| 03:30 | <Hixie> | i need a complicated but small table for summary="" stuff |
| 10:38 | <Hixie> | hsivonen: "if an old draft" -> "of an old draft" |
| 10:47 | <Hixie> | well |
| 10:47 | <Hixie> | shockingly, it seems no two browsers interoperate on the meaning of history.back(). |
| 10:47 | <Hixie> | i would never have guessed that |
| 10:53 | <takkaria> | excellent |
| 10:53 | <takkaria> | given that it's used all over the place, that's quite surprising |
| 11:02 | <Hixie> | http://www.hixie.ch/tests/adhoc/dom/level0/history/ |
| 11:03 | <Hixie> | i'm gonna spec IE's behaviour |
| 11:03 | <Hixie> | (since the spec doesn't match anyone currently) |
| 11:09 | gsnedders | stretches and heads for lunch, if anyone wants to join him |
| 11:09 | <Lachy_> | Hixie, why do the tests need to be cached and run again? I just tried them in Opera and Firefox and got the same result both times each test was run in each browser |
| 11:17 | <Hixie> | if you're on a slow connection you'll get unreliable results if you don't let them get cached first. |
| 11:21 | <Lachy_> | ok. I'm on a fast conncetion, so didn't affect me |
| 12:51 | <hsivonen> | Hixie: typo fixed. thanks |
| 12:55 | <Lachy> | hsivonen, for the question "What’s the doctype for XHTML5 documents?" you might want to mention that <!DOCTYPE html> is also a well-formed DOCTYPE that can be used for the purpose of producing a polyglot document |
| 12:57 | <hsivonen> | Lachy: I don't want to encourage people to write polyglot documents |
| 12:59 | <Lachy> | ok |
| 13:12 | <jgraham> | hsivonen: Nice FAQ. Any progress on being syndicated on planet.mozilla? Your blog deserves more exposure... |
| 13:13 | <hsivonen> | jgraham: planet is bug https://bugzilla.mozilla.org/show_bug.cgi?id=502224 |
| 13:14 | <jgraham> | http://article.gmane.org/gmane.comp.python.general/629043 and following are quite interesting |
| 13:15 | <jgraham> | (and we got a patch out of it too, which I haven't really looked at hard yet but shows that at least some of the people are awesome some of the time) |
| 13:26 | <gsnedders> | http://mail.python.org/pipermail/python-list/2009-July/718802.html saner archives |
| 13:26 | <gsnedders> | (like, with next message links) |
| 13:27 | <jgraham> | gsnedders: Ah. Useful. I wonder hy gmane sucks so much |
| 13:27 | <jgraham> | *why |
| 13:27 | <gsnedders> | From that original post, I guess I probably ought to read the thread sometime (but I guess not now :)) |
| 13:35 | <hsivonen> | jgraham: if you want more perf, I suggest splitting up the data state the way it's split in the V.nu tokenizer |
| 13:35 | <Dashiva> | Link to hsivonen's FAQ? |
| 13:36 | <hsivonen> | Dashiva: http://hsivonen.iki.fi/xhtml2-html5-q-and-a/ |
| 13:36 | <Dashiva> | Thanks |
| 13:36 | <jgraham> | hsivonen: How is that? |
| 13:37 | jgraham | actually wants a non-broken trunk more than perf right now |
| 13:37 | <hsivonen> | jgraham: you don't need to check the content model flag per character if you bake the content model flag into the tokenizer state |
| 13:37 | <jgraham> | Oh so you have dataStateCdata and dataStateRcdata and so on? |
| 13:37 | <hsivonen> | jgraham: yeah |
| 13:38 | <gsnedders> | I know that in the PHP impl. that the PHP DOM extension (which is written in C) is the bottleneck for medium to large HTML documents! |
| 13:46 | <hsivonen> | gsnedders: what kind of storage does it use for the children of an element? |
| 13:47 | <gsnedders> | hsivonen: One with non-constant perf. characteristics. More exactly, I don't know. |
| 13:47 | <hsivonen> | gsnedders: ok |
| 13:52 | <gsnedders> | hsivonen: Well, actually, accessing it I think is constant, but appendChild is non-constant |
| 13:52 | <hsivonen> | gsnedders: it seems to me that you want linked lists of siblings and child arrays are bad |
