| 01:29 | <Hixie> | so apparently the security concern eric is worried about is an attacker smuggling a <base> in the <body> before a <script> with a relative src="" |
| 01:29 | <Hixie> | as far as I can tell, IE does indeed ignore <base href> outside <head> |
| 01:29 | <jamesr> | all your <base> are belong to us? |
| 01:29 | <Hixie> | that surprises me, though |
| 01:29 | <Hixie> | does anyone have any more detailed data on what IE does with <base>? |
| 01:29 | <Hixie> | my data is all out of date and i don't have a recent IE to test |
| 01:42 | <beverloo> | Would someone have a link to Aryeh's contenteditable / execCommand spec at hand? |
| 01:44 | <beverloo> | got it: http://aryeh.name/gitweb.cgi?p=editcommands;a=blob_plain;f=editcommands.html |
| 01:54 | <Hixie> | testing IE9 it seems the same as IE8, which is to say, in both quirks and non-quieks it ignores <base href> in the body |
| 01:54 | <Hixie> | i am surprised |
| 01:55 | <AryehGregor> | The blog post said IE has behaved that way since IE7. |
| 01:55 | <AryehGregor> | Why are you surprised? |
| 01:55 | <Hixie> | i didn't read the blog post |
| 01:55 | <AryehGregor> | Also, under what circumstance is an attacker going to be able to smuggle <base> into the body but not any other XSS vector? |
| 01:55 | <Hixie> | and i'm surprised because it seems like this would cause them compat issues |
| 01:56 | <AryehGregor> | He linked to this: http://blogs.msdn.com/b/ie/archive/2005/08/29/457667.aspx |
| 01:56 | <AryehGregor> | Which is, indeed, entitled "All your <base> are belong to us". |
| 01:56 | <Hixie> | the idea is that the victim has a blacklist approach and forgets <base> |
| 01:56 | <Hixie> | i knew about that blog post but i didn't realise it affected quirks mode as well |
| 01:56 | <Hixie> | they're usually so scared of changing quirks mode |
| 01:56 | <Hixie> | anyway |
| 02:05 | <Hixie> | blimey |
| 02:05 | <Hixie> | IE has 5 quirks modes? |
| 02:05 | <AryehGregor> | Hixie, maybe aCaseSensitive in WebKit does something along the lines of NFK{C,D}? |
| 02:06 | <Hixie> | AryehGregor: apparently it uses the platform user-friendly matching algorithm, or something |
| 02:06 | <AryehGregor> | I think IE10 has one quirks mode, three compatibility modes, and one standards mode. |
| 02:06 | <Hixie> | i think my wording is more accurate. :-P |
| 02:06 | <jamesr> | Hixie: http://hsivonen.iki.fi/doctype/ie8-mode.png |
| 02:07 | <jamesr> | but that doesn't include IE9/10 variations |
| 02:07 | <Hixie> | blimey |
| 02:08 | <Hixie> | their QA team must be... large |
| 02:08 | <Hixie> | or ill |
| 02:08 | <TabAtkins> | Both. |
| 02:11 | <TabAtkins> | Hixie: I thought that IE did the implied-body magic for known elements; it was only unknown elements (like the new sectioning elements) that got swallowed by the implied <head>. |
| 02:11 | <Hixie> | some known elements too |
| 02:11 | <TabAtkins> | Jeezus. |
| 02:12 | <AryehGregor> | IE clearly doesn't do an implied body for, say, <style>. |
| 02:12 | <AryehGregor> | In any version. |
| 02:13 | <TabAtkins> | Yes, of course not. That's a <head> element. |
| 02:13 | <Hixie> | tab is referring to the mail i just sent |
| 02:13 | <TabAtkins> | Hixie's claiming that <form> doesn't trigger implied-body. |
| 02:13 | AryehGregor | didn't see it |
| 02:13 | <AryehGregor> | Oh. That would be . . . insane. |
| 02:13 | <Hixie> | i literally just sent it seconds ago |
| 02:13 | <Hixie> | tab is just stalking me |
| 02:13 | <AryehGregor> | But then, all pre-HTML5 HTML parsers were insane, so . . . |
| 02:13 | <Hixie> | yeah well the HTML5 parser isn't that sane either |
| 02:20 | <AryehGregor> | Hixie, you just resolved <http://www.w3.org/Bugs/Public/show_bug.cgi?id=12570>; in r6314. |
| 04:13 | <Hixie> | wow, the patents part of the CLA for W3C Community Groups is surprisingly toothless |
| 04:14 | <Hixie> | it only covers your patents on parts of the spec that you actually write yourself |
| 04:17 | <Hixie> | also it seems there's a trivial way to get out of the licensing requirements: you just form an independant corporation to which you transfer your patents on the condition of receiving all royalties |
| 04:24 | <roc> | Hixie: that "independent corporation" approach actually works around a lot of patent agreements, even ones in which money changes hands |
| 04:25 | <roc> | "you have a cross-licensing agreement with us? Fine, we'll spin out our patents to an 'independent' entity to sue you" |
| 04:25 | <Hixie> | why would people not require that any license agreement follow the patents? |
| 04:26 | <roc> | you'd think |
| 04:26 | <roc> | maybe some do |
| 04:26 | <Hixie> | that's the kind of thing i do even when playing monopoly... |
| 04:27 | <roc> | This appears to be a case of that: http://thepriorart.typepad.com/the_prior_art/2010/04/mobilemedia-ideas-v-apple.html but I'm not 100% sure |
| 04:29 | <roc> | it may involve some kind of workaround where your staff develop some technology but the patent is never actually owned by your company, it goes straight to a holding company |
| 06:09 | <Hixie> | so if i add this download="" feature... is it something i shouldn't be adding to the w3c spec? |
| 06:10 | <Hixie> | i don't understand whether we're under a feature freeze or not |
| 06:55 | <nessy> | difficult call - not sure if an attribute falls under feature freeze - I'd think only big things such as adaptive streaming or an audio api |
| 08:11 | <matjas> | hmm, so no browser supports document.innerHTML yet? |
| 08:11 | <matjas> | Ms2ger (assuming you read the logs): http://ms2ger.freehostia.com/tests/html5/dynamic-markup-insertion/innerhtml-02.xhtml 404s |
| 09:38 | <annevk> | matjas, yeah, it's a new feature |
| 09:42 | <jgraham> | annevk: Is your nextNode stuff intended to be correct if the filter function manipulates theDOM? |
| 09:46 | <matjas> | annevk: is there an open Opera bug ticket for document.innerHTML or should I file one? |
| 09:53 | <annevk> | jgraham, when "iterator collection" is mutated referenceNode and pointerBeforeReferenceNode need to be updated, that is not defined yet |
| 09:53 | <annevk> | jgraham, however, once that is defined it should be alright |
| 09:54 | <annevk> | matjas, I can't find one so fast, feel free |
| 09:57 | <matjas> | annevk: done |
| 09:57 | <matjas> | Mozilla: http://bugzil.la/563320 WebKit: http://webk.it/60316 IE: https://connect.microsoft.com/IE/feedback/details/680040/implement-document-innerhtml |
| 10:04 | <jgraham> | annevk: Yes, I suppose that would work. And the detached flag. Although if I understand your intent correctly, it doesn't seem to match existing browsers http://software.hixie.ch/utilities/js/live-dom-viewer/saved/1080 |
| 10:04 | <jgraham> | In particular it seems they stop if they encounter a node they have seen before, but don't set pointerBeforeReferenceNode |
| 10:08 | <annevk> | jgraham, what about the detached flag? |
| 10:08 | <jgraham> | annevk: Maybe I don't understand what it is for |
| 10:09 | <jgraham> | Why do you check it after the filter function? What causes it to be set? |
| 10:10 | <annevk> | jgraham, the filter could set it by invoking detach() |
| 10:10 | <annevk> | jgraham, which reminds me, the current algorithm does not handle recursive invocation either |
| 10:11 | <annevk> | jgraham, I don't really understand the results from your NodeIterator btw |
| 10:11 | <annevk> | jgraham, it has SHOW_ELEMENT set yet Gecko and WebKit (not Opera) are happily returning Text nodes |
| 10:13 | <jgraham> | The text node never gets returned, but it does get set as the referenceNode |
| 10:13 | <annevk> | doh, I should read better |
| 10:14 | <jgraham> | (which also doesn't match the spec I guess) |
| 10:16 | <annevk> | Can that situation only occur at the end of the collection and during modifications? |
