| 01:14 | <mcarter> | Hixie, whats the rationale behind the framing of each payload in the TCPConnection spec? |
| 09:57 | <zcorpan_> | hmm, perhaps the offline-webapps should expand a bit what the difference is between localStorage["status"] = "Idling."; and window["status"] = "Idling."; |
| 12:20 | Dashiva | wonders if anyone has been offended yet |
| 12:32 | <takkaria> | Dashiva: what by? |
| 12:37 | <Dashiva> | 'primary' |
| 12:37 | <jgraham__> | By hsivonen suggesting that cameras are mainly designed to take pictures, you mean? |
| 12:37 | <Dashiva> | Yes. The nerve! |
| 12:41 | <takkaria> | I'm glad someone finally said it, Rob scared me off from replying to the alt thread again |
| 12:43 | <Dashiva> | takkaria: It was brought up once before, and it got some pretty nasty replies |
| 13:44 | <hsivonen> | sometimes the alt thread makes me wonder if I'm suffering from metacognitive miscalibration |
| 13:44 | <hsivonen> | I take the comments above as indication that I might not be. thanks |
| 13:50 | <MikeSmith> | I see mention of "fast-path DOM" at http://wiki.mozilla.org/Mozilla_2/Work_List |
| 13:51 | <MikeSmith> | wondering just what that might mean |
| 14:36 | <Philip`> | MikeSmith: Judging by e.g. http://markmail.org/message/sudwtkjlsrkf47zx it seems to be about optimising the JS-DOM interface by avoiding the XPCOM overhead |
| 14:37 | <MikeSmith> | Philip`: cool, thanks for the pointer |
| 14:44 | <hsivonen> | I think I've now said all I have to say about alt |
| 14:45 | <hsivonen> | now I'm just going to wait and see what if anything WAI is going to tell the HTML WG to do |
| 14:49 | <MikeSmith> | hsivonen: hope it will free up more time for you work on validator.nu |
| 14:49 | <MikeSmith> | btw, do you have a v.nu TODO list online somewhere? |
| 14:49 | <hsivonen> | MikeSmith: there's the bugzilla, but it doesn't have all TODO on it |
| 14:50 | <hsivonen> | and yeah, I want to do a lot more coding |
| 14:51 | <MikeSmith> | what's the status on the work you've been doing to improve the error output from jing? |
| 14:51 | <MikeSmith> | is that done, or you still need to do more work? |
| 14:51 | <hsivonen> | MikeSmith: the status is that I'm blocking on George Bina pushing out a new upstread snapshot |
| 14:51 | <MikeSmith> | ok |
| 14:52 | <MikeSmith> | yeah, I remember you saying that a while back |
| 14:52 | <MikeSmith> | what's the current likelihood of George folding back in your changes upstream? |
| 14:53 | <hsivonen> | MikeSmith: I think the likelihood isn't too great |
| 14:53 | <hsivonen> | we'll see |
| 15:06 | <hsivonen> | I think not replying on the mailing list is a key skill |
| 15:07 | <hsivonen> | the trouble I have is that I'm not sure how badly the alt stuff is going to come back and bite Validator.nu if I don't explain what I'm doing |
| 15:07 | <hsivonen> | that is, that I'm not seeking to be evil here |
| 15:18 | <Dashiva> | Geez |
| 15:19 | <hsivonen> | Dashiva: ? |
| 15:19 | <Dashiva> | I'm pondering a dilemma. What would they choose between a) always perfect alt, everything else in wcag intentionally violated and b) optional alt |
| 15:20 | <takkaria> | they would spin round the question such that they didn't have to answer. :) |
| 15:21 | <Dashiva> | Yes, I expect most answers would be "We want both, and a pony" |
| 15:21 | <hsivonen> | Dashiva: and you'd be immoral for asking :-/ |
| 15:22 | <Dashiva> | That's okay, I was cited in the original formal complaint, so my innocence is lost |
| 15:23 | <hsivonen> | I think it isn't okay. it creates and unhealthy working environment and polarizes people needlessly |
| 15:29 | <Dashiva> | I guess I still have problems taking it 100% seriously. It's just so... absurd... at times. |
| 15:32 | <hsivonen> | perhaps I'm taking it too seriously, but it really bothers me that discussions about different ways to address an issue have a constant undertone that people who try to do something they believe is for the better are immoral, closet bigots, etc. |
