used by enough people to show it's probably a useful idea |
| 23:22 | <Philip`> | (People used special video plugins, so it was apparent that video was useful, so it got integrated into Flash, and more people used it, and it got integrated into HTML) |
| 23:22 | <jwm> | though I believe video codecs shouldn't be in the browser itself |
| 23:22 | <Philip`> | (Probably a similar path would have to be taken by other technologies that haven't proved themselves yet) |
| 23:22 | <jwm> | heh |
| 23:29 | <TabAtkins> | Anyone interested in CSS flexbox, could you take a look at http://www.xanthir.com/:wih and tell me if any of the examples could use improving? |
| 23:29 | <TabAtkins> | I've just created pictures for everything. |
| 23:32 | <jwm> | cool |
| 23:32 | <jwm> | looks nice |
| 23:32 | <jwm> | kind of how jquery does its docs |
| 23:33 | <TabAtkins> | It's the wonder of Markdown and fairly simple styling. |
| 23:33 | TabAtkins | loves Markdown so much, he'd marry it if he weren't already, and it were legal to marry a technical specification. |
| 23:35 | <othermaciej> | wouldn't it be neat if browsers could parse text/markdown into a DOM natively? |
| 23:35 | <TabAtkins> | It would indeed be neat. |
| 23:37 | <roc> | wouldn't it be neater if browses didn't have to build in support for everyone's favourite markup language? |
| 23:37 | <othermaciej> | probably also a security hole |
| 23:38 | <othermaciej> | browsers certainly don't *have* to |
| 23:39 | <jwm> | you can just a javascript component :) |
| 23:39 | <jwm> | load* |
| 23:39 | <Rik`> | TabAtkins: with the changes, is it possible to have all children occupy 1/(number of children) |
| 23:39 | <jwm> | javascript booklet |
| 23:40 | <TabAtkins> | Rik`: That's the default, actually. width:auto makes everyone have width:1fl, which distributes all space evenly. |
| 23:40 | <TabAtkins> | Rik`: If you're using padding and border and such, just pair that with a box-sizing:border-box too. |
| 23:41 | <Rik`> | even if one child has more text inside than the medium width ? |
| 23:41 | <TabAtkins> | jwm: I already made a js script that parses <pre>[[csv here]]</pre> into a table, so doing markdown is definitely possible too. |
| 23:41 | <TabAtkins> | Rik`: Yes, it doesn't pay attention to the contents unless it has to. |
| 23:41 | <Rik`> | cool |
| 23:41 | <TabAtkins> | Rik`: Like if you do width:calc(fit-content+1fl) |
| 23:42 | <Rik`> | that was something I couldn't do when I tried with webkit |
| 23:42 | <TabAtkins> | webkit's flexbox impl is kinda broken anyway. |
| 23:45 | <jwm> | I'd never heard of flexbox |
| 23:47 | <TabAtkins> | jwm: http://www.w3.org/TR/css3-flexbox is the current draft, which Firefox mostly implements and Webkit somewhat less implements. I'm rewriting it right now. |
| 23:48 | <TabAtkins> | (Don't really bother yourself reading the current draft. Just read mine. It's easier to understand.) |
| 23:48 | <Rik`> | any known effort for other layout modules ? |
| 23:49 | <TabAtkins> | After I'm done with Flexbox I'll be tackling Template Layout. |
| 23:50 | <Rik`> | oh cool ! |
| 23:51 | <jwm> | will this allow for proper vertical positioning? :) |
| 23:52 | <TabAtkins> | jwm: You can do proper vertical positioning with this, yeah. Proper abspos centering using flexes is a near-future goal too. ^_^ |
| 23:52 | <TabAtkins> | <div display:flex><div margin:1fl>I'm centered!</div></div> |
| 23:52 | <jwm> | ohh nice |
| 23:52 | <jwm> | tears of joy here |
| 23:52 | <jwm> | it'll only be IE 21 until it is implemented to standards though |
| 23:52 | <Rik`> | right now, you have box-pack and box-align, right ? |
| 23:53 | <TabAtkins> | Rik`: No, those are dropped in favor of directly using flex units. |
| 23:53 | <Rik`> | when I say right now, I meant in browsers |
| 23:53 | <jwm> | I never liked using em in browsers |
| 23:54 | <jwm> | I'm actually trying to come up with some html5/jquery inspired web designs right now |
| 23:54 | <jwm> | to use for clients |
| 23:54 | <TabAtkins> | Rik`: Yeah, right now the current draft is implemented in FF and webkit. |
| 23:54 | <jwm> | I'd like to make stuff scale |
| 23:54 | <jwm> | where the entire design scales to the window size |
| 23:54 | <TabAtkins> | So box-pack and box-align must be used. |
| 23:56 | <Rik`> | TabAtkins: Alternately, you may use flex in the display property, which is equivalent to declaring block flex |
| 23:57 | <Rik`> | declaring block flex ? |
| 23:57 | <TabAtkins> | Rik`: Sorry, that's a leak from my "split display" proposal. |
| 23:57 | <TabAtkins> | Where display becomes a shorthand for display-outside and display-inside. |
| 23:57 | <Rik`> | I did get that, I didn't get the block: flex part |
| 23:58 | <TabAtkins> | Ah, no, I meant "display:block flex" |
| 23:58 | <TabAtkins> | Let me go revise that. |
| 23:59 | <TabAtkins> | Better now? |
