| 01:51 | <heycam> | Hixie, how many objects need to do this? |
| 01:51 | <heycam> | (wondering whether it's worth having a special xattr for it) |
| 01:51 | <heycam> | i think the sentence should be fine |
| 01:52 | <Hixie> | right now i only know of one, and i wouldn't say it needs it per se -- just that i would have used such a feature if it was available |
| 01:52 | <heycam> | "In the ECMAScript binding, host objects that implement interface Blah have a [[Call]] method that does ...." |
| 01:52 | <heycam> | k |
| 01:56 | <Hixie> | what would "...." be? |
| 01:56 | <Hixie> | "invokes the [[Get]] operator with the same argument"? |
| 01:56 | <Hixie> | (just putting this in a comment for now in the spec in case i find we do need it) |
| 01:56 | <heycam> | well not quite, you'll have to deal with different numbers of arguments i suppose |
| 01:56 | <heycam> | what happens if you pass 0 or more than 1 arg? |
| 01:57 | <Hixie> | dunno, haven't tested |
| 02:03 | <heycam> | so i'd do "... [[Call]] method that behaves as follows: 1. If called with 0 or more than 1 argument, throw a TypeError. 2. Let arg be the single argument passed. 3. Invoke [[Get]] on the object with ToString(argument) as the property name. 4. Return Result(3)." |
| 02:03 | <heycam> | or so |
| 02:03 | <Hixie> | k, thanks |
| 02:05 | <heycam> | c |
| 02:05 | <heycam> | ww sorry |
| 07:09 | <Hixie> | annevk42: yt? wondering what the timeline you're expecting for cssom is, and whether you think that'd be a good spec to have define when the 'resize' and 'scroll' events fire |
| 07:10 | <hsivonen> | argh. lots of email about validation over the weekend |
| 07:24 | <sinclair> | hello all |
| 07:25 | <sinclair> | so, anyone in here using O3D? |
| 07:28 | <sinclair> | im trying to find a online chat community who is working on O3D |
| 07:28 | <sinclair> | hello |
| 08:47 | <annevk42> | Hixie, I basically haven't found time to work on CSSOM for the past year or so |
| 08:48 | <annevk42> | Hixie, though resize and scroll could I suppose be part of the subset that is CSSOM Vie |
| 08:48 | <annevk42> | w |
| 08:53 | <jgraham> | What happens if I have some HTML like <body bgcolor="black"><article><font color="white">... and the <article> contents ends up in a feed? ill it be unreadable? Taking some but not all style information seems like it could be dangerous |
| 08:53 | <jgraham> | s/ill/will/ |
| 08:55 | <Dashiva> | jgraham: Well, that's "wrong" to begin with |
| 08:55 | <jgraham> | Dashiva: In what way? |
| 08:55 | <jgraham> | Under the assumption that anything that works in a UA is conforming |
| 08:55 | <Dashiva> | You should set text color on the body element as well |
| 08:56 | <Dashiva> | Always specify foreground and background together |
| 08:56 | <jgraham> | Dashiva: According to whom? Not according to the supposed no-author-requirements HTML spec |
| 08:57 | <Dashiva> | According to the best practices that are supposed to replace them |
| 08:57 | <Dashiva> | No requirements includes the freedom to shoot yourself in the foot |
| 08:58 | <jgraham> | Yeah but I'm trying to work out why that's a good thing |
| 08:58 | <Dashiva> | Has anyone claimed that? |
| 08:59 | <jgraham> | That having no author requirements is a good thing? Sure |
| 08:59 | <Dashiva> | That's a choice between evils, not goods |
| 08:59 | <jgraham> | Can you explain more? |
| 08:59 | <Dashiva> | Requirements come with ratholes, bikesheds, and aren't enforced |
| 09:00 | <Dashiva> | The argument (as I read it) goes that avoiding all those weighs up for the loss of the guideance authority that could be provided |
| 09:00 | <jgraham> | Lack of requirments will come with a huge rathole, at the bottom of which will be the original ratholes but renamed "best practices" |
| 09:01 | <jgraham> | So I don't see any possibility for gain here |
| 09:01 | <Dashiva> | That is the gain |
| 09:01 | <Dashiva> | That huge rathole will always be there, that's what good authoring is all about |
| 09:01 | <Dashiva> | But without requirements, we outsource the war over where the hole should be |
| 09:02 | <jgraham> | As far as I can tell the main agenda here is that some people don't like the HTML5 draft because it doesn't endorse something that they hold dear to them. So they have decided that the only way out is to try to remove all restrictions and be told "do what you like" |