| 13:54 | <gsnedders> | hsivonen: Yeah, myself and ezyang were intending on trying to write something quicker, though for distributability purposes we need something that runs in interpreted code, so it'll be fun if we get that running quicker than the DOM extension in C |
| 13:56 | <gsnedders> | IIRC it's based upon a very old gdome |
| 14:16 | <MikeSmith> | hsivonen: you there? got a problem with v.nu update |
| 14:17 | <hsivonen> | MikeSmith: I'm here |
| 14:17 | <MikeSmith> | hsivonen: I'm rebuilding now to see if I can reproduce it |
| 14:18 | <MikeSmith> | maybe same problem that David Eisenberg reported |
| 14:19 | <hsivonen> | the classcast exception can usually be remedied by rerunning python build/build.py all |
| 14:21 | <MikeSmith> | hsivonen: this may be a different error |
| 14:21 | <MikeSmith> | anyway, I'm re-running now |
| 14:22 | <MikeSmith> | yeah, this seems different |
| 14:22 | <MikeSmith> | hang on, I'll post the log |
| 14:24 | <MikeSmith> | hsivonen: http://people.w3.org/mike/v.nu/build-log.txt |
| 14:25 | <hsivonen> | MikeSmith: hmm. I may have broken the tree by failing to commit all local changes |
| 14:26 | <MikeSmith> | hsivonen: that's what I suspected might be the case |
| 14:26 | <MikeSmith> | easy thing to do |
| 14:26 | <hsivonen> | MikeSmith: try updating now |
| 14:26 | <hsivonen> | MikeSmith: sorry |
| 14:26 | MikeSmith | retries |
| 14:30 | <hsivonen> | http://paulmwatson.wordpress.com/2009/07/06/im-finding-the-technical-vs-mar/ |
| 14:30 | <MikeSmith> | hsivonen: thanks for the checkin -- working as expected now |
| 14:38 | <Dashiva> | "I use XHTML because it simply feels good." |
| 14:41 | <MikeSmith> | words have failed, I guess we need to start drawing pictures |
| 14:41 | <MikeSmith> | maybe somebody could do a comic strip |
| 14:44 | <Dashiva> | The earlier post repeats the "mobile devices are too weak to parse HTML, and XHTML is much easier to parse" story |
| 14:44 | <hsivonen> | oh I forgot to have a Q and A about that |
| 15:06 | <Lachy> | hsivonen, are there any issues you're aware of with the html5.enable pref not showing up in the Linux build of the Firefox nightlies? |
| 15:06 | <jgraham> | Lachy: wfm (but you can create the pref if it doesn't exist already) |
| 15:06 | <Lachy> | I don't have Linux myself, but someone here at work wanted to try it and apparently, the pref isn't available for him |
| 15:07 | <Lachy> | jgraham, yeah, that's what I said to do on the internal discussion list |
| 15:30 | <jgraham> | http://simonwillison.net/2009/Jul/6/xhtml/#c47115 appears to be an "interestingly" phrased Firefox bug report |
| 15:55 | <MikeSmith> | jgraham: I think dude must be using the "Convert glyphs into boxes with little numbers inside them" extension for Firefox |
| 15:55 | <MikeSmith> | I think it's one of the developer features |
| 15:57 | <MikeSmith> | as far as I can see, he complaint is that the little-number-boxes with lines underneath them should be a slightly different shade of blue than the one they show up in |
| 15:58 | <MikeSmith> | and he has a point there -- the blue not quite right |
| 15:58 | <MikeSmith> | so, good eye, John Handelaar |
| 15:59 | <MikeSmith> | a lot of people don't catch those kind of subtleties |
| 15:59 | <MikeSmith> | it takes a real pro, like him |
| 16:19 | <gsnedders> | eighty4: you were saying before about coming around this evening to pick up your stuff. still got any intention of doing so? |
| 16:22 | jgraham | assumes gsnedders just wants help finding the washing machine |
| 16:22 | gsnedders | would like to deny that, but it is in part true |
| 16:23 | <gsnedders> | One way or another, this _will_ end tonight. |
| 16:23 | <gsnedders> | (I expect that's a misquote, though…) |