| 10:17 | <annevk> | I could modify step 3.1 to account for it |
| 10:19 | <jgraham> | I imagine that internally they update the referenceNode irrespective of whether the node matches the filter |
| 10:19 | <jgraham> | So you would just need to move that inside the loop |
| 10:20 | <jgraham> | I think it is only observable after modification or at the end though |
| 10:20 | <annevk> | that does not work if something throws an exception |
| 10:22 | <annevk> | if you update referenceNode all the time and filter throws an exception or sets detach() it will have the wrong value |
| 10:24 | <jgraham> | It seems to get set to null in that case |
| 10:25 | <jgraham> | Oh, only in gecko |
| 10:25 | <jgraham> | In WebKit it seems to be the node that you were trying to filter |
| 10:26 | <jgraham> | So you could set referenceNode to node at the start of the loop |
| 10:28 | <annevk> | And set it to null if an exception was raised or detach() was called? And that makes the whole thing no longer function? |
| 10:28 | <jgraham> | http://software.hixie.ch/utilities/js/live-dom-viewer/saved/1081 is what I looked at |
| 10:30 | <annevk> | It does not even throw an exception in Gecko? |
| 10:31 | <annevk> | oh it does |
| 10:31 | <annevk> | you were not printing that |
| 10:32 | <annevk> | if you just do throw "x" referenceNode does not become null |
| 10:33 | <annevk> | it is only null when things are detached |
| 10:34 | <annevk> | so invoking detach() just initializes referenceNode to null |
| 10:34 | jgraham | wonders what the point of detach is |
| 10:35 | <annevk> | so you can clear up resources |
| 10:35 | <annevk> | silly legacy thing |
| 10:38 | <annevk> | thanks for helping out btw |
| 10:38 | <jarek> | Hi |
| 10:38 | <jarek> | I'm trying to implement CSS parser, but it's rather hard for me to understand some of the regular expressions used in specification |
| 10:38 | <jarek> | e.g.: \\[0-9a-f]{1,6}(\r\n|[ \n\r\t\f])? |
| 10:39 | <jarek> | is there somewhere a document written in human language that would explain which strings are allowed for a given token? |
| 10:40 | <annevk> | other than the CSS specification and its dependencies? I do not think so |
| 10:40 | <jarek> | so I will have to read a book on regular expressions in order to understand the specification? :/ |
| 10:42 | <annevk> | no |
| 10:42 | <annevk> | something on yacc |
| 10:43 | <jarek> | I have to implement this with javascript |
| 10:43 | <jarek> | afaik there is no such tool like yacc that would generate js code |
| 10:43 | <jgraham> | Well you will likely need to understand regexp, yes |
| 10:45 | <jgraham> | And there are some parser-generators avaliable that will create javascript |
| 10:45 | <jgraham> | But your chances of success if you don't understand the input seem rather small |
| 10:46 | <jarek> | there is peg.js, but it genarates the whole parser (I just need tokenizer) |
| 10:48 | <jgraham> | Well I would read http://flex.sourceforge.net/manual/Patterns.html#Patterns and convert to javascript regexp |
| 10:48 | <jarek> | also, CSS2 and CSS3 specifications seem to be using slightly different tokens |
| 10:48 | <jarek> | e.g. there is BAD_STRING token in CSS2, but I can't find it in CSS3 specs |
| 10:49 | <annevk> | CSS3 is out of date |
| 10:49 | <annevk> | ironically enough |
| 10:52 | <jarek> | annevk: should I implement the tokenizer according to CSS2 specification and later update it when there is new revision of http://www.w3.org/TR/css3-syntax/? |
| 10:52 | <jarek> | or perhaps I should use http://www.w3.org/TR/css3-syntax/ as the only reference? |
| 10:53 | <annevk> | jgraham, so I tested again and I think referenceNode only becomes an "invisible" node due to modification |
| 10:54 | <annevk> | jarek, you should use CSS2 with modifications |
| 10:54 | <annevk> | jarek, css3-syntax is hopelessly out of date |
| 11:08 | <annevk> | ah |
| 11:08 | <annevk> | recursive calls throw an exception as well |
| 11:08 | <annevk> | in WebKit some kind of callstack exception |
| 11:08 | <annevk> | in Gecko INVALID_STATE_ERR |
| 11:09 | <annevk> | in Opera the same as in Gecko |
| 11:10 | <annevk> | in WebKit invoking a different method (e.g. previousNode) does work, it does not in Opera/Gecko |
| 11:11 | <annevk> | I guess I'll side with Opera/Gecko |
| 11:11 | <jgraham> | simplicity ++ |
| 11:12 | <Ms2ger> | specs++ |
| 11:13 | <annevk> | name for this flag? |
| 11:13 | <annevk> | "locked flag"? |
| 11:14 | <Ms2ger> | wfm |
| 11:14 | <jgraham> | yeah, whatever |
| 11:15 | <Ms2ger> | For detach(), s/initialize/set/? |
| 11:19 | <annevk> | I use initialize for readonly attributes |
| 11:19 | <annevk> | because they cannot really be set |
| 11:20 | <jgraham> | I'm not sure that is better |
| 11:21 | <jgraham> | Initialise implies that it is being set at the point it is created |
| 11:25 | <annevk> | better term? |
| 11:25 | <annevk> | set just seems wrong |
| 11:25 | <annevk> | but maybe we should use it anyway |
| 11:25 | <bga_> | lol http://xkcd.com/927/ |
| 11:25 | <annevk> | for now: http://dvcs.w3.org/hg/domcore/raw-file/tip/Overview.html#dom-nodeiterator-nextnode |
| 11:26 | <jgraham> | annevk: Better than initialise. I think it is understood that the redonlyness of attributes from the external point of view doesn't matter to internal algorithms |
| 11:30 | <AnselmBradford> | hi all, does anyone know of a browser that supports the scoped attribute on the style element? |
| 11:45 | <annevk> | jgraham, done |
| 11:45 | <annevk> | Anyone an idea how to generalize 3.1 and 3.2 so we can have one algorithm for nextNode() and previousNode()? |
| 12:00 | <annevk> | WebKit allows arguments to be omitted when creating a NodeIterator but defaults whatToShow to 0 |
| 12:00 | <annevk> | oops |
| 12:00 | <annevk> | Guess that is because it is the same as undefined for them |
| 12:35 | <Ms2ger> | matjas, updated the link to my innerhtml tests in the spec |
| 12:35 | <matjas> | Ms2ger: \o/ |
| 12:36 | <annevk> | I am going to use dl class=switch btw |
| 12:36 | <remysharp> | If I wanted to delete localStorage using `delete localStorage.foo` it works, but am I asking for trouble as it's not spec'ed out (or that I've seen) |
| 12:38 | <annevk> | it is specified |
| 12:38 | <annevk> | "deleter void removeItem(in DOMString key)" |
| 12:38 | <annevk> | see Web IDL |
| 12:39 | <remysharp> | honestly, I'm not very good at reading Web IDLs - in fact, I don't think I even know what IDL stands for - interface something lang? ? |
| 12:39 | <Ms2ger> | description, I guess |
| 12:39 | <annevk> | definition I thought |
| 12:40 | <Ms2ger> | The "deleter" means that function is called when you do `delete` |
| 12:40 | <remysharp> | glad I'm not the only one then :) |
| 12:40 | <annevk> | see http://dev.w3.org/2006/webapi/WebIDL/ |
| 12:40 | <remysharp> | cool - cheers for the clarification annevk & Ms2ger |
| 12:40 | <Ms2ger> | Np |
| 12:41 | <Ms2ger> | Oh hey, :dir |
| 12:41 | <annevk> | what about it? |
| 12:42 | <annevk> | though now you mention it, I wonder why that is not :ltr and :rtl |
| 12:43 | <Ms2ger> | Aharon filed a Gecko bug to implement it |
| 12:47 | <annevk> | http://dvcs.w3.org/hg/domcore/raw-file/tip/Overview.html#concept-nodeiterator-traverse |
| 12:57 | <annevk> | does this sound correct: |
| 12:58 | <annevk> | nm |
| 12:58 | <annevk> | it doesn't |
| 13:00 | <matjas> | annevk: to support :dir(upsidedown) ;) |
| 13:01 | <annevk> | matjas, isn't dir just for writing direction? |
| 13:02 | <annevk> | or whatever the proper term for that is |
| 13:02 | <matjas> | </lame-joke> |
| 13:03 | <annevk> | vertical text exists of course, but is not a semantic thing (at least that's how it is being approached) |
| 13:19 | <MikeSmith> | AnselmBradford: there is a Webkit developer who's been implementing it |
| 13:19 | <MikeSmith> | but I don't know if the code landed yet or not |