| 09:02 | <Dashiva> | Well, the basic complaint holds: What use are requirements that no one actually requires? |
| 09:03 | <jgraham> | Sometimes this is wrapped up in theoretical RFC 2119 arguments |
| 09:04 | <jgraham> | Dashiva: The point of authoring requirements in a language with error correcting clients is not to describe the only possible way of doing things |
| 09:04 | <jgraham> | It is to reduce the authoring surface of the language so that it does not grow without bounds and so there are as few sharp edges as possible |
| 09:05 | <Dashiva> | But nobody reads those requirements |
| 09:05 | <jgraham> | Dashiva: Demonstrably untrue |
| 09:05 | <Dashiva> | They read second- or third-hand references that also includes local author intent, best practices, misconceptions, urban legends, etc etc |
| 09:06 | <jgraham> | There has been a huge movement toward following standards. Yes there is a lot of cargo cult nonsense. But there is enough essentially good stuff that we should no thro the whole thing away just to make the lives of spec authors better |
| 09:06 | <jgraham> | We even have a design principle about that |
| 09:06 | <Dashiva> | Nobody's saying we should throw it away |
| 09:06 | <jgraham> | Dashiva: They really are |
| 09:07 | <Dashiva> | They may think so |
| 09:07 | <Dashiva> | But those references people use will still exist |
| 09:07 | <Dashiva> | It's only mark's assholes that will be weakened |
| 09:09 | <jgraham> | Dashiva: I would expect one effect of removing conformance requirements to be that people will start to reject HTML5 for cargo-cult reasons |
| 09:09 | <Dashiva> | Oh, sure. But there wouldn't be anything for them to reject. It would be a user agent spec full and whole. |
| 09:09 | <jgraham> | Dashiva: Not really. |
| 09:10 | <jgraham> | It still will define a language |
| 09:10 | <jgraham> | It just won't tell you which bits are a really bad idea |
| 09:10 | <Dashiva> | Now we're back at html4 ground |
| 09:10 | <jgraham> | ? |
| 09:10 | <Dashiva> | The language in the spec does not matter for authors, authors and users only go via user agents |
| 09:11 | <jgraham> | But we just discussed this. Authors learn from somewhere and that somewhere often has looked at the spec |
| 09:11 | <jgraham> | And authors do use vaidators |
| 09:12 | <jgraham> | and validators have read the spec for sure |
| 09:12 | <Dashiva> | But again the authors don't go directly to the spec |
| 09:12 | <Dashiva> | They see the validator's rules, which would include all kinds of best practices |
| 09:13 | <jgraham> | Dashiva: What kind of best practices are encoded by validator.w3.org or validator.nu, beyond what is in the spec? |
| 09:13 | <Dashiva> | All the author requirements |
| 09:14 | <jgraham> | I mean today |
| 09:14 | <Dashiva> | Today there isn't much need, henri gets it put into the spec instead |
| 09:14 | <Dashiva> | In particular the downplayed errors, he seems to be the main driving force there |
| 09:15 | <jgraham> | Dashiva: The concept of downplayed errors is silly |
| 09:15 | <jgraham> | That is a UI issue |
| 09:15 | <Dashiva> | If so, that strenghtens the point |
| 09:16 | <Dashiva> | These are things he doesn't want the validator to mark as errors. Now he's working inside the spec, if the spec dropped author reqs he would just go ahead and do it directly |
| 09:16 | <Philip`> | What if someone else writing a different validator has different ideas of what should be considered an error or a warning? |
| 09:17 | <jgraham> | Dashiva: So in the future what do you envison? validator.w3.org uses best practices document A. commercial.validator.org uses weaker best practices document B to iucrease marketshare. Validator validate.net is written by areholes and has a whole load of additional requirements. An author tries to validate her document. What does she conclude? |
| 09:17 | <Dashiva> | Then you get competition in who validates the best |
| 09:17 | <Philip`> | Putting those requirements in the spec prevents the benefits of competition, because there will only be one official blessed notion of validity |
| 09:17 | <Philip`> | which seems similar to the argument against putting machine-readable schemas in the spec |
| 09:17 | <jgraham> | Dashiva: Re: dowwnplayed errors, Henri can already choose to downplay any error he likes. It is a pure UI issue |