| 15:36 | <Philip`> | hsivonen: I don't think anyone ever seeks to be evil - they just see themselves from a twisted perspective, and then act in a way that they consider perfectly rational :-) |
| 15:37 | <Dashiva> | We definitely need people like you (and more of them). |
| 15:38 | <Dashiva> | Can't let it turn into a war of attrition, where they proclaim "victory" just because everyone gives up trying to reason with them |
| 15:40 | <jgraham__> | Gez would be a lot more convincing if he didn't argue in tautologies |
| 15:41 | <jgraham__> | "<img> without alt is defined as an incomplete structure incomplete structures are bad, therefore img without alt is bad" |
| 15:41 | <jgraham__> | is totally logical but pretty pointless |
| 15:45 | <takkaria> | I think some people just don't accept presenting an image as a valid thing to want to do in HTML |
| 15:52 | <hsivonen> | jgraham__: whether it is bad isn't the whole story. it's whether the alternative is worse |
| 15:55 | <Dashiva> | Haha. I just imagined a conspiracy theory that would explain it all: They have to make sure non-conforming documents are horribly inaccessible, because if non-conforming documents were possible to consume, it would hurt the push for mandatory accessibility. |
| 16:08 | <jgraham__> | hsivonen: I know. |
| 16:08 | <hsivonen> | jgraham__: yeah. my point was that the tautology wasn't the only problem |
| 16:09 | <jgraham__> | Sure. |
| 16:10 | <jgraham__> | FWIW I'm not sure that the Authoring Tool Accessibility Guidelines are very helpful |
| 16:11 | <hsivonen> | I should probably reread them, but I don't feel like doing it right now. |
| 16:11 | <jgraham__> | For example they seem to encourage KompoZer-style dialogs that make it very easy to generate bad alt text |
| 16:12 | <MikeSmith> | hsivonen: I'm not sure I'd say you're taking it too seriously, but I would wonder whether the time and energy any and everybody is spending on it might be more productively spent on other things |
| 16:12 | <jgraham__> | Although they do say that it should be possible to check the accessibility of a document, they don't seem to take the stance that this should be a human-led process |
| 16:13 | <hsivonen> | MikeSmith: I think it's clear that it would be more productive for *everyone* to drop the issue. |
| 16:13 | <hsivonen> | MikeSmith: the part I'm worried about is only some people dropping it and design getting done by being out-shouted |
| 16:15 | <MikeSmith> | hsivonen: whatever decision might possibly be reached right now would not be the final decision about it anyway |
| 16:15 | <jgraham__> | That's a legitimate fear given that e.g. John Foliot has stated that he views the situation as trench warfare |
| 16:15 | <MikeSmith> | I mean I can see this discussion will get re-opened when we go to last call |
| 16:18 | <hsivonen> | the cost of having the courtesy to discuss stuff instead of just being rude and not talking certainly has gone up lately |
| 16:19 | <BenMillard> | I have a message rule which automatically marks as read all messages with subject line mentioning any of various keywords |
| 16:19 | <BenMillard> | sadly, I added "alt" to that list a few months ago |
| 16:19 | <MikeSmith> | it seems to me it was well known (or should have been well known) that this and other contentious changes to authoring-conformance criteria related to accessibility would provoke just the kinds of reactions just we are seeing now |
| 16:21 | <Dashiva> | MikeSmith: It would indeed be wonderful if someone could convince them to leave the issue alone at least until LC |
| 16:22 | <MikeSmith> | Dashiva: would be wonderful, but it's not going to happen, of course |
| 16:25 | <BenMillard> | MikeSmith, what are you getting at with the "it was well known" comment? it looks like you're suggesting such changes set out to be controversial rather than being genuine attempts at good design...although I may be reading too much between the lines |
| 16:25 | <MikeSmith> | BenMillard: you're reading too much between the lines |