| 13:19 | <MikeSmith> | I think not though |
| 13:21 | <AnselmBradford> | MikeSmith: thanks :) |
| 13:25 | <AnselmBradford> | MikeSmith: Any idea where I'd be able to find out when it lands ... http://trac.webkit.org/timeline or? |
| 13:25 | <MikeSmith> | AnselmBradford: hang on, I will get you the bug number |
| 13:26 | <MikeSmith> | https://bugs.webkit.org/show_bug.cgi?id=49142 |
| 13:29 | <AnselmBradford> | cheers |
| 13:39 | <annevk> | mutation feedback requested |
| 13:39 | <annevk> | if the mutated node is not a descendant of root and not an ancestor of the reference node terminate these steps |
| 13:39 | <annevk> | if pointerBeforeReferenceNode is false and there is a node preceding where referenceNode was previously, set referenceNode to that node |
| 13:39 | <annevk> | if pointerBeforeReferenceNode is true and there is a node following where referenceNode was previously, set referenceNode to that node |
| 13:39 | <annevk> | if pointerBeforeReferenceNode is true and there is no node following where referenceNode was previously, set referenceNode to the node preceding where referenceNode was previously and set pointerBeforeReferenceNode to false |
| 13:39 | <annevk> | I don't think there are any other conditions |
| 13:40 | <jgraham> | When do these steps run? Does it apply irrespective of which node is mutated? |
| 13:43 | <annevk> | whenever the "iterator collection" is mutated |
| 13:44 | <annevk> | maybe once mutation is defined a little better we can have more details |
| 13:45 | <annevk> | but for now this seems sufficient |
| 13:45 | <jgraham> | Oh, so this is supposed to be a seperate algorithm… |
| 13:45 | <annevk> | http://dvcs.w3.org/hg/domcore/raw-file/tip/Overview.html#iterator-collection |
| 13:47 | <jgraham> | What does mutating a collection mean? Is it only operations that change the order of the collection or any operation that mutates nodes in the collection? |
| 13:48 | <annevk> | yeah |
| 13:48 | <annevk> | I guess it should say "whenever the iterator collection has nodes inserted or removed" |
| 13:49 | <jgraham> | Does that work if I remove a node and insert it somewhere else? |
| 13:49 | <jgraham> | I mean should the algorithm run once or twice? |
| 13:50 | <jgraham> | (also, with replaceChild) |
| 13:51 | <annevk> | does it matter if it runs once or twice? |
| 13:51 | <jgraham> | I don't know :) I haven't fully understood it yet |
| 13:52 | <jgraham> | In fact I'm not sure I understand it at all |
| 13:52 | <jgraham> | e.g. the second step says "if there is a node before where referenceNode was previously" |
| 13:52 | <jgraham> | But reference node might not have been moved |
| 13:53 | <jgraham> | And I assume the intent isn't to move referenceNode in that case |
| 13:54 | <annevk> | look at the first clause |
| 13:55 | <annevk> | I should add and is not the reference node to the first clause |
| 13:57 | <annevk> | jgraham, so the algorithm is terminated if referenceNode is not affected |
| 13:58 | <hsivonen> | Hmm. so when you set innerHTML on a title element, the appropriate end tag remains undefined? Firefox sets it to "title". Ragnarök and Chrome don't. |
| 13:58 | <hsivonen> | see http://software.hixie.ch/utilities/js/live-dom-viewer/saved/1084 |
| 13:59 | <annevk> | the behavior of Gecko is not really nice |
| 14:00 | <jgraham> | annevk: Yes, OK |
| 14:01 | <annevk> | thanks: http://dvcs.w3.org/hg/domcore/raw-file/tip/Overview.html#iterator-collection |
| 14:01 | <hsivonen> | Hixie: if the Gecko behavior isn't what you meant, it would have been nice to have a note affirming that in the spec |
| 14:02 | <hsivonen> | Hixie: for document.createElement("title").innerHTML = "</title>"; that is |
| 14:02 | <jgraham> | annevk: "where the referenceNode was was previously" isn't too clear |
| 14:02 | <annevk> | yeah |
| 14:02 | <annevk> | not sure how to do that |
| 14:02 | <annevk> | or do it better, that is |
| 14:04 | <jgraham> | What do you actually mean? If I have [a b c * d e] and mutate it to [c * d a b e], where * is the reference node, which node do you have in mind? |
| 14:04 | <jgraham> | e? |
| 14:06 | <annevk> | * is not the reference node |
| 14:06 | <annevk> | it's either [c] or [d] |
| 14:07 | <annevk> | you have a reference node |
| 14:07 | <annevk> | and an indicator that is either before or after the reference node |
| 14:07 | <annevk> | for direction |
| 14:09 | <jgraham> | I don't understand :) |
| 14:09 | <jgraham> | Why isn't it possible to set up a situation like I described? |
| 14:10 | <annevk> | the reference node is an actual node |
| 14:10 | <jgraham> | Yes, I meant * to be an actual node |
| 14:10 | hsivonen | filed https://bugzilla.mozilla.org/show_bug.cgi?id=672775 |
| 14:10 | <jgraham> | Sorry, maybe that wasn't clear |
| 14:10 | <annevk> | if you mutate it to that the referenceNode is not affected so nothing happens |
| 14:11 | <annevk> | only the list mutates |
| 14:11 | <annevk> | but how that happens is clear from the definition of collection |
| 14:11 | <annevk> | or follows from |
| 14:11 | <hsivonen> | so... |
| 14:11 | <jgraham> | Well the mutated node can be a desendant of root and an ancestor of referenceNode |
| 14:12 | <jgraham> | in my example |
| 14:12 | <jgraham> | If * is a child of c in the actual tree |
| 14:12 | <hsivonen> | if innerHTML is being set on style, xmp, iframe, noembed, noframes or script, is there any use in putting the tokenizer in the RAWTEXT or script data state without an appropriate end tag instead of just putting the tokenizer in the PLAINTEXT state? |
| 14:14 | <hsivonen> | doh. there actually *is* already a green note in the spec saying all these things |
| 14:14 | <hsivonen> | I wonder when that note appeared |
| 14:14 | <hsivonen> | could I have missed it when I implemented fragment parsing? hard to believe |
| 14:14 | <jgraham> | hsivonen: I seem to remember looking at this at some point. But I don't know if I caused a note to appear |
| 14:15 | <annevk> | jgraham, how is c mutated there? |
| 14:16 | <annevk> | oh you move c forward |
| 14:17 | <hsivonen> | Sadly, Atom ended up like this: http://www.xkcd.com/927/ |
| 14:17 | <annevk> | jgraham, depending on direction it should be either b or e |
| 14:17 | hsivonen | checks the www-tag archive to see if the xkcd URL has already been posted there |
| 14:18 | <hsivonen> | not according to list search |
| 14:18 | <annevk> | nope |
| 14:18 | <annevk> | you're not subscribed? |
| 14:18 | <hsivonen> | annevk: I'm not |
| 14:23 | <hsivonen> | annevk: maybe I should subscribe, though. I have a number of other W3C list subscribed and going into a bucket that I don't read actively but occasionally browse and respond to |
| 14:23 | <annevk> | jgraham, i would love a better way to have that described though |
| 14:23 | <hsivonen> | (in the case of www-tag, I read the archives and then copy and paste when replying) |
| 14:24 | <hsivonen> | (I probably end up reading a smaller propertion of www-style messages than www-tag messages even though I subscribe to www-style) |
| 14:26 | <jgraham> | annevk: Yeah, it is not quite trivial. I guess I might try talking about the state of the collection before and after the mutation, with some names for the set of elements that is moved during the mutation, the set that is deleted, the set that is inserted, and the rest |
| 14:27 | <jarek_> | does CSS3 allow comments between "!" and "important"? |
| 14:27 | <jarek_> | according to specification it's not allowed, but for some reason it does validate on http://jigsaw.w3.org/ |
| 14:28 | <annevk> | jgraham, I guess the only problem is removal of the referenceNode |
| 14:28 | <annevk> | if the referenceNode is removed by itself you simply need the prev or next node |
| 14:29 | <annevk> | if it's an ancestor you need the ancestor prev or next node |