| 09:18 | <Dashiva> | jgraham: Again, sure. It's not material to my point. The point is that he puts effort into convincing Hixie to mark them as downplayed. That means he cares. |
| 09:19 | <jgraham> | Philip`: The problem with putting schemas in the spec is that it is not possible to express all conformance requirements in schema and there is the possibility of divergence between the actual requirements and the schema (due either to the limitations or to bugs) |
| 09:20 | <Dashiva> | This situation exists even now, of course |
| 09:20 | <Dashiva> | There are requirements outside the HTML5 spec |
| 09:20 | <jgraham> | Dashiva: And I still maintain it is silly to have the requirements in the spec. So I don't see that I'm taking an inconsistent position |
| 09:20 | <Dashiva> | This whole thing would just be moving one set of requirements into some other (possible more than one) location |
| 09:21 | <jgraham> | s/requirements/downplayed errors/ |
| 09:21 | <Dashiva> | It's not your position. His effort suggests that even if the spec did not have author requirements, he personally cares enough about these issues that his validator would have best practices for them |
| 09:22 | <Dashiva> | So no, validators would not just be windows onto the UA spec |
| 09:22 | <jgraham> | Dashiva: I don't understand your argument at all |
| 09:22 | <jgraham> | specifically the sentence "validators would not just be windows onto the UA spec |
| 09:22 | <jgraham> | " |
| 09:23 | <Dashiva> | [10:17:05] <jgraham> Dashiva: What kind of best practices are encoded by validator.w3.org or validator.nu, beyond what is in the spec? |
| 09:23 | <jgraham> | "The spec" in this case is not just a "UA spec" though |
| 09:24 | <Dashiva> | Now we're just repeating ourselves |
| 09:24 | <Dashiva> | The validator doesn't have its own requirements because they're in the spec. If they're not in the spec, they would be in the validator instead (so the validator itself wouldn't change) |
| 09:26 | <jgraham> | There is no disagreement about what the requirements are. There is only the issue of whether some errors should be displayed with greater weight. That, to me, is a UI issue and I think that it shouldn't be in the spec |
| 09:27 | <Dashiva> | Leave that issue. It's just an example to show that henri personally cares about some of the requirements, to support the claim that the validator would still have requirements if they weren't in the spec |
| 09:27 | <jgraham> | So I disagree with Henri here. But I don't feel bad about it too much because I disagree with Henri less than almost anyone else involved with HTML5 :) |
| 09:28 | <jgraham> | Dashiva: Your argument seems totally unjustified. Unless you can point to things that are not in the spec but Henri has included as requirements anyway I don't think that you can conclude that he would do such a thing |
| 09:29 | <jgraham> | I also don't think it's very relevant to whether, overall, there is any benefit to be gained by removing all the author requirements from HTML5 |
| 09:29 | <jgraham> | Anyway, I guess I should do something more useful... |
| 09:30 | <Dashiva> | We were there earlier |
| 09:31 | <Dashiva> | It's related because it claims that validators wouldn't change much. And validators and non-primary sources are where authors go |
| 09:31 | <Dashiva> | => Things would not change for authors and users |
| 09:32 | <jgraham> | I still disagree but there se go |
| 09:32 | <jgraham> | *we |
| 09:32 | <Dashiva> | Sure, you might get small civil wars over what best practices are best, but that would be hidden to most people. And it already happens, just look at WCAG Samurai |
| 09:35 | <jgraham> | Actually the accessibility space is a great example of wwhy this kind of thing can be a disaster. There are so many documents relating to WCAG that it is almost impossible to tell where you should look for any given thing. And so you ask an "expert" and get an answer. And ask another "expert" and get a different answer. And so on... |
| 09:36 | <Philip`> | Does anyone take any notice of the WCAG Samurai's suggestions, or is it evidence that everyone just follows the most official sounding guidelines (from the W3C) and ignores everything else regardless of merits? |
| 09:36 | <Dashiva> | I think the evidence is that people ignore everything, period |
| 11:23 | <hsivonen> | Dashiva: actually, I don't get to put stuff in the spec. If I did, td@align and object@classid would be valid by now. |