| 16:26 | <Dashiva> | It's a bit funny seeing people go "This is silly, surely we'll be working on HTML8 before 2022" seeing as how HTML4 was over a decade ago. Why the sudden change in expectancies :) |
| 16:27 | <Dashiva> | hsivonen: Is "Unnofficial" typoed on purpose? |
| 17:44 | <eighty4> | gsnedders: is it ok if I do it tomorrow? |
| 17:44 | <eighty4> | gsnedders: I'm a bit tired :) |
| 17:44 | <eighty4> | gsnedders: and I can help you find the washing machine tomorrow :) |
| 17:45 | <eighty4> | I'm free all day! |
| 17:45 | <eighty4> | absolutly free! I have nothing at all to do |
| 17:45 | eighty4 | hides in the corner... |
| 17:45 | <eighty4> | having nothing to do is scary |
| 18:06 | <Dashiva> | No matter how much data against @summary is discredited, I think jarring that there's absolutely no data (regardless of credibility) in favor of it |
| 18:15 | <gsnedders_> | eighty4: yeah, I guess |
| 18:16 | <gsnedders> | hmm, oh well, if me at work does want to persist, then so be it. |
| 18:17 | <gsnedders> | eighty4: Also, see Habari #580 for something to do :P |
| 18:19 | <eighty4> | oh, ETags :) |
| 18:19 | <gsnedders> | That's take a fair amount of work to impl. well :) |
| 18:19 | <eighty4> | gsnedders: you're working tomorrow? |
| 18:19 | <gsnedders> | eighty4: yeah |
| 18:20 | <eighty4> | then just let me know in here when you'll be home |
| 18:20 | <gsnedders> | eighty4: I could even let you know when I leave the office, or when I arrive at the office :P |
| 18:21 | <eighty4> | I'll be in here and/or in #habari most of the day. |
| 18:22 | <gsnedders> | As you've probably noticed, I'm not in #habari at work |
| 18:22 | <eighty4> | I might walk out to IKEA or something but that'll be in the morning |
| 18:23 | <eighty4> | no, haven't noticed that :) I'm not actually active in any channels durin working hours |
| 18:23 | <gsnedders> | I'm just not in them :) |
| 18:24 | <gsnedders> | (I am here, but this is vaguely work related) |
| 18:24 | <eighty4> | it's very work related |
| 18:24 | <eighty4> | :) |
| 18:25 | <eighty4> | I have cheez doodles |
| 19:26 | <gsnedders> | Hixie: http://www.w3.org/mid/Pine.LNX.4.62.0907060805240.1053⊙hdc — with what locale? :P |
| 20:05 | <tantek> | gsnedders, Hixie - I still think a "vote" on a specific feature sets a bad precedent. science isn't democracy. |
| 20:06 | <gsnedders> | tantek: I was more attempting to make a joke |
| 20:45 | <smedero> | hrm. can someone remind me why I would get an "unexpected-doctype" exception in html5lib. |
| 20:46 | <smedero> | (for some reason I thought it didn't give a crap about the doctype...) |
| 20:46 | <gsnedders> | smedero: Because it's not at the start of the document. But as an exception? |
| 20:48 | <sayrer_> | I actually don't want a vote either |
| 20:49 | <smedero> | I should say it was a html5lib.html5parser.ParseError |
| 20:49 | <smedero> | @gsnedders but yes, that makes sense. |
| 20:49 | <sayrer_> | I doesn't seem like anything has changed from a technical perspective |
| 20:49 | <sayrer_> | It doesn't |
| 20:50 | <gsnedders> | My computer doesn't think 1 + 1 = 2 anymore? |
| 20:50 | gsnedders | goes back to less useful stuff :P |
| 20:50 | <sayrer_> | how does it do on 5 > 2 |
| 20:50 | <gsnedders> | False. Maybe it has a virus or something. |
| 20:52 | <Hixie> | tantek: i agree, but sam hasn't put forward any other suggestions for how to resolve this, and he refuses to make an actual determination himself |
| 20:56 | <tantek> | Hixie, from briefly reading the recent email thread that gsnedders linked to, it appeared that there was *some* progress being made in the level of dialog (more scientific) with regards to the methodology of analysis of summary. |
| 20:57 | <gsnedders> | tantek: I think what progress there is is minimal. |
| 20:57 | <sayrer_> | gsnedders: yes, nothing seems to warrant a line item vote on the matter |