| 14:29 | <annevk> | not sure what else is there |
| 14:29 | <annevk> | replaceChild I guess |
| 14:30 | <annevk> | should prolly treat that as removal as well |
| 14:40 | <annevk> | jgraham, http://dvcs.w3.org/hg/domcore/raw-file/tip/Overview.html#iterator-collection |
| 14:41 | <Ms2ger> | jarek_, does the spec say it explicitly? You can usually put comments anywhere between tokens |
| 14:41 | <Ms2ger> | (Assuming ! and important are two tokens) |
| 14:42 | <jgraham> | annevk: Saying "referenceNode attribute value" everywhere rather than "referenceNode" is bad |
| 14:42 | <jarek_> | Ms2ger: CSS3 Syntax module defines IMPORTANT_SYM token as "!{w}important" |
| 14:42 | <jarek_> | CSS2 defines it as: "!"({w}|{comment})*{I}{M}{P}{O}{R}{T}{A}{N}{T} |
| 14:42 | <jgraham> | It is confusing because of DOM attributes |
| 14:43 | <jgraham> | and verbose |
| 14:43 | <Ms2ger> | -syntax is irrelevant |
| 14:43 | <Ms2ger> | As was said earlier |
| 14:43 | <annevk> | jgraham, that style is rather consistently used |
| 14:43 | <annevk> | jgraham, and besides the point |
| 14:43 | <jgraham> | annevk: OK, that style rather consistently sucks :) |
| 14:44 | <annevk> | yeah whatever |
| 14:46 | <annevk> | pretty pleased with this new definition |
| 14:47 | <jgraham> | annevk: It is still not clear what the "node preceeeding the node that was removed" is |
| 14:48 | <jgraham> | A node that isn't in the collection doesn't have a node preceding it |
| 14:50 | <annevk> | jgraham, if it isn't in the collection it would already terminate |
| 14:50 | <annevk> | jgraham, again, see step 1 of the algorithm |
| 14:51 | <jgraham> | annevk: Hmm? If I remove a node from a collection, how can it also be in the collection? |
| 14:51 | <jgraham> | It *was* in the collection before it was remove |
| 14:51 | <jgraham> | d |
| 14:52 | <annevk> | oh right |
| 14:52 | <annevk> | this needs to run before they are actually removed |
| 14:53 | <jgraham> | OK. It should also be clear what the :mutated node" is. Especially since the algorithm is run "whenever nodes [plural] are removed from the collection" |
| 14:53 | <annevk> | jgraham, replacing "was" with "is being" helps you? |
| 14:54 | <annevk> | <p>Whenever a <span title=concept-node>node</span> is removed from the |
| 14:54 | <annevk> | <span>iterator collection</span>, these steps must be run: |
| 14:54 | <annevk> | and that? |
| 14:54 | <hsivonen> | hmm. it's an unusual situation that Sam's latest blog post is well over a month old |
| 14:54 | <annevk> | add being after is again |
| 14:55 | <jgraham> | I thought you wanted to add :before"? |
| 14:55 | <jgraham> | s/:/"/ |
| 14:56 | <annevk> | oh |
| 14:56 | <annevk> | like |
| 14:56 | <annevk> | <p>Before a <span title=concept-node>node</span> is removed from the |
| 14:56 | <annevk> | <span>iterator collection</span>, these steps must be run: |
| 14:56 | <annevk> | could do that too |
| 14:57 | <annevk> | refresh |
| 15:00 | <jgraham> | annevk: Well I can't be sure it is right, but at least I understand it now :) |
| 15:01 | <annevk> | yeah, needs more tests and stuff |
| 15:06 | danj | Crys over Vagrant |
| 15:07 | <annevk> | TreeWalker.root is fairly meaningless, no? |
| 15:07 | <annevk> | I wonder why that exists |
| 15:07 | <annevk> | oh never mind |
| 15:07 | <annevk> | although |
| 15:07 | <annevk> | " |
| 15:07 | <annevk> | Moves to and returns the closest visible ancestor node of the current node. If the search for parentNode attempts to step upward from the TreeWalker's root node, or if it fails to find a visible ancestor node, this method retains the current position and returns null." |
| 15:07 | <annevk> | does not seem to account for currentNode being outside the TreeWalker root |
| 15:23 | <annevk> | http://mxr.mozilla.org/mozilla-central/source/content/base/src/nsTreeWalker.cpp#143 I wonder why that security check is there |
| 15:33 | <Ms2ger> | https://bugzilla.mozilla.org/show_bug.cgi?id=156452 |
| 15:34 | <Ms2ger> | if you hadn't found it yet |
| 15:41 | <annevk> | hmm |
| 15:50 | <smaug____> | that is really old code |
| 15:50 | <gsnedders> | Motherboard with processor/memory given ATX2 power cables plugged in should turn on it's power-on LED, right? |
| 15:50 | <smaug____> | and as far as I know, not needed anymore |
| 16:34 | <dglazkov> | good morning, Whatwg! |
| 16:38 | <AryehGregor> | Good morning, dglazkov! |
| 16:41 | <annevk> | I have the feeling scrolling is now the opposite direction in OS X Lion |
| 16:41 | <annevk> | That is somewhat confusing |
| 16:41 | <zewt> | heh people complaining about that simultaneously in two channels i'm in |
| 16:43 | <AryehGregor> | What does "the opposite direction" mean? |
| 16:44 | <annevk> | fingers up, content goes up |
| 16:45 | <annevk> | well, move fingers from top to bottom and you scroll upwards |
| 16:45 | <annevk> | used to be the other way around |
| 16:46 | <zewt> | it says a lot about how much a company cares about their users when they're willing to destroy users' hard-baked UI habits and muscle memory approaching a decade and a half of use |
| 16:46 | <AryehGregor> | That's the way touchscreens all work, right? |
| 16:47 | <AryehGregor> | The idea being that you're sort of pushing the content down while the screen stays in place. |
| 16:47 | <jgraham> | That sounds awful |
| 16:47 | <annevk> | AryehGregor, yeah, it's like on the iPhone |
| 16:47 | <AryehGregor> | As though it's printed on a little piece of paper stuck behind the screen that you're physically pushing. |
| 16:47 | <annevk> | except there is no screen under my fingers |
| 16:50 | Philip` | presumes it's futile to ask if the scroll direction is a configurable option |
| 16:51 | <annevk> | you can configure it |
| 16:52 | <Philip`> | Oh, that's fortunate |
| 16:52 | <annevk> | it's a checkbox called "scroll direction: natural\n content tracks finger movement" |
| 16:55 | <annevk> | I guess I'll try to get used to it |
| 16:55 | <zewt> | and then good luck using any other system, heh |
| 16:56 | <annevk> | oh, Safari has two finger sideways scrolling for history |
| 16:56 | <annevk> | quite neat |
| 16:57 | <annevk> | zewt, they already trip me up with the keys |
| 16:57 | <zewt> | i use mousewheel tilt for that |
| 16:57 | <annevk> | the way it is animated is quite nice |
| 16:57 | <zewt> | annevk: heh i had to use windows in a vm on a macbook for a week or so once, and the buttons that were mapped to alt and windows key were flipped |
| 16:57 | <zewt> | drove me absolutely insane for a few days, then as I was getting used to it I went home and was back to a regular keyboard, where I could go insane again |
| 16:58 | <annevk> | computers are terrible |
| 17:05 | <swarren08> | annevk you can change the way it scrolls in the new Lion |
| 17:06 | <swarren08> | under trackpad / scroll & zoom , uncheck the scroll directional |
| 17:09 | <annevk> | I know |
| 17:10 | <swarren08> | cool. So what do you think of it? |
| 17:12 | <annevk> | dunno, I mostly went back to work on DOM Traversal |
| 17:12 | <annevk> | looks a bit neater though, and being able to play albums directly from the screen saver is a nice touch |
| 17:12 | <zewt> | (except now all node traversal will be backwards) |
| 17:14 | <swarren08> | I cant wait to get myself a mac |
| 17:14 | <annevk> | can't wait for competition to catch up personally |
| 17:14 | <annevk> | but it seems like that might be a while |
| 17:15 | <swarren08> | that is true |
| 17:16 | <swarren08> | apple seems to come up with this technology faster than others, which is great |
| 17:16 | <swarren08> | obviously it is helping them out immensly. Did anyone happen to see the stats on the recent quarter for apple? it was definitly insane to see |
| 17:26 | <annevk> | the Gecko and WebKit TreeWalker implementations look remarkably similar |