| 11:24 | <hsivonen> | v.nu has non-spec-backed warnings especially around the area of authors shooting themselves in the foot by not using UTF-8 |
| 11:59 | <annevk42> | hsivonen, sync XHR is superevil |
| 11:59 | <annevk42> | also to spec |
| 12:00 | <annevk42> | though I think I have it nailed down now |
| 12:06 | <hsivonen> | I wonder if the script execution code in any browser is clear to someone who just reads it |
| 12:06 | <hsivonen> | anyway, it seems I managed to make http://mxr.mozilla.org/mozilla-central/source/content/base/test/test_bug461555.html not crash |
| 12:06 | <hsivonen> | yay |
| 12:09 | hsivonen | observes that it would be appropriate to call Lamborghinis dump trucks if the large majority of Lamborghini drivers used them as dump trucks |
| 12:37 | gsnedders | guesses that is some reference to a post on the list |
| 12:53 | <jgraham> | gsnedders: No actually it is true that most owners of italian sportscars use them mainly for taking stuff to the dump. Researchers say "the have such small boots that it can require hundreds of trips to dispose of the cuttings every time you trim your garden hedge" |
| 12:53 | <gsnedders> | :P |
| 12:57 | <beowulf> | i looked at a disabled car parking space the other day, it'd be almost impossible to implement something like that incorrectly |
| 12:58 | <beowulf> | and easy to do as well, just paint some lines |
| 13:14 | <annevk42> | I will toally include <supercalifragilisticexpialidocious> in all my documents from now on |
| 13:14 | <jgraham> | hsivonen: I am interested to know what you think of the argument that making <font> non conforming has positive implications for the learnability of the language (due to making it smaller) |
| 13:14 | <jgraham> | And by helping authors avoid doing things that are belived to be bad |
| 13:15 | <annevk42> | (why: http://www.w3.org/mid/5D1706DF-0909-418F-8592-24535A5C19A1⊙if ) |
| 13:15 | <jgraham> | Specifically the difference beteen <font> and <span style> is that authors using <span> must be aware of CSS and have taken a concious decision to use inline style in place of seperated style |
| 13:15 | <jgraham> | That seems like a positive ting |
| 13:16 | <jgraham> | If a tool emits <span style> in place of <font> it is neutral |
| 13:16 | <jgraham> | So therefore removing <font> seems to have an overall positive balance |
| 13:17 | <hsivonen> | somehow I'm more annoyed when OpenOffice.org or Nvu emit style='' |
| 13:17 | <jgraham> | hsivonen: Is that just an aethetic preference though? |
| 13:17 | <jgraham> | *aesthetic |
| 13:18 | <hsivonen> | jgraham: I tend to think that HTML already has enough weird corners that learnability needs to be achieved by prioritizing what to reveal and what order to a learner |
| 13:18 | <hsivonen> | jgraham: no, if you use TextWrangler's grep find&replace to zap style='', you then need to separately zap spans, too |
| 13:18 | <jgraham> | hsivonen: What weird corners are part of the conforming language? Optional tags I guess |
| 13:18 | <Philip`> | It seems a pain if you use <font> for most styling until you reach a point where you want e.g. small-caps and you have to switch to <span style> and now have to understand how to use both and how they interact, rather than simply using the more powerful mechanism from the start |
| 13:19 | <hsivonen> | jgraham: optional tags is the #1 thing, yes |
| 13:19 | <jgraham> | Philip`: Indeed |
| 13:19 | <hsivonen> | #1 weird corner, not #1 thing to learn |
| 13:20 | <hsivonen> | Philip`: I agree. <font color> makes some sense for GUI tool-generated visual empahasis as long as the GUI tool is simple enough not to support all of CSS |
| 13:20 | <hsivonen> | also, with <font face> and <font size>, you lose |
| 13:20 | <jgraham> | I note that we tried the wwhole <font>-for-GUI-tools only thing and people didn't go for it |
| 13:21 | <hsivonen> | jgraham: we tried a slightly different thing |
| 13:21 | <hsivonen> | jgraham: we tried <font style> for GUI tools |
| 13:21 | <hsivonen> | jgraham: also, I see sayre's point about survival in syndication |
| 13:22 | <jgraham> | hsivonen: I may misremembert but I'm not sure if the distincion makes a big difference |
| 13:22 | <jgraham> | hsivonen: It seems relatively dangerous to keep some but not all style during syndication since you could e.g. render some text invisible |