| 16:26 | <BenMillard> | phew :) |
| 16:27 | <BenMillard> | I've not seen much review of data table accessibility since that section got updated |
| 16:28 | <BenMillard> | maybe people like what we did (?!) |
| 16:29 | <Dashiva> | Would be nice |
| 16:31 | <BenMillard> | Lachy & jgraham__, does your presentation show people how to contribute productively to HTML5's development? |
| 16:31 | <jgraham__> | BenMillard: I want to cover that for sure |
| 16:32 | <jgraham__> | I was planning to use you as an example :) |
| 16:32 | <BenMillard> | jgraham, cool :) |
| 16:34 | <MikeSmith> | jgraham__: does it give examples of how to not contribute productively? ;) |
| 16:34 | <jgraham__> | MikeSmith: Were it only that I were that brave |
| 16:34 | <MikeSmith> | heh |
| 16:34 | <BenMillard> | I have a dream where I give a presentation about the path I took from my bad initial reactions to HTML5, through the process of studying data tables and hanging around here in IRC, meeting WHATWG people at the W3C thing last year, to the understanding I have now for how the work is done |
| 16:35 | <MikeSmith> | jgraham__: It reminds me of something .. here in Japan, there are many signs in public (on trains and such) that have cartoon illustrations of bad/discouraged behavior |
| 16:36 | <MikeSmith> | e.g., illustration of a woman doing her makeup on the train |
| 16:36 | <MikeSmith> | and statement being basically, Don't do this. |
| 16:36 | <jgraham__> | MikeSmith: I remember those; when I was in Japan we wore our rucksacs on the train which was a social faux-pas |
| 16:37 | <jgraham__> | Someone pointed us at the pictures |
| 16:37 | <MikeSmith> | I wonder if we could come up with some illustrations of bad behavior that we'd like to discourage in discourse about HTML |
| 16:37 | <MikeSmith> | jgraham__: yeah, problem with rucksacs is that the trains are so crowded |
| 16:37 | <MikeSmith> | so bags ended up bumping into peopel |
| 16:38 | <BenMillard> | MikeSmith, flamethrowers, trolls, sheep and bandwagons come to mind? |
| 16:38 | <MikeSmith> | BenMillard: ah yes |
| 16:38 | <jgraham__> | MikeSmith: The train wasn't particularly crowded and I was pretty annoyed that people seed to think it was OK to just push me out of the way when it was easier to go round me |
| 16:38 | <jgraham__> | But such is the joy of different cultures |
| 16:38 | <MikeSmith> | I also imagine a "painting the bikeshed" illustration |
| 16:38 | <jgraham__> | Especially when you're sleep deprived |
| 16:39 | <jgraham__> | s/seed/seemed/ |
| 16:39 | <MikeSmith> | jgraham__: ah yeah, the pushing-out-of-the-way thing is hard to get used to |
| 16:40 | <MikeSmith> | anywhere else, somebody doing it would likely end up getting a punch in the nose |
| 16:40 | <jgraham__> | heh |
| 16:41 | <MikeSmith> | One way i've found to educate people about it is to push back |
| 16:41 | <MikeSmith> | but unfortunately, some of the main culprits are old ladies |
| 16:41 | <MikeSmith> | so that doesn't work so well for that case |
| 16:42 | <MikeSmith> | or rather, works a bit too well |
| 16:43 | <hsivonen> | MikeSmith: I once considered writing observations about how to really annoy CSS WG members on www-style |
| 16:43 | <jgraham__> | Japan is a fascinating place. It has very different cultural norms to almost everywhere else in the "developed world" |
| 16:43 | <hsivonen> | MikeSmith: but then I decided not to write it considering that my wannabe article from years ago offended people |
| 16:44 | <MikeSmith> | well, some people are just waiting to be offended, actively looking around for something to get offended about |
| 16:44 | <MikeSmith> | so sometimes nice to oblige them |
| 16:45 | <Philip`> | Other times people just write offensive things ;-) |
| 16:45 | <MikeSmith> | with some token of offensiveness to keep them busy |
| 16:45 | <Dashiva> | Is the css wg still trying to decide if "public" includes logging IRC? |
| 16:45 | <annevk> | No, that's settled now |