| 17:27 | <Ms2ger> | How do they compare to Opera's? |
| 17:28 | <annevk> | dunno and could not tell you either |
| 17:46 | <annevk> | writing everything down is a little more complicated than I initially thought |
| 17:47 | <annevk> | though I guess three-five days is not that bad |
| 17:47 | <annevk> | events was just simpler |
| 17:50 | <Hixie> | hsivonen: no idea to what you are referring, but i'm happy to add notes to the spec, drop me a mail or a bug |
| 17:51 | <hober> | annevk: you'll get used to the new scrolling in a couple of days, and then you won't want to go back. |
| 17:55 | <jgraham> | hober: Spoken like a good mini-Steve |
| 17:56 | <jgraham> | :) |
| 17:56 | <hober> | jgraham: :) |
| 18:04 | <AryehGregor> | Hixie, you resolved <http://www.w3.org/Bugs/Public/show_bug.cgi?id=12570>; in r6314. You should close the bug as FIXED. |
| 18:05 | <AryehGregor> | (I don't know why we have an informative reference to PDF but not the zillions of other standards mentioned in non-normative text, though) |
| 18:10 | <Hixie> | looks like it's already marked fixed |
| 18:10 | <Hixie> | the pdf one is mentioned in normative text |
| 18:10 | <Hixie> | or at least, in main prose, as opposed to notes and examples |
| 18:10 | <Hixie> | if there's anything else you think should get referneces, let me know |
| 18:23 | <jarek_> | CSS3 selectors specification defines NOT token as ":"{N}{O}{T}"(" while CSS3-mediaqueries specification defines a token with the same name as {N}{O}{T} |
| 18:24 | <TabAtkins_> | Yes, it's inconsistent. Been pointed out before, hasn't been fixe dyet. |
| 18:24 | <AryehGregor> | Oh, Ms2ger marked it FIXED even though technically only the editor is supposed to. Oh well, I doubt anyone will care. |
| 18:25 | <AryehGregor> | Hixie, I didn't see any reference to PDF in normative text. It's mentioned in the prose, but only preceded by "e.g." or "for example". |
| 18:25 | <Ms2ger> | Meh, I assume the guy won't try to process-troll me over getting his bug fixed |
| 18:25 | <AryehGregor> | Well, that's why it's an informative reference, of course. |
| 18:25 | <jarek_> | perhaps it would be easier if I had written two tokenizers - one for selectors and another one for everything else (the second parser would call the first one) |
| 18:25 | <jarek_> | CSS3 selectors are pretty complex |
| 18:25 | <AryehGregor> | But I don't see the difference between entire paragraphs marked as examples or notes, and sentences within normative text marked as examples or notes. |
| 18:26 | <AryehGregor> | Still, it makes no difference in the end. |
| 18:26 | <AryehGregor> | As I said in the bug, examples of other formats mentioned in examples include H.264, AAC, MP4, MPEG-4, |
| 18:26 | <AryehGregor> | AMR, 3GPP, Theora, Vorbis, Ogg, Speex, FLAC, and Dirac, and that's without even trying to find them. |
| 18:43 | <MikeSmith> | AryehGregor: yep |
| 18:44 | <MikeSmith> | dude who filed that bug is basically trolling, actually |
| 18:45 | <MikeSmith> | that seems to be his whole M.O., a |
| 18:45 | <Hixie> | AryehGregor: i'm happy to include more references |
| 18:45 | <Hixie> | AryehGregor: file a bug or send mail letting me know what you want references for |
| 18:46 | <MikeSmith> | Hixie: well, don't feed the trolls man |
| 18:47 | <MikeSmith> | I don't think AryehGregor is suggesting more references |
| 18:49 | <Hixie> | i'm all for adding more references |
| 18:49 | <MikeSmith> | yeah, well |
| 18:50 | <MikeSmith> | the point is whether they are actually useful |
| 18:50 | <MikeSmith> | as opposed to gratuitous |
| 18:50 | <Hixie> | we are _long_ past the point of gratuitous |
| 18:51 | <Ms2ger> | Isn't adding gratuitous references a stated goal of the HTMLWG? |
| 18:52 | <MikeSmith> | further gratuoisity proliferation us not a win |
| 18:52 | <MikeSmith> | or however you spell that, christ |
| 18:53 | <Hixie> | i don't think the references make any difference whatsoever :-) |
| 18:53 | <MikeSmith> | Ms2ger: good example of trolling-trolling |
| 18:53 | <MikeSmith> | clever |
| 18:54 | Ms2ger | likes trolling trolls |
| 18:54 | <MikeSmith> | Hixie: good, remove all of them, then |
| 18:55 | <Ms2ger> | MikeSmith, agreed, and we should import all other web specs into WA1.0 at the same time :) |
| 18:55 | <MikeSmith> | yeah, let's make one giant monolithic spec |
| 18:56 | <MikeSmith> | that is clearly a great thing |
| 18:56 | <Ms2ger> | It worked for CSS! |
| 18:57 | <Hixie> | i'd much rather we had one spec than 100 |
| 18:58 | <jcranmer> | maybe here's a better place |
| 18:58 | <jcranmer> | is there a nice way in CSS to get a height that's the same as the line-height? |
| 18:59 | <TabAtkins_> | No. |
| 18:59 | <TabAtkins_> | (Unless the line-height is the same as the font-size.) |
| 18:59 | <jcranmer> | so my best bet is to set the line-height myself and hope for the best? |
| 19:01 | <MikeSmith_> | Hixie: me too |
| 19:01 | <MikeSmith_> | honestly |
| 19:02 | <MikeSmith> | but some delusional people seem to prefer specs that are less than 5MB |
| 19:03 | <Hixie> | i don't mind having a multipage copy |
| 19:03 | <Hixie> | but if the concern is just having lots of pages to read, you get fewer total pages if you have 1 spec than 100 |
| 19:04 | <Hixie> | since each spec has a lot of duplicate boilerplate, cross-spec stuff, etc |
| 19:04 | <MikeSmith> | true |
| 19:05 | <MikeSmith> | I think CSS as gone to the other extreme |
| 19:05 | <Ms2ger> | Splitting specs is fine, IMO, if you can find good split-lines |
| 19:05 | <Ms2ger> | These are almost non-existent, unfortunately |
| 19:05 | <TabAtkins_> | jcranmer: yes. |
| 19:06 | <TabAtkins_> | Ms2ger: That's not true. For example, different layout algos for CSS can split apart pretty easily. |
| 19:06 | <Ms2ger> | I was talking WA1.0 mostly |
| 19:06 | <Hixie> | CSS has a number of problems, dunno if splitting is a problem itself there or a solution that's just been swamped by other issues |
| 19:06 | <Hixie> | CSS definitely has a "velocity" problem |
| 19:07 | <Ms2ger> | Also, editors |
| 19:07 | <Hixie> | but i think that's more an issue of the editing style |
| 19:07 | <TabAtkins_> | Editors are our largest problem. Secondary is that we're bound by W3C process, which made 2.1 a glorious mess. |
| 19:07 | <Hixie> | yeah |
| 19:08 | <Hixie> | i've been speaking with jeff and i have no faith that he's interesting in changing the process |
| 19:09 | <jcranmer> | hmm |
| 19:10 | <jcranmer> | 9pt/13px |
| 19:10 | <TabAtkins_> | We should add CSS unit conversion to google calc. |
| 19:12 | <jcranmer> | so 9pt = 12px |
| 19:13 | <jcranmer> | how about I call it 0.9em and everyone's happy? |
| 19:13 | <AryehGregor> | MikeSmith, Hixie: Personally I don't include any references at all in the edit commands spec. The spec is already going to be horribly broken if you try doing something stupid like printing it out, since all the xrefs will be broken, so I'm just linking to sections of other specs directly when I want to reference something. |
| 19:13 | <beverloo> | TabAtkins: 1vm = ?, 1% = ?. Too much depends on the context imo |
| 19:13 | <AryehGregor> | beverloo, do it for absolute units only. |
| 19:13 | <TabAtkins_> | beverloo: Yeah, only the physical units I mean. |
| 19:13 | <jcranmer> | pt, pc, px, in, cm |
| 19:13 | <TabAtkins_> | Converting between in, pt, px |
| 19:14 | <jcranmer> | mm |
| 19:14 | <beverloo> | How does that handle dpi? |
| 19:14 | <TabAtkins_> | The ratio is fixed in CSS. |
| 19:14 | <TabAtkins_> | 1in = 96px |
| 19:14 | <jcranmer> | the ratios is essentially fixed |
| 19:14 | <jcranmer> | it's way too broken otherwise |
| 19:14 | <TabAtkins_> | Not essentially, actually. ^_^ |
| 19:14 | <beverloo> | Fair enough |
| 19:15 | <jcranmer> | px: pixel units — 1px is equal to 0.75pt. |
| 19:15 | <jcranmer> | ah, it is fixed |