| 20:57 | <sayrer_> | that is supposed to be a last resort at the w3c |
| 20:58 | <gsnedders> | Basically I don't think we're going to get much more compromise from either side, and I think we're more or less stuck, with the arguments going round in circles. |
| 20:59 | <sayrer_> | I don't think we're stuck |
| 20:59 | <sayrer_> | just stop answering emails about it |
| 20:59 | <sayrer_> | then we're not stuck |
| 21:00 | <gsnedders> | I haven't answered any emails about it almost a year ago :P |
| 21:00 | <gsnedders> | s/haven't/stopped/ |
| 21:00 | jgraham | wonders how that makes us less suck |
| 21:00 | <jgraham> | *stuck |
| 21:00 | <tantek> | gnsedders - I disagree - specifically, Shelley's recent message: http://lists.w3.org/Archives/Public/public-html/2009Jul/0152.html has a good tone and a reasonable listing of concerns with the methodology of analysis of summary. At a minimum, those concerns could (should) be documented on a wiki page, and then addressed. |
| 21:01 | <sayrer_> | jgraham: well, we are stuck if you are wed to making the attributes lack of success result in a conformance error |
| 21:02 | <tantek> | gsnedders - in addition, framing the debate as "from either side" unnecessarily harms the ability for all sides to make scientific progress on various possibilities by polarizing the issue. |
| 21:03 | <tantek> | gsnedders, you're right about being stuck in circles due to it being discussed in email. if you capture the issues and follow-ups in a wiki page, it becomes much easier to represent loops and duplicates (via hyperlinks) than in email. |
| 21:04 | <gsnedders> | tantek: I think we went through stuff like what Shelley's message is about at TPAC last year with PF, though, and within a few weeks the arguments were back to what they were before that meeting |
| 21:04 | <tantek> | gsnedders - so why weren't those points recorded on a wiki then so they could be properly assessed? (rather than being repeated in email) |
| 21:05 | <jgraham> | sayrer_: So your proposition is "accept my point of view and the arguments will go away"? |
| 21:05 | <gsnedders> | tantek: I think because nobody has bothered to put them on the wiki. |
| 21:06 | <jgraham> | tantek: We have a wiki page. Is is hardly fair and balanced and attempts to make it so would unlikely go down well |
| 21:06 | <sayrer_> | jgraham: so you do think it is really important to have it result in a conformance error? |
| 21:06 | <sayrer_> | like, life or death, technical showstopper? |
| 21:06 | <gsnedders> | It would be a lot easier if we just made no comment about what was conforming :P |
| 21:07 | <tantek> | jgraham - Wikipedia has had to deal with far worse conflicts than this, by recording multiple points of view (and arguments for/against), and has done so. So I disagree with your assessment that it is "unlikely [to] go down well" - offering Wikipedia conflict documentation as evidence. |
| 21:07 | <sayrer_> | well, that is a proposal I intend to make, but we needn't go whole hog there to resolve this specific issue |
| 21:07 | <sayrer_> | the author conformance requirements have a variety of motivations |
| 21:07 | <sayrer_> | and they are inconsistently applied |
| 21:09 | <jgraham> | tantek: http://esw.w3.org/topic/HTML/SummaryForTABLE (note also http://esw.w3.org/topic/HTML/SummaryForTABLE?action=info) |
| 21:09 | <gsnedders> | I think from a syntax point of view, conformance should be what produces an obvious DOM tree and the subset of the spec's algorithm and what works in legacy UAs |
| 21:09 | <sayrer_> | legacy UAs aren't html5-conformant |
| 21:09 | <sayrer_> | not remotely |
| 21:10 | <gsnedders> | Indeed, but there is a subset of input that is parsed the same. |