| 16:46 | <jgraham__> | I remember that one of the www-style flamewars made me realise that working groups / mailing lists for WGs are not able to deal with deliberate disruptive influences well. |
| 16:47 | <jgraham__> | It's too easy to tarpit everyone |
| 16:47 | <MikeSmith> | jgraham__: aye |
| 16:48 | <jgraham__> | BenMillard (or anyone I guess): do you have anything particular about the community that you think should go in the @media talk? |
| 16:48 | <MikeSmith> | I guess that issue you mentiong is one of the biggest issues with trying to do purely consensus-driven decision making |
| 16:49 | <Dashiva> | BenMillard: Any chance of that dream becoming reality? |
| 16:50 | <jgraham> | MikeSmith: The interesting thing is that the WHATWG sidestepped the issue by not maintaining an illusion of decision making being consensus driven |
| 16:50 | <MikeSmith> | jgraham: the record of browser implementation progress that has been made with parts of HTML5 seems worth mentioning in your talk |
| 16:50 | <MikeSmith> | how long is the talk? |
| 16:50 | <jgraham> | MikeSmith: 1 hour. I think we'll aim for about 45+15 |
| 16:51 | <jgraham> | (the 15 being for questions) |
| 16:51 | <jgraham> | So the plan is: |
| 16:51 | <MikeSmith> | have you seen the slides I did for my presentation at XTech? |
| 16:51 | <jgraham> | 1. Introduction |
| 16:51 | <jgraham> | 2. Design Principles |
| 16:51 | <jgraham> | 3. Demo how HTML5 features can be used in projects today |
| 16:51 | <jgraham> | 4. Communit |
| 16:52 | <jgraham> | y |
| 16:52 | <jgraham> | MikeSmith: Yeah, I saw the slides. We will crib from them for 3 I guess |
| 16:52 | <MikeSmith> | cool |
| 16:53 | <hsivonen> | jgraham: I have a couple of community observations both of which may not be politically correct |
| 16:53 | <MikeSmith> | I think the record of implementation progress for HTML5 has been exceptional compared to other specs |
| 16:53 | <MikeSmith> | and that is always worth mentioning |
| 16:54 | <hsivonen> | jgraham: 1) the WHATWG community formed when WHATWG was seen as illegitimate by many, so it attracted people who looked at the substance instead of well-known logo |
| 16:54 | <MikeSmith> | especially to counter the "The spec will take 15 years to complete??!!" stuff |
| 16:54 | <hsivonen> | jgraham: 2) "running code" process vs. consensus process |
| 16:55 | <MikeSmith> | jgraham: I also have some general HTML5 overview slides here: |
| 16:56 | <MikeSmith> | http://www.w3.org/2008/Talks/04-24-smith/ http://www.w3.org/2008/Talks/04-24-smith/index.pdf |
| 16:56 | <MikeSmith> | feel free to crib from those too, if you find anything useful |
| 16:56 | <jgraham> | MikeSmith: Thanks |
| 16:56 | <MikeSmith> | when I made them, I felt free to crib from previous presentations that annevk and zcorpan had done :) |
| 16:57 | <jgraham> | hsivonen: I agree with both your observations |
| 16:57 | <hsivonen> | jgraham: do you have my ten points on file? |
| 16:57 | <jgraham> | hsivonen: No, I don't think so |
| 16:58 | <hsivonen> | jgraham: http://pastebin.ca/992541 |
| 16:58 | <BenMillard> | Dashiva, seems unlikely as nobody has ever heard of me :| |
| 16:59 | <Dashiva> | If you make it, they will ask ;) |
| 17:01 | <hsivonen> | regarding consensus process, see the first paragraph of http://en.wikipedia.org/wiki/Management_by_perkele (the content in the third para and onwards is questionable) |
| 17:01 | <BenMillard> | jgraham__, "09 You can participate." giving specific URLs, such as to Hixie's message with 6 steps to joining HTMLWG, as well as an irc:// URI for this channel and other useful logistical things would be valuable |
| 17:02 | <BenMillard> | until you're already in the community, finding where the community takes place is almost impossible |
| 17:03 | <BenMillard> | contributions don't need to be Grand Scientific Experiments; Mythbusters "science with a small s" is infinitely more useful than speculation |