| 19:15 | <AryehGregor> | As of like a year ago. |
| 19:15 | <AryehGregor> | Before that it was unofficially not fixed but mostly fixed in practice. |
| 19:15 | <TabAtkins_> | Was it wrong before, jcranmer? |
| 19:15 | <jcranmer> | I thought it was only in CSS3 |
| 19:15 | <TabAtkins_> | Oh, the px/in thing. Yeah. |
| 19:15 | <AryehGregor> | I think he means "fixed" as opposed to "variable", not as opposed to "broken". |
| 19:15 | <TabAtkins_> | We set it when we were resolving 2.1 issues. |
| 19:16 | <jcranmer> | I should say |
| 19:16 | <jcranmer> | people assume 96dpi in too many pages, that to make any other value would break the web |
| 19:16 | <jcranmer> | that's what I meant by "it's way too broken otherwise" |
| 19:16 | <TabAtkins_> | Yes, exactly. |
| 19:19 | <TabAtkins_> | Huh. I never realized that I had to actually put the </script> for the script to execute. |
| 19:19 | <TabAtkins_> | (I'm testing some interfaces before I write the real page.) |
| 19:20 | <AryehGregor> | I only realized that when I started using Live DOM viewer. |
| 19:32 | <hsivonen> | Hixie: it turned out that the note I was asking for already exists |
| 19:36 | <hsivonen> | is there something genuinely relative unusual about Swedish names that I'm not aware of or is this person just throwing an arbitrary collection of country names out there for the sake of illustration? http://www.kalzumeus.com/2010/06/17/falsehoods-programmers-believe-about-names/ |
| 19:38 | <Hixie> | hsivonen: excellent :-) |
| 19:39 | <swarren08> | Whats the best way to learn HTML5. Im going to be honest, watching everyone talk makes me feel like i know nothing. |
| 19:41 | <Philip`> | Spend a year reading the spec and the mailing lists and IRC :-) |
| 19:42 | <hsivonen> | swarren08: people on this channel develop HTML5 itself and may, therefore, not have a good idea of what's it like to learn for others |
| 19:42 | <hsivonen> | swarren08: but what Philip` says should work |
| 19:42 | <swarren08> | its amazing watching you guys |
| 19:43 | <jarek_> | swarren08: I would from https://developer.mozilla.org |
| 19:43 | <jgraham> | hsivonen: Maybe the way that Swedish sorts characters being different to e.g. German? |
| 19:43 | <jarek_> | s/I would/I would start |
| 19:43 | <jgraham> | Philip`: You are mean :) |
| 19:44 | <swarren08> | So you guys actually develop HTML5? or what you call the Living Standard? |
| 19:44 | <jgraham> | hsivonen: Although that isn't really about names. I'm not aware of anything common and strange about Swedish names |
| 19:45 | <jgraham> | swarren08: Nah Hixie does that. We just hold the palm leaves and feed him grapes |
| 19:45 | <AryehGregor> | swarren08, yes. |
| 19:45 | <AryehGregor> | Also, what jgraham says. |
| 19:45 | <Hixie> | wait, i get grapes?! |
| 19:45 | <Hixie> | where is this! |
| 19:45 | <swarren08> | ha ha ha |
| 19:45 | <AryehGregor> | We're really all Hixie's minions, he does all the work. |
| 19:45 | <AryehGregor> | We just do menial labor for him and take some of the credit. |
| 19:45 | <swarren08> | Well I do thank you guys that do work on it |
| 19:46 | <hsivonen> | AryehGregor: but only you are contractually Hixie's minion :-) |
| 19:46 | <Hixie> | hah |
| 19:46 | <AryehGregor> | Yes, that's true. |
| 19:46 | <Hixie> | i dunno if i'd call what y'all do "menial" |
| 19:46 | <AryehGregor> | Other people are his minions for fun. |
| 19:46 | <TabAtkins_> | "fun" |
| 19:46 | <AryehGregor> | I'm heartless and only do minion work for money. |
| 19:46 | <Hixie> | "indispensible" maybe |
| 19:46 | <swarren08> | If you guys dont mind, id like to hang around in this channel to watch you guys |
| 19:47 | <AryehGregor> | swarren08, go ahead, that's how lots of people get started in standards work. |
| 19:47 | <Ms2ger> | You're welcome |
| 19:47 | <Hixie> | swarren08: you are welcome to hang around :-) |
| 19:47 | <AryehGregor> | Ask any questions you have. |
| 19:47 | <Ms2ger> | Within a year or two, you'll be a minion as well :) |
| 19:47 | <AryehGregor> | Like, it's how I got started, and TabAtkins_, and others too. |
| 19:47 | <swarren08> | Ive been building websites for years now |
| 19:47 | <Hixie> | swarren08: if you hang around long enough, you might find yourself being useful, that's how pretty much all of us started :-) |
| 19:47 | <wilhelm> | Just remember to leave your sense of logic at the door, please. |
| 19:47 | <swarren08> | I finally decided to take a look into HTML5 and CSS3 since its going to be standard soon |
| 19:47 | <swarren08> | Logic? whats that? :) |
| 19:48 | <Ms2ger> | People who write websites? These really exist? :) |
| 19:48 | <swarren08> | apperently im a dieing race now |
| 19:48 | <swarren08> | lol |
| 19:48 | <swarren08> | i was tought using notepad |
| 19:48 | <jgraham> | I'm not sure. As far as I cn tell the world is full of people who write web browsers |
| 19:48 | <jgraham> | I though the websites just formed out of the aether |
| 19:48 | <swarren08> | taught* |
| 19:48 | <Ms2ger> | That's my experience as well |
| 19:49 | <Ms2ger> | And fishy aether, usually |
| 19:49 | <swarren08> | How long have you all been doing this? as in working on the standard? |
| 19:50 | <jgraham> | Well the WHATWG started in 2004 or so |
| 19:50 | <swarren08> | yea after the conference |
| 19:51 | <jgraham> | Of course some people date from before that |
| 19:51 | <AryehGregor> | Some people here have been working on browsers since the Netscape days. |
| 19:51 | <Hixie> | i've been working on specs since late 1998, iirc |
| 19:51 | <Hixie> | if you could posting on mailing lists "working on specs" |
| 19:51 | <swarren08> | Wow |
| 19:52 | <Ms2ger> | But Hixie's just old |
| 19:52 | <Hixie> | it's true |
| 19:52 | <swarren08> | I feel like i may be the youngest one here |
| 19:52 | <jgraham> | swarren08: Well if your user name is from your year of birth |
| 19:52 | <hober> | swarren08: probably not, unless your nick means you were born in 2008. good work on typing & language acquisition, if so |
| 19:52 | <swarren08> | lol |
| 19:53 | <swarren08> | actually its my graduating year of high school |
| 19:53 | <swarren08> | ive had that sn since about the 6th grade |
| 19:53 | <AryehGregor> | We have people pretty close to your age here. |
| 19:53 | <AryehGregor> | I graduated high school in 2005. There are at least a couple people here younger than me. |
| 19:53 | <swarren08> | Wow |
| 19:54 | <jgraham> | yeah, gsnedders is about 3 or 4 now |
| 19:54 | <jgraham> | (OK, more like 19) |
| 19:54 | <swarren08> | and you work on the standards also AryehGregor? |
| 19:54 | <Ms2ger> | And it's still jgraham who has the infantile humour? :) |
| 19:54 | <AryehGregor> | swarren08, yes, currently contracting for Google to write this: http://aryeh.name/spec/editcommands/editcommands.html |
| 19:55 | <jgraham> | Ms2ger: Not to mention thee grey hair :) |
| 19:55 | <swarren08> | woa ehy that is you, i was looking at that yesterday |
| 19:55 | <AryehGregor> | I'm pretty sure the 90th percentile age among regulars in this room is well under 30. |
| 19:55 | <swarren08> | Thats awesome |
| 19:56 | <AryehGregor> | But really, nobody cares. You're judged on your contributions, not your age. |
| 19:56 | <Hixie> | (i'm 31, was 18 or 19 or so when i started posting to the lists) |
| 19:56 | <jgraham> | Hmm, I wonder if that is actually true |
| 19:56 | <swarren08> | I bet you all have or are in college |
| 19:56 | <Ms2ger> | ...Or your degrees |
| 19:56 | <swarren08> | -have + have been in |
| 19:57 | <AryehGregor> | Right, nobody cares much about degrees either. I mean, most of the people here have or are pursuing degrees in some technical field, in some cases Ph.D.'s, but it doesn't really matter. |
| 19:57 | <AryehGregor> | We just get to make fun of the Ph.D.'s by calling them "Dr." if we want to annoy them. |