| 13:23 | <hsivonen> | jgraham: yeah, it only works for <font color> red |
| 13:23 | <hsivonen> | if a user hits select all, picks a color and the editor generates a zillion <font>s, you lose |
| 13:24 | <hsivonen> | if the user selects two words and picks red, not generating <font color> is quite silly |
| 13:24 | <Philip`> | jgraham: Why is rendering some text invisible dangerous? |
| 13:25 | <Philip`> | jgraham: You could achieve the same effect by replacing all the text in your feed with whitespace |
| 13:25 | <jgraham> | hsivonen: Like I said I think the latter case is a wash compard to <span style>. At least, as Philip` said, if you select two words and select small caps you end up with the same markup |
| 13:25 | <Philip`> | jgraham: ...unless you mean "dangerous" as in "likely to go wrong", rather than "permitting malicious behaviour" |
| 13:25 | <jgraham> | Philip`: Not dangerous in the security sense, but in the UI sense |
| 13:25 | <hsivonen> | jgraham: I think small caps is on a more advanced level of typography than red |
| 13:26 | <hsivonen> | jgraham: different users. different classes of editor UIs |
| 13:26 | <jgraham> | hsivonen: I'm not sure there is any correlation between people who would use small caps and people who would use an advanced editor |
| 13:27 | <hsivonen> | jgraham: note how in my email, I didn't take the time to argue Sam's <font color> bait per se |
| 13:27 | <jgraham> | And if the use case is people who don't read the markup, I don't really see how <font> is an advantage |
| 13:28 | <Philip`> | hsivonen: Maybe a better example would be highlighting (as in <span style=background:yellow>...</span>), which in (at least) OO.o has almost exactly the same UI exposure as font colour |
| 13:28 | <hsivonen> | <font color> survives syndication and legacy tools better |
| 13:28 | <hsivonen> | Philip`: indeed. much better example. |
| 13:30 | jgraham | is being slow and doesn't understand why that is a bette example |
| 13:31 | <Philip`> | jgraham: Because it's conceptually very similar to font colour (as evidenced by word processor UIs), and can be implemented with very similar markup if you use <span style>, but not if you use <font> |
| 13:31 | <Philip`> | so using <font> introduces an unnecessary distinction between those similar concepts |
| 13:32 | hsivonen | notes that <font color> is one special thing compared to style="" like @summary is one special legacy thing compared to aria-describedby |
| 13:32 | <Philip`> | which makes everything harder to understand, and is more fragile when e.g. syndicators strip style attributes but not font elements |
| 13:32 | <jgraham> | Oh. OK that's what I though. But I wasn't sure if it was supposed to support hsivonen's point |
| 13:32 | <hsivonen> | or maybe aria-label considering that it's human-readable text hidden in an attribute |
| 13:32 | <Philip`> | I haven't been concentrating enough to read what hsivonen's point was so I don't know |
| 13:33 | <hsivonen> | my point is that if you want to make a principled markup about stuff that was in HTML 4, you'd better make sure your principle applies to <font color>, too |
| 13:33 | <hsivonen> | otherwise, you are just picking favorites while pretending to have a principle |
| 13:34 | <hsivonen> | s/principled markup/pricipled point/ |
| 13:34 | <hsivonen> | or something |
| 14:00 | <annevk42> | hsivonen, you could group all the font messages into one |
| 14:01 | <annevk42> | hsivonen, your page uses a legacy element called "font" please don't |
| 14:27 | <annevk42> | licensing⊙fo: will we now get an influx of laywers on whatwg⊙wo? |
| 14:27 | <annevk42> | oh, I guess blocking from third parties is still blocked |
| 14:37 | <gsnedders> | annevk2: Licensing⊙fo is mainly volunteers, not all lawyers |
| 14:40 | <annevk42> | s/blocking/email/ |
| 15:34 | gsnedders | squints |
| 15:34 | <gsnedders> | There's a mysterious gap of trains between 15:42 and 17:42 from Nyköping to Linköping, right in the rush hour :\ |
| 15:35 | <jgraham> | gsnedders: You really really want to take the bus |
| 15:35 | <gsnedders> | Ah, OK |
| 15:35 | <gsnedders> | How long does that take? |
| 15:36 | <jgraham> | 1.5 hours |
| 15:36 | <jgraham> | http://www.flygbussarna.se/ |
| 15:36 | <gsnedders> | Yes, I've found taht already :P |
| 15:36 | <gsnedders> | *that |