| 21:10 | <jgraham> | gsnedders: I think there are significant reasons not to bloat the markup language with all the old, failed, ideas |
| 21:10 | <sayrer_> | jgraham: what are those reasons? |
| 21:11 | <sayrer_> | I am honestly curious |
| 21:11 | <sayrer_> | I don't see these attributes being reclaimed for other purposes soon |
| 21:12 | <jgraham> | sayrer_: It increases the language surface that authors will encounter. There is value in "there whould be one and only one obvious way to do it" |
| 21:13 | <sayrer_> | that is a very subjective area to venture into, but I agree it is a legitimate concern |
| 21:13 | <tantek> | jgraham - that wiki page certainly has a lot of good documentation, but does little to address follow-ups of the points made. |
| 21:15 | <tantek> | jgraham - I agree with you about minimizing the "language surface that authors will encounter" - I've been referring to that as the principle of minimal vocabulary and have documented it as a microformats design principle: http://microformats.org/wiki/minimal-vocabulary |
| 21:16 | <tantek> | in short, lower cognitive load which greatly helps ease of use (for authors and developers) |
| 21:17 | <sayrer_> | a counter example might be presentational elements |
| 21:18 | <sayrer_> | since those let authors do simple formatting without learning CSS |
| 21:21 | <takkaria> | I've tried editing various wiki pages on the w3c |
| 21:21 | <takkaria> | I seem to recall a couple of edits being reverted staright away |
| 21:21 | <takkaria> | at that point I gave up |
| 21:21 | <gsnedders> | takkaria: Well, you obviously aren't doing it right :P |
| 21:30 | <takkaria> | I think it was around the time I bothered actively debating with Rob Burns |
| 21:30 | <takkaria> | I think I was editing a page he had some investment in |
| 21:30 | <jgraham> | tantek: I think that wiki page is very unbalanced e.g. it frames the issue as "why summary should not be provided vs why summary should be provided whilst the actual issue is about the most effective way to increase the accessibility of HTML tables |
| 21:31 | <jgraham> | tantek: and my perception is that any edits I make that change existing content would be reverted |
| 21:32 | <jgraham> | So there is no point in investing my limited time in making those changes |
| 21:33 | <jgraham> | (it is possible that I m wrong. However I think that other people have the same impression; the esw wiki is viewed as poisoned from the pov of HTML) |
| 21:43 | <smedero> | I think there was one instance of an edit war on the ESW wiki between Lachlan and Laura... if memory serves. I believe Laura suggested it was a mistake... that she accidently clobbered Lachlan's updates. |
| 21:44 | <smedero> | *shrug* I find most of the wiki pages on there to be insanely difficult to parse. |
| 21:46 | <jgraham> | smedero: Agreed. Maybe that is one of the reasons I am so loathe to edit them; it is impossible to work out where to insert things and heavy refactoring would likely remove information that people were attached to |
| 21:46 | <smedero> | yep. |
| 21:46 | <jgraham> | We should make unit tests for the wiki page :) |
| 21:47 | <gsnedders> | jgraham: Are you going to get a intern to do that? :P |
| 21:48 | <jgraham> | (a series of statements that should be invariant under refactorings. Then we could replace the wiki page with the unit tests) |
| 21:48 | <smedero> | There was a suggested template structure... and people sorta stuck to it but then they misinterpreted the spirit of the headers... or perhaps... the average person doesn't share the same definition of "use case" that a software engineer has. |
| 21:53 | <tantek> | jgraham, then let me make the modest proposal of forking from the esw wiki, and copying any content relevant to HTML to the whatwg wiki, where the policies of the whatwg can be used to help encourage better community participation on the wiki. |