| 19:57 | <Ms2ger> | Except for Hixie, who's happy with his BSc |
| 19:57 | <swarren08> | ah |
| 19:57 | <TabAtkins_> | Sure, but my college education has very little effect on my standards contribution. |
| 19:57 | <AryehGregor> | Who's the Opera person with a Ph.D. who doesn't like being called Dr.? jgraham? |
| 19:58 | <TabAtkins_> | Yeah, jgraham |
| 19:58 | jgraham | hides |
| 19:58 | <jgraham> | AryehGregor: A glance at the stats page suggests that more than 10% of the top 25 people might well be over 30 |
| 19:58 | <TabAtkins_> | On that note, who's got a math degree and wants to help me implement Coons patch mesh gradients? |
| 19:58 | <AryehGregor> | othermaciej also has a Ph.D., but reasonable people would actually believe that based on his behavior, unlike jgraham. |
| 19:58 | <TabAtkins_> | I got them working in squares, but bezier-curve sides are a bit more difficult... |
| 19:59 | <swarren08> | I know Apple, Opera and Mozilla are the main companies who were behind the Living Standard, is thier others? |
| 19:59 | <Ms2ger> | Maciej has a PhD? Really? |
| 19:59 | <Ms2ger> | |
| 19:59 | <swarren08> | Forgot Google |
| 19:59 | <AryehGregor> | TabAtkins_, I have a math degree, but it's pure math, so I get to contemptuously reply that such drudgework has nothing to do with the kind of math I studied (i.e., *real* math). |
| 19:59 | <TabAtkins_> | AryehGregor: Nuts to your real math. |
| 19:59 | <Ms2ger> | AryehGregor++ |
| 19:59 | <TabAtkins_> | I need a Math BS here. |
| 20:00 | <AryehGregor> | Oh, I'm wrong. |
| 20:00 | <AryehGregor> | Only a master's. |
| 20:00 | <AryehGregor> | His sister has a Ph.D. |
| 20:00 | <AryehGregor> | His master's is from MIT, though. |
| 20:00 | <swarren08> | dang |
| 20:00 | <AryehGregor> | TabAtkins_, I have a math BS. |
| 20:00 | <jamesr> | MEng? |
| 20:00 | <AryehGregor> | It's just a pure math Bs. |
| 20:00 | <AryehGregor> | BS. |
| 20:00 | <TabAtkins_> | Also: omigod it's so frustrating being locked out of email. I have so many threads to respond to, including one where I now have a working counterexample to someone's statement. |
| 20:00 | <AryehGregor> | jamesr, http://en.wikipedia.org/wiki/Maciej_Stachowiak |
| 20:00 | <swarren08> | you guys are smart people |
| 20:00 | <Hixie> | that's debatable |
| 20:00 | <jgraham> | BS is the worst name for a type of degree |
| 20:00 | <TabAtkins_> | BS |
| 20:01 | <Ms2ger> | Hixie is |
| 20:01 | <Hixie> | and indeed frequently debated! |
| 20:01 | <AryehGregor> | swarren08, nobody cares about the degrees. They only care that you're smart. Of course, smart people often get advanced degrees. |
| 20:01 | <Ms2ger> | AryehGregor, ... or they do something useful :) |
| 20:01 | <AryehGregor> | I said "often". |
| 20:01 | <jgraham> | And often people get advanced degrees without being unusually smart |
| 20:01 | <AryehGregor> | Also, the two are not mutually exclusive. |
| 20:01 | Ms2ger | went the degree route |
| 20:02 | <AryehGregor> | Well, yes, but smart people are more likely to get advanced degrees. Which is 90% of the reason anyone bothers, to signal that they're likely to be smarter than average. |
| 20:02 | <jgraham> | Hmm, I did it because it was fun |
| 20:02 | <AryehGregor> | Yeah, people do that too. |
| 20:02 | <swarren08> | Does anyone actually work for Google, Apple, Mozilla or Opera? that would be pretty awesome if someone does |
| 20:02 | <AryehGregor> | I'd have gone on to get a Ph.D. if it was fun, probably. |
| 20:02 | <AryehGregor> | swarren08, basically everyone you're talking to works for one of those. |
| 20:02 | <swarren08> | Really? |
| 20:02 | Ms2ger | doesn't |
| 20:03 | <swarren08> | Wow |
| 20:03 | <Hixie> | Ms2ger: how do i know which interface should be marked 'partial'? |
| 20:03 | <AryehGregor> | Hixie, TabAtkins_, and I are Google. Ms2ger volunteers for Mozilla, although apparently doesn't work for them. jgraham is Opera. |
| 20:03 | <TabAtkins_> | Ms2ger: You contribute to Moz heavily enough that it count. |
| 20:03 | <jgraham> | But one of the brightest, most productive people I work with doesn't have a degree so there are certainly exceptions |
| 20:03 | <swarren08> | Thats is amazing |
| 20:03 | <AryehGregor> | jamesr is Google, hober is Apple. |
| 20:03 | <Ms2ger> | Hixie, the one in the obsolete section? :) |
| 20:03 | <kbrosnan> | roc dbarron mozilla |
| 20:03 | <Ms2ger> | wilhelm is Opera |
| 20:03 | <TabAtkins_> | swarren08: Though Aryeh and I were working on this stuff before we were hired by Google. |
| 20:03 | <Hixie> | Ms2ger: oh, i got confused |
| 20:03 | <swarren08> | Ive always wanted to talk to people who worked for those companies |
| 20:03 | <TabAtkins_> | And were in fact hired by Google *because* of the standards work we o. |
| 20:03 | <jamesr> | anyone know what "kwac" or "Obigo X10" are? (i'm looking at the view-mode media feature conformance report from public-webapps) |
| 20:04 | <AryehGregor> | Etc. |
| 20:04 | <gsnedders> | jgraham: I don't think much stuff in general applies to jl, though. |
| 20:04 | <Hixie> | Ms2ger: i saw you put it on HTMLElement and forgot that we had a duplicate of that in the microdata section for w3c's nonsense |
| 20:04 | <jgraham> | Most people were working on his stuf before they were hired to work on it |
| 20:04 | <Ms2ger> | You didn't notice the comment about that in the context? :) |
| 20:04 | <jgraham> | gsnedders: If you decapitate him he will die? |
| 20:04 | <wilhelm> | For seven years now. Shit, I'm getting old. |
| 20:04 | <Ms2ger> | Which reminds me of krijn's quote "What else do I need to do to get hired by Google" |
| 20:05 | <swarren08> | Here is an observation, ive only notice you guys speak of W3C 1 time in the past 20 minutes(maybe more) |
| 20:05 | <jgraham> | wilhelm: And yet you are still younger than me. Bastard |
| 20:05 | <Ms2ger> | We like not to think about the W3C |
| 20:05 | <swarren08> | What is the difference between what you guys do and what W3C does? |
| 20:05 | <Ms2ger> | And if we do, we rant |
| 20:05 | <gsnedders> | wilhelm: You're not the youngest person at Opera anymore! Though there again, we have younger interns than me this summer… |
| 20:05 | <swarren08> | im sorry if its a sore spot |
| 20:05 | <AryehGregor> | swarren08, roughly, we do the work and the W3C gets in the way. |
| 20:05 | <wilhelm> | jgraham: But I have the same level of education as Mr. swarren08. Can I borrow some of your diplomas? |
| 20:05 | <Ms2ger> | We write specs, the W3C publishes specs? :) |
| 20:05 | <Hixie> | we actually make progress, w3c releases press-releases claiming our progress as theirs |
| 20:05 | <Hixie> | wait, was that impolitick |
| 20:05 | <TabAtkins_> | swarren08: We're both standards bodies. The WHATWG was formed originally because the W3C didn't want to work on HTML. |
| 20:05 | <Hixie> | crap |
| 20:05 | <smaug____> | it really doesn't matter too much where the standardization happens, if it happens |
| 20:06 | <swarren08> | yea XHTML is what W3C was interested in |
| 20:06 | <AryehGregor> | No, but it matters if the W3C distracts us with bureaucratic nonsense that interferes with our actual work. |
| 20:06 | <Ms2ger> | Hixie, I should quote that on your wikipedia page :) |
| 20:06 | <jgraham> | wilhelm: I'm not sure they will be of much help to you. There isn't a big market for astrophysicists |
| 20:06 | <wilhelm> | Oh. |
| 20:06 | <AryehGregor> | And knowingly publishes outdated and incorrect standards that it refuses to update for procedural reasons. |
| 20:06 | <AryehGregor> | Etc. etc. etc. |
| 20:06 | <TabAtkins_> | swarren08: Later, the W3C requested that HTML be developed in it again, so now it's developed jointly between the two. |
| 20:06 | <Hixie> | hey there's no k on the end of impolitic |
| 20:06 | <Hixie> | who knew |