| 15:36 | <jgraham> | http://www.flygbussarna.se/Default.aspx?lang=EN is even better |
| 15:37 | <gsnedders> | Yes, I found that actually :P |
| 15:37 | <gsnedders> | (I didn't find the Swedish.) |
| 15:37 | <jgraham> | Oh well wwhy are you asking how long it takes then? |
| 15:38 | <gsnedders> | Because I hadn't clicked through the three pages to the timetable and hadn't downloaded it yet, and I thought you'd know |
| 15:38 | gsnedders | guesses jgraham has probably flown from STN to NYO like he will be doing next week |
| 15:39 | <jgraham> | gsnedders: Just a little |
| 15:40 | gsnedders | was just emailing Bibbi because she has the key to his flat, and he won't get to Linköping until a bit before 19:00 |
| 15:40 | <jgraham> | gsnedders: What day of the wweek? |
| 15:40 | <gsnedders> | jgraham: Friday. |
| 15:40 | gsnedders | is really inconvenient |
| 15:41 | jgraham | wonders if he can fit a new keyboard in his hand luggage |
| 15:41 | <jgraham> | gsnedders: Oh everyone will still be here for Friday beer |
| 15:41 | <gsnedders> | Ah |
| 15:41 | <jgraham> | If you mean Fri. 26th |
| 15:41 | <gsnedders> | Isn't a keyboard a bit big |
| 15:41 | <gsnedders> | yeah |
| 15:41 | <jgraham> | gsnedders: Probably :( |
| 15:42 | <gsnedders> | hmm, fits in my laptop bag (which holds my 17" MBP) |
| 15:42 | <jgraham> | gsnedders: It is highly likely that bibbi will still be at the office then. You should of course check |
| 15:42 | gsnedders | just emailed her |
| 15:43 | <jgraham> | gsnedders: Do you know your address? |
| 15:43 | jgraham | is curious |
| 15:43 | <gsnedders> | Off the top of my head? No |
| 15:56 | <gsnedders> | jgraham: But yeah, then I have the weekend to come to terms with where I am in the world :P |
| 16:11 | <jgraham> | gsnedders: I think I am busy that weekend but we should organise something social for the first weekend in July |
| 17:16 | <takkaria> | Opera does seem to gobble up whatwg people |
| 17:21 | <Philip`> | There must be sinister motives |
| 17:35 | gsnedders | remembers talking with MikeSmith over lunch at TPAC about the percentage of Opera employees involved in W3C WGs |
| 17:54 | <dglazkov> | Philip`: do you have a bug in Chromium tracker for the canvas failures? |
| 18:00 | <Philip`> | dglazkov: No - I've never properly looked at Chromium's behaviour on tests at all |
| 18:06 | <jgraham> | takkaria: When are you coming over? |
| 18:20 | <takkaria> | jgraham: Sunday morning |
| 18:21 | <annevk42> | are you all still there in August? |
| 18:21 | <takkaria> | I'm there til mid-September |
| 18:22 | <annevk42> | at one point I made this remark I'd visit Linkoping on another time then during the Christmas party, but I haven't made the time yet |
| 18:22 | <takkaria> | btw, do you pronounce Linköping as "linchurping"? I can't figure it out from my phraseboo |
| 18:23 | <Philip`> | How exciting, an IE compat list update |
| 18:24 | <Philip`> | Sites added: 7m.cn beijing.com.cn benq.com.cn btbroadbandinformation.com btopenworld.com chinanews.com.cn chinatelecom.com.cn donews.com foundertech.com gmail.com kooxoo.com mydown.com narrowad.com tobacco.gov.cn |
| 18:24 | <annevk42> | "linkshohping" or so |
| 18:24 | <annevk42> | http://en.wikipedia.org/wiki/Linköping might have better tips |
| 18:24 | <Philip`> | Sites removed: about.com desjardins.com guardian.co.uk hyves.net hyves.nl intuit.com me.com mozilla.com mozilla.org |
| 18:24 | <krijnh> | hyves.nl :S |
| 18:24 | <jgraham> | takkaria: Ah OK. I just thought I should check in cae you were on the same flight as me on Saturday |
| 18:25 | <jgraham> | Although I guess you are not flying from Stanstead |
| 18:25 | <dglazkov> | Philip`: ok |
| 18:25 | <dglazkov> | Philip`: I'll look over. Thanks for the test suite :) |
| 18:25 | <Philip`> | The IE compat list has very little activity, given that it's got three thousand entries and only a few dozen changes per month |
| 18:26 | <takkaria> | jgraham: no, I'm coming from manchester |
| 18:26 | <takkaria> | jgraham: are you doing Stanstead->Schiphol->Linkoping? |
| 18:26 | <jgraham> | takkaria: It has a soft k pronounciation so you ronounce it like lin-sher-ping |
| 18:26 | <takkaria> | right, gotcha |
| 18:26 | <jgraham> | or so |
| 18:26 | <Philip`> | Soft k? Crazy language :-( |
| 18:26 | <jgraham> | (imagine you are slurring the word shopping...) |
| 18:27 | <annevk42> | just write it on a piece of paper or laptop and show that to people |