| 21:53 | <jgraham> | tantek: That seems like a reasonable idea |
| 21:55 | <tantek> | if Lachy has had trouble with his edits being reverted on the esw wiki, perhaps he might be able to help with such a migration to the whatwg wiki. |
| 21:56 | <tantek> | perhaps takkaria would similarly be interested in migrating to the whatwg wiki |
| 21:57 | <Lachy> | I gave up on the esw wiki based on edit wars with Rob Burns and, if smedero is right, Laura. |
| 21:58 | <Lachy> | I think the edit war with Laura was something about documenting use cases for video accessibility |
| 21:58 | <Lachy> | which is why I did document most of that on the whatwg wiki already |
| 22:07 | <takkaria> | heh |
| 22:07 | <takkaria> | the idea of migrating to the whatwg wiki is all well and good |
| 22:07 | <takkaria> | but people who aren't involved with whatwg have in the past refused to deal with it |
| 22:38 | <tantek> | takkaria - start citing the whatwg wiki in the Issues tracker. the more you make use of it, the more others will as well. |
| 22:45 | <Hixie> | btw in case any edit wars turn up on the whatwg wiki - feel free to create whatever pages you want; if two people disagree about how a page should exist on the whatwg wiki, then fork it and both maintain an independent copy |
| 22:45 | <Hixie> | it's just a scratch space |
| 22:48 | <tantek> | Hixie - that's a good start for documenting multiple points of view. |
| 22:48 | <Lachy> | Is there a page on the whatwg wiki documenting the summary issue? |
| 22:48 | <Hixie> | not to my knowledge |
| 22:49 | <Hixie> | christ, www-font makes public-html look positively friendly |
| 22:49 | Lachy | goes to create a "Table Summary" page |
| 22:50 | <Lachy> | Hixie, do you recall what evidence the proponents for summary="" claim supports their position? |
| 22:51 | <Lachy> | The only thing I can think of is that the attribute is read out by default in some screen readers |
| 22:51 | <Hixie> | the PFWG has consensus on the idea that summary="" is successful |
| 22:51 | <Lachy> | yeah, but what's that consensus based on? |
| 22:51 | <Lachy> | maybe there's something in the esw wiki about it. |
| 22:52 | <hober> | That's like having consensus that the Roman Empire never fell. |
| 22:52 | <gsnedders> | It didn't. |
| 22:52 | <Hixie> | the consensus is based on the expert opinions of their constituents, i believe |
| 22:53 | <Lachy> | so it's an appeal to authority, who's consensus is itself based on an appeal to authority? |
| 22:53 | <Lachy> | wow |
| 22:53 | <Hixie> | digg is weird |
| 22:54 | <gsnedders> | Hixie: Looking at the HTML 5 codec thing? |
| 22:54 | <Hixie> | article "Decoding the HTML 5 video codec debate - Ars Technica" has the following link suggested as "related by keyword": "5 Bizarre Sexual Conditions That Can Ruin Your Life" |
| 22:55 | <ojan> | 5. Discussing video codecs in bed |
| 22:56 | <Hixie> | i can't scroll the digg comments page in safari |
| 22:56 | <Hixie> | it keeps scrolling back to the top |
| 22:56 | <Hixie> | wtf |
| 22:56 | <Hixie> | i hate digg |
| 22:56 | <gsnedders> | Hixie: Any interesting replies to my comments? |
| 22:57 | <Lachy> | gsnedders, I don't see any |
| 22:57 | <gsnedders> | Lachy: Do you have all threads expanded? |
| 22:59 | <Hixie> | i was just looking to see who was trashtalking me, but it turned out to me someone who's comments are all paranoid rants, so i didn't feel the need to investigate further :-) |
| 22:59 | <Lachy> | gsnedders, I do now. There's some responses to you |
| 23:00 | <Lachy> | "Ian Hickson is personally responsible for the failure of the video tag, because he has made too many enemies in all three camps..." :-D |
| 23:00 | <Hixie> | check out the guy's other comments though |