| 17:04 | <Dashiva> | Just documenting status quo can be helpful, and that's largely just someone having to put in the hours |
| 17:05 | <BenMillard> | Dashiva, exactly |
| 17:07 | <MikeSmith> | hsivonen: thanks for the pointer. I remember somebody mentioning this term when we were at XTech |
| 17:12 | <BenMillard> | by the way, my detailed notes from Sight City 2008 are coming together here: http://projectcerbera.com/blog/2008/05#day10 |
| 17:12 | <BenMillard> | I've got to document 9th May 2008 then it's done, I think |
| 17:13 | <BenMillard> | people I talked to mostly had no idea HTML5 was being developed |
| 17:13 | <BenMillard> | they seemed happy about us wanted them to be involved, though |
| 17:23 | <Philip`> | BenMillard: You have a link to http://projectcerbera.local/ which doesn't work |
| 17:35 | <jgraham> | hsivonen: Thanks for the links |
| 18:06 | <MikeSmith> | jgraham: another suggestion to consider, fwiw: was thinking about whatever publisher it was that put out the "Missing Manual" series of tech books |
| 18:07 | <MikeSmith> | the HTML5 spec could be seen as the "missing" spec for HTML |
| 18:07 | <Philip`> | (Apparently Pogue Press / O'Reilly) |
| 18:07 | <MikeSmith> | the spec defines how conformant implementations must behave |
| 18:08 | <MikeSmith> | the spec that UA/browser implementors have been missing for the last 10+ years |
| 18:08 | <Philip`> | That sounds like a good idea - publish the HTML5 spec as a book, such that it's slightly cheaper than printing hundreds of pages on your own laser printer, and use the money to fund future development |
| 18:09 | <jgraham> | Philip`: Who would buy the HTML spec as a book? |
| 18:09 | <Philip`> | jgraham: People who would otherwise print it out, and realise that the book is cheaper |
| 18:09 | <jgraham> | Philip`: e.g.? |
| 18:09 | <Philip`> | jgraham: I believe I've heard Chris Wilson talking about the spec as being an inch thick, so I assume he's printed it |
| 18:10 | <annevk> | marcosc prints specs |
| 18:10 | <jgraham> | Oh well. |
| 18:10 | <jgraham> | Still a market of 10-100 people isn't really going to make the book cheape |
| 18:10 | <jgraham> | r |
| 18:11 | <annevk> | true |
| 18:11 | <jgraham> | I think the time is about right for the first html5-specific book aimed at web-devs though |
| 18:12 | Philip` | remembers there was a Perl 6 book two or three years ago |
| 18:13 | <Philip`> | That's already totally obsolete, and Perl 6 isn't even implemented yet, so HTML 5 is in a far better position |
| 18:14 | <MikeSmith> | there are services that can print paperback books these days on-demand |
| 18:16 | <MikeSmith> | hey, there are some people that hate some of the ideas in the HTML5 spec so much (e.g., handling on non-well-formed content) that maybe there is a market for printing the spec on rolls of toilet paper |
| 18:17 | <Philip`> | The HTML5 spec-annotation system should be extended so you can vote for how much you like a section on a scale of 1 to 5 |
| 18:18 | <Philip`> | then we can automatically extract the most suitable sections for those applications such as toilet paper |
| 18:18 | <MikeSmith> | heh |
| 18:18 | <MikeSmith> | we're doing some great productive brainstorming here |
| 18:20 | <MikeSmith> | but I think Doug Crockford is way ahead of us all, as usual |
| 18:20 | <MikeSmith> | with his JavaScript: The Good Parts book |
| 18:21 | <MikeSmith> | great cross-selling opportunity for O'Reilly/Amazon or whoever |
| 18:21 | <MikeSmith> | JavaScript: The Good Parts book bundled with JavaScript: The Bad Parts toilet paper |
| 18:22 | Philip` | thinks that's an awesome idea |
| 22:35 | <Hixie> | mcarter__: otherwise you don't know when to fire the event on the other side |
| 22:36 | <mcarter__> | Hixie, why not fire an event every time any data comes through? |
| 22:40 | <Hixie> | BenMillard: you can always give your talk as a video blog and put it on youtube :-) |
| 22:41 | <BenMillard> | Hixie, interesting idea |
| 22:41 | <Hixie> | mcarter__: because if side A sends two strings to side B, you want two events on side B |