| 18:27 | <jgraham> | takkaria: No STN-NYO w/ryanair |
| 18:27 | <annevk42> | they'll know what you mean :) |
| 18:27 | <jgraham> | Philip`: You're telling me. I still can't pronounce anything |
| 18:28 | <jgraham> | the sj and tj sound is... interesting |
| 18:28 | <gsnedders> | annevk42: I'm there until the 31st |
| 18:29 | <gsnedders> | jgraham: This Sat? |
| 18:29 | <jgraham> | gsnedders: yes |
| 18:29 | <gsnedders> | jgraham: You getting the 6:30am flight? :P |
| 18:29 | <jgraham> | gsnedders: No the on in the evening |
| 18:30 | <gsnedders> | jgraham: Hehe, I was expecting not :) |
| 18:30 | gsnedders | wonders why he can't print |
| 18:30 | <takkaria> | I'm getting the 0600 out of manchester on sunday morning |
| 18:30 | <takkaria> | hellish really |
| 18:30 | <gsnedders> | Ow. |
| 18:31 | gsnedders | is flying at 13:30 from STN -> NYO |
| 18:31 | <takkaria> | the alternative is getting to Linkoping airport at 11 |
| 18:31 | <gsnedders> | (Next Friday) |
| 18:31 | <takkaria> | (11pm, that is) |
| 18:31 | <gsnedders> | Oh well, I get to cross another person off the list of cabal members to meet :P |
| 18:36 | <annevk42> | takkaria, just party until 3:30 and then take an hour and a half to get to the airport :) |
| 18:56 | <annevk42> | I wonder if http://www.cssquirrel.com/comic/?comic=22 influences my chances of getting invited to future BBQ parties |
| 19:01 | jgraham | wonders if it is churlish to point out that alt=comic doesn't really help blind users much |
| 19:01 | <gsnedders> | hah |
| 19:02 | <annevk42> | surely it has longdesc? |
| 19:02 | <jgraham> | Curiously not |
| 19:02 | <annevk42> | I'm so glad I live in the Netherlands |
| 19:03 | <jgraham> | annevk42: Why? |
| 19:03 | <annevk42> | I can already feel the wrath of John Foliot for these highly inappropriate jokes |
| 19:03 | <Hixie> | writing alt text for comics is actually surprisingly easy |
| 19:03 | <annevk42> | alt=":-)" |
| 19:05 | <annevk42> | Hixie, Google didn't do it though, for their huge Chrome comic book |
| 19:05 | <Hixie> | (photographs are a pain) |
| 19:05 | <Hixie> | annevk42: google put the chrome comic book online? |
| 19:05 | <Hixie> | sweet |
| 19:05 | <Hixie> | uri? |
| 19:05 | <annevk42> | like ages ago |
| 19:05 | Hixie | has a book version of that |
| 19:06 | <annevk42> | seems to be here among other places: http://www.google.com/googlebooks/chrome/ |
| 19:06 | <Philip`> | Hixie: http://www.google.com/search?q=chrome+comic+book |
| 19:06 | <annevk42> | fancy |
| 19:06 | <Hixie> | hm yeah, no alt text |
| 19:06 | <Hixie> | that is sad |
| 19:07 | <Hixie> | on another note, is it possible to remove duplicates from a list of strings in O() less than sorting a string? |
| 19:07 | <Hixie> | er |
| 19:07 | <Hixie> | sorting the list of strings |
| 19:08 | <Philip`> | Yes, if you don't care about memory usage |
| 19:09 | <Philip`> | (e.g. have a hash-set containing all the strings you've seen so far, and for each new string you do an O(1) test to see if it's new) |
| 19:09 | <Philip`> | (so it ends up as O(n) if you've got n strings) |
| 19:09 | <Philip`> | (vs O(n log n) for sorting) |
| 19:10 | <Philip`> | (which doesn't seem a hugely important difference in practice, since n will usually be 1 or sometimes 2) |
| 19:26 | Philip` | decides to adopt a policy of not reading analogies |
| 19:29 | annevk42 | wonders if he should point out that the new iPhone compass allows you to switch between true North and magnetic north |
| 19:36 | jgraham | wonders wtf the anaology was supposed to mean |
| 19:36 | <jgraham> | and also wtf he can't spell |
| 19:37 | <Hixie> | Philip`: hmm, thanks |
| 19:38 | <Hixie> | wait was my compass analogy bad? or is this someone else's compass analogy |
| 19:38 | <annevk42> | you got a reply |
| 19:38 | <Hixie> | ah |
| 19:40 | <jgraham> | Hixie: FWIW I think "modifying user behaviour" is something that conformance requirements are supposed to do in the sense that the state "author behaviour given conformance requirements" is expected to be different from "author behaviour given no conformance requirements" |
| 19:41 | <annevk42> | people often get the compass interface wrong; don't fix the interface because experts tell you so; instead educate the masses |
| 19:41 | <annevk42> | so as anology that'd mean |
| 19:42 | <annevk42> | people often get summary wrong; don't fix summary because experts tell you so; instead educate the masses |