| 23:01 | <Lachy> | this one? "Well apparently Ian Hickson has somehow acheived the post where he is personally responsible for HTML 5, and he routeinly trash talks all the other members of the committee. I don't understand how that happened, but this is mostly a result of push back on this Ian Hickson character..." |
| 23:01 | <Hixie> | i mean other comments in other threads |
| 23:01 | <Lachy> | oh, here? http://digg.com/users/B1665r |
| 23:02 | <Hixie> | yeah |
| 23:02 | <Hixie> | it strikes me that saying that he only writes paranoid rants is not going to help me any in convincing him that i don't trashtalk people regularly |
| 23:03 | <Hixie> | though he wouldn't have known i was talking abotu him if you hadn't pasted what he wrote and his username into the channel :-P |
| 23:04 | <Lachy> | just get krijnh to clean up the logs and he'll never know |
| 23:04 | <Hixie> | i don't hide my words :-) |
| 23:05 | <gsnedders> | I just talk incoherently, and claim not to use words. |
| 23:05 | <gsnedders> | Though I guess UAX #29 would disagree |
| 23:07 | <Lachy> | gsnedders, no-one ever knows what you're talking about anyway :-) |
| 23:07 | <gsnedders> | Lachy: esp. on Twitter |
| 23:09 | <Lachy> | LOL, some of these rationale for including summary="" from the esw wiki are funny. |
| 23:10 | <Lachy> | "19. Including @summary solves a real problem." |
| 23:10 | <Lachy> | "20. Including @summary removes an obstacle to accessibility advocates promoting the use of HTML5." |
| 23:15 | <tantek> | Lachy - perhaps the [citation-required] template/convention of Wikipedia would be useful for claims that are made without supporting evidence. |
| 23:16 | <Lachy> | tantek, unfortunately, the esw wiki seems to lack any useful templates like that |
| 23:16 | <tantek> | Lachy, but you could introduce such templates into the whatwg wiki |
| 23:16 | <Lachy> | yeah, that would be possible |
| 23:16 | <tantek> | and thus continue your fork of the Table Summary page there with more diligence |
| 23:17 | <hober> | pedantic correction: it's {{Citation needed}} |
| 23:17 | <tantek> | regarding "PFWG has consensus on the idea that summary="" is successful" - it is better (more scientific) to document data/opinions, even bad data / opinions, and debunk it, than not to document it at all. |
| 23:17 | <Lachy> | hober, it's it {{cite}} ? |
| 23:18 | <Lachy> | s/it's/Isn't/ |
| 23:18 | <tantek> | debunk *them* (them being the opinions as it were, not the people) |
| 23:18 | <tantek> | if necesary, label opinions as "appeal to authority", or "based on personal anecdote" etc., whatever you think is the appropriate critique. |
| 23:19 | <tantek> | because even data based on personal anecdote is better than no data. |
| 23:19 | <tantek> | it just has be considered with the appropriate weight relative to other kinds of data (data on studies of a hundred websites for example) |
| 23:23 | <Lachy> | I'm not going to copy and paste most of the existing content. I think a better approach is to sift through the existing page and e-mail archives and simply document any supported claims, preferably with citations for each |
| 23:24 | <Lachy> | But not tonight. I'm getting quite tired. |
| 23:25 | <tantek> | Lachy yes that seems reasonable |
| 23:26 | <tantek> | There may also be value in documenting which claims are *unsupported* in a separate section, just to highlight the fact that the claims are unsupported - so that if such claims are reiterated in email - you can follow-up with the response that the claim is unsupported per URL. |
| 23:26 | <Lachy> | I started the page. http://wiki.whatwg.org/wiki/Table_Summary |
| 23:27 | <Lachy> | feel free to continue with it and I'll try and get back to it tomorrow evening |
| 23:27 | <Hixie> | what's it for? |