| 22:41 | <Hixie> | mcarter__: not one, or three |
| 22:42 | <BenMillard> | Hixie, it's a question of time and money and equipment, where I have little of any |
| 22:42 | <mcarter__> | Hixie, yeah, but protocols generally already take care of framing... and if you're inventing a new protocol, you can just make it \r\n based if you want simple framing |
| 22:43 | <mcarter__> | Hixie, what i mean, is that our transport protocol can not have framing, and you can effectively do the same thing in your client by parsing the incoming data |
| 22:43 | <Hixie> | mcarter__: if you frame on newlines then you can't send multiline text, and TCP/IP doesn't have framing at a more useful level than "byte" as far as I can tell. |
| 22:43 | <Hixie> | mcarter__: i don't want to require that every js author implement a byte buffer to collect strings, that's just annoying fluff that we should take care for them |
| 22:44 | <Hixie> | mcarter__: adding framing also means we can later expand the protocol to support both strings and arbitrary binary blobs |
| 22:44 | <Hixie> | mcarter__: e.g. when we implement Blobs as proposed in Web API |
| 22:44 | <Philip`> | BenMillard: In case you don't have psychic powers and didn't see it while you weren't here: http://krijnhoetmer.nl/irc-logs/whatwg/20080517#l-368 |
| 22:45 | <mcarter__> | Hixie, I can imagine a scenario where you want to modify an existing server, say a jabber server, just by making it understand the initial handshake, then leaving the rest of its code untouched. With the framing added to the data transport, its not as easy as that |
| 22:47 | <Hixie> | mcarter: that wasn't one of the use cases i was considering, but in any case if you're going to that kind of lengths adding a simple translator on the end isn't a big deal |
| 22:47 | <mcarter> | perhaps not |
| 22:49 | <BenMillard> | Philip`, oops, thanks for that! |
| 22:50 | <Hixie> | http://www.openajax.org/runtime/wiki/Feature_Requests_Summary_Page and the pages it link to are hard to read for some reason |
| 22:50 | <Hixie> | i don't know why |
| 22:51 | <Hixie> | i think it just has too muh fluff |
| 22:51 | <Hixie> | instead of being simple and direct |
| 22:51 | <Hixie> | i just feel like i'm wading in mud when i read it |
| 22:51 | <Hixie> | which is really sad |
| 22:51 | <Hixie> | because this would be really important feedback |
| 22:51 | <BenMillard> | Hixie, the "Active Feature Requests" duplicates the first column in the third column, only difference being the third column items are linked |
| 22:52 | <Hixie> | yeah, for example |
| 22:52 | <Hixie> | and the wiki pages themselves have about 50% boilerplate |
| 22:53 | <Hixie> | so it's like wading through generic material to get to the actual data |
| 22:53 | <BenMillard> | Hixie, they use <th> for the normal headers and <td colspan><strong> for spanned ones, annoyingly :( |
| 22:54 | <BenMillard> | I see ordered lists written using (1) foo (2) bar rather than <ol> |
| 22:54 | <BenMillard> | I always find that hard to read |
| 22:54 | <mcarter> | are those artifacts of mediawiki, or are they deliberately putting those in? |
| 22:54 | <Hixie> | yeah |
| 22:55 | <BenMillard> | oh, just realised the entire 2nd column in that table just says "coming later" :D :D |
| 22:55 | <BenMillard> | so really, it's just a set of <h3><ul> and not a table at all (yet) |
| 22:56 | <BenMillard> | Hixie, you aren't going mad. that site has made unhelpful choices for presenting its content |
| 22:56 | <Hixie> | glad to hear it's not just me :-) |
| 22:59 | <BenMillard> | Hixie, I like how it talks about "heroic JavaScript" in contrast to the "vastly insufficient" CSS: http://www.openajax.org/runtime/wiki/Better_UI_Layout |
| 22:59 | BenMillard | imagines Superman's outfit with a J prefixed to the S |
| 23:06 | <Hixie> | heh |
| 23:52 | <Hixie> | i think <title> should be made optional for documents intended to be used inside iframes |
| 23:59 | <takkaria> | sounds sensible |