| 19:42 | <Hixie> | jgraham: i agree that, to the same extent as a compass is intended to modify user behaviour, a validator (and conformance requirements) are intended to modify user behaviour. |
| 19:43 | <jgraham> | Right |
| 19:43 | jgraham | is way too tired for the compass analogy |
| 19:45 | <annevk42> | it seems to me that true noth + gps obsoletes the need for a compass (maybe even gps by itself if you can walk around a bit) and any kind of trouble the masses might have in understanding it, but maybe that takes the anology too far |
| 19:46 | <Philip`> | I would expect the masses would have somewhat more trouble understanding a GPS device than a compass... |
| 19:47 | <annevk42> | I'm thinking of a map + arrow |
| 19:49 | <annevk42> | I'm not sure what's wrong with getting a better compass though. According to Wikipedia ships widely use a Gyrocompass, which points to true north |
| 19:50 | <jgraham> | annevk42: http://en.wikipedia.org/wiki/Gyrocompass should explain the problem |
| 19:50 | <Philip`> | http://en.wikipedia.org/wiki/File:Kreiselkompass_Schnitt_Anschütz.jpg - looks like a great device to take on an orienteering trip |
| 19:50 | <annevk42> | there's a lot of issues with the magnetic north: http://en.wikipedia.org/wiki/North_Magnetic_Pole#Magnetic_North_and_Magnetic_Declination |
| 19:51 | <annevk42> | jgraham, as I said the new iPhone does true north too |
| 19:51 | <Philip`> | I prefer the approach of not really caring much about where I'm going, and just looking at the sun to work out which direction is more likely to be north than south |
| 19:51 | <jgraham> | annevk42: It is kind of a pain if your phone battery dies when your are days from anywhere :) |
| 19:52 | <jgraham> | Philip`: To be fair it took you months to find Sainsburys so I don't really look to you as a source of wisdom on navigation |
| 19:52 | <Hixie> | lol |
| 19:53 | <Philip`> | jgraham: True :-( |
| 19:54 | gsnedders | wonders how hard it is to find Sainsbury's |
| 19:54 | <gsnedders> | (I presume the Cambridge one, this is) |
| 19:54 | <gsnedders> | I mean, it's hardly as if it's that far from King's… |
| 19:55 | <Philip`> | It's easy to find - you just walk to Sidney Sussex, then turn around, and that's it |
| 19:56 | <Philip`> | It's right behind the Big Issue seller |
| 19:57 | <gsnedders> | I know where it is |
| 19:57 | <gsnedders> | I just wonder how it can be hard! |
| 19:57 | <Philip`> | It isn't |
| 19:57 | gsnedders | must confess he constantly forgets where Sidney Sussex is though |
| 19:57 | <Philip`> | It's right next to Sainsbury's |
| 19:58 | <gsnedders> | (And then I get told, "it's opposite Sainsbury's", and I headdesk at forgetting that again.) |
| 19:58 | Philip` | only knows where Sidney Sussex is because he had some supervisions there; he has no idea where any other colleges are |
| 19:59 | <gsnedders> | Philip`: Do you know where Clare is? Caius? |
| 19:59 | <gsnedders> | Catz? Queens'? |
| 20:01 | gsnedders | is still mildly amused by the fact that he seems to know Cambridge better than Philip` |
| 20:02 | <Philip`> | I could probably find them by walking around a bit and looking at the signs outside them, but I'd forget within a few hours |
| 20:16 | <aroben> | any idea where the definition of the Storage interface went? |
| 20:17 | <Hixie> | Web Storage |
| 20:17 | <Hixie> | http://dev.w3.org/html5/webstorage/ |
| 20:18 | <aroben> | thanks! |
| 20:43 | <gsnedders> | Philip`: Hint: They are all adjacent to King's |
| 21:14 | <annevk2> | Hixie, "contains-p" does not work as variable in ECMAscript; I suggest containsP ;) |
| 21:28 | <Hixie> | annevk2: that's not better :-) |
| 21:28 | <annevk2> | at least it works |
| 21:30 | <Hixie> | well, contains-p does too (nobody said that was js, after all) |
| 21:30 | <Hixie> | bbiab, lunch |
| 21:30 | <virtuelv> | contains_p |
| 21:31 | <annevk2> | yeah, seems very likely authors will try it somewhere else o_O |
| 21:33 | <Philip`> | Kristof would try it in VBScript, I guess |
| 21:36 | <annevk2> | so does it work there? |
| 21:38 | <Philip`> | No |
| 21:40 | <annevk2> | heh, that was rhetorical question |
| 21:41 | <Philip`> | And that was a rhetorical answer |
| 22:26 | <Hixie> | annevk2: fine fine :-P |
| 22:43 | <annevk2> | oops, forgot to look for DOMStringList usage within HTML5 |