| 20:06 | <TabAtkins_> | swarren08: With the caveats that other people have mentioned. |
| 20:07 | <AryehGregor> | Hixie, there is if you went to school in the 19th century. How old did you say you are again? |
| 20:07 | <swarren08> | So you guys do the work and W3C takes credit? |
| 20:07 | <swarren08> | am i getting that right? |
| 20:07 | <Ms2ger> | You'll note that AryehGregor is rather anti-W3C |
| 20:07 | <Hixie> | i was in school in the 20th century! |
| 20:07 | <AryehGregor> | swarren08, we're being sarcastic and bitter. |
| 20:07 | <Hixie> | swarren08: the truth is less cynical than that |
| 20:07 | <swarren08> | Ah... i feel so dumb |
| 20:07 | <Hixie> | also more complicated |
| 20:07 | <wilhelm> | swarren08: Many here wear hats from several organizations. |
| 20:07 | <AryehGregor> | swarren08, the W3C is an organization, it doesn't do anything but provide some web hosting and things like that. It doesn't employ the people who write or edit almost any of the actual standards. |
| 20:07 | <wilhelm> | Best tool for the job, &c. |
| 20:08 | <Hixie> | the w3c is an institution that provides support for standards development |
| 20:08 | <swarren08> | i had no idea the standards would have so much "drama" behind it |
| 20:08 | <Hixie> | the whatwg is essentially a mailing list |
| 20:08 | <AryehGregor> | Plus a wiki. |
| 20:08 | <jgraham> | swarren08: It's a bit like a soap opera |
| 20:08 | <Ms2ger> | Who described "working group" as "the intersection of web technology and religion"? |
| 20:08 | <AryehGregor> | The W3C imposes lots of requirements on how exactly standards are to be developed, as a condition of hosting them. |
| 20:08 | <swarren08> | ive begun to notice that jgraham |
| 20:08 | <Hixie> | we need the w3c because the w3c has managed to convince a large number of companies to grant licenses for their patents |
| 20:08 | <cygri> | Ms2ger: i thought that was RDF |
| 20:08 | <Ms2ger> | cygri++ |
| 20:08 | <jgraham> | At christmas we will have a special episode where it will be revealed that Hixie is actually TimBL's lovechild |
| 20:08 | <AryehGregor> | In exchange they provide their brand name, patent licensing, etc. Plus Microsoft only participates in the W3C, not the WHATWG. |
| 20:09 | <Ms2ger> | cygri, but RDF isn't webtech |
| 20:09 | <swarren08> | Ive noticed microsoft doesnt do much with HTML5 |
| 20:09 | <jgraham> | Yes, the patent licesing is a big deal |
| 20:09 | <swarren08> | or at least didnt till IE9 |
| 20:09 | <AryehGregor> | So we're forced to participate there for political reasons, which means we have to obey their rules, which we all think are stupid and a waste of everyone's time. |
| 20:09 | <Ms2ger> | They're a couple of years behind |
| 20:09 | <AryehGregor> | swarren08, Microsoft is implementing HTML5 like everyone else, and they're catching up at it. They provide useful feedback on the things they implement, like other browsers. |
| 20:09 | <AryehGregor> | They just came somewhat late to the game. |
| 20:10 | <AryehGregor> | IE10 likely won't be far behind other browsers' HTML5 support. |
| 20:10 | <Hixie> | Ms2ger: what was the stuff in web workers you weren't sure about? |
| 20:10 | <Hixie> | it's not so much that the w3c's rules are stupid, so much as they are anachronistic |
| 20:10 | <Hixie> | they haven't evolved with the times |
| 20:10 | <Ms2ger> | DedicatedWorkerGlobalScope, SharedWorkerGlobalScope, AbstractWorker |
| 20:10 | <swarren08> | To be honest, i would have figured Microsoft would be more interested in this stuff since they have 40+% of the world using IE |
| 20:11 | <AryehGregor> | They still annoy us because they tend to make opaque institutional decisions based on internal discussions instead of having discussions individually on public lists, and they also refuse to acknowledge the WHATWG, which forces us to deal with it more than we'd like. |
| 20:11 | <AryehGregor> | But MS is definitely a part of the picture here. |
| 20:11 | <Hixie> | which is highly ironic because the w3c's technologies are largely responsible for moving humanity's culture in the way that its processes now need to be updated to take into account |
| 20:11 | <Hixie> | Ms2ger: k |
| 20:11 | <AryehGregor> | Yeah, it's kind of ridiculous when the W3C doesn't update to account for changes brought about by the success of the web. |
| 20:11 | <Ms2ger> | We don't really hate the W3C as much as we usually make it look |
| 20:12 | <Hixie> | Ms2ger: did you leave those as [Supplemental] |
| 20:12 | <Ms2ger> | I didn't touch them |
| 20:12 | <Hixie> | k |
| 20:12 | <Hixie> | cool |
| 20:12 | <Hixie> | thanks |
| 20:12 | <swarren08> | I definitly see where your coming from AryehGregor, ive wondered the same things at time. Wont it hurt them down the line |
| 20:12 | <Hixie> | woot, your patch applied cleanly! |
| 20:12 | <AryehGregor> | Like, what point is there in a spec maturity cycle that requires the spec to freeze for years before it's called complete when major browsers are all on release cycles from six weeks to a year? |
| 20:12 | <Hixie> | Ms2ger++ |
| 20:13 | Hixie | waits for Ms2ger to rename himself Ms3ger |
| 20:13 | <Ms2ger> | Old joke, sorry :) |
| 20:13 | <Hixie> | :-P |
| 20:13 | <AryehGregor> | Hixie, surely "Ms2ger" evaluates to 0 and becomes 1? |
| 20:13 | <Hixie> | maybe i just get a type mismatch |
| 20:13 | <Ms2ger> | AryehGregor, :( |
| 20:13 | <Hixie> | or maybe Ms2ger overloaded the ++ operator |
| 20:13 | <AryehGregor> | No, NaN. |
| 20:14 | <AryehGregor> | In PHP it would be 1. |
| 20:14 | <Ms2ger> | operator++() { return this; } |
| 20:14 | <wilhelm> | swarren08: It already has. |
| 20:15 | <AryehGregor> | Yeah, the formation and success of the WHATWG put a serious dent in the W3C's preeminence as a web standards body . . . |
| 20:15 | <Hixie> | swarren08: eventually if the w3c doesn't adapt there's going to come a point where someone gets annoyed enough that they set up an alternative patent policy |
| 20:15 | <AryehGregor> | Anyway, I should get back to working. |
| 20:15 | <Hixie> | swarren08: at which point they're basically doomed |
| 20:15 | Hixie | has been trying really hard to convince them to change, without any suggestion of success |
| 20:15 | <AryehGregor> | Hixie, you need to have implementers credibly threaten to abandon them. |
| 20:15 | <AryehGregor> | That's the only thing that will force them to change. |
| 20:16 | <Hixie> | that will happen once there's a patent policy, i expect |
| 20:16 | <AryehGregor> | Anything else will be voted down by the AC on reactionary grounds. |
| 20:16 | <swarren08> | Then why hasent someone done that? You all basically work for big companies and see where the web is going |
| 20:16 | <Ms2ger> | Or, the <dfn title=dom-Ms2ger-operator++>++</dfn> operator must return the <span>context object</span> |
| 20:16 | <swarren08> | You all should have more say for whats going on |
| 20:16 | <Hixie> | swarren08: it's a huge amount of really painful work |
| 20:16 | <AryehGregor> | In practice we do have all the say. |
| 20:16 | <AryehGregor> | The W3C annoys us, but doesn't interfere too much in the end. |
| 20:16 | <AryehGregor> | Just wastes our time. |
| 20:17 | <Hixie> | swarren08: so for someone to do it they have to be _really_ pushed over the edge |
| 20:17 | <jgraham> | swarren08: If we did that there wouldn't be anything to complain about |
| 20:17 | <Ms2ger> | IME, it's mainly the HTMLWG that wastes our time |
| 20:17 | <Hixie> | swarren08: i've gotten close, but not enough |
| 20:17 | <AryehGregor> | Yes, other WGs are much saner than the HTMLWG. |
| 20:17 | <Hixie> | Ms2ger: i'm getting my time wasted by plenty of other WGs too now |
| 20:17 | <swarren08> | so what your telling me |
| 20:17 | <Ms2ger> | WebApps wastes some, but not enough for me to whine |
| 20:17 | <jgraham> | and the peanut gallery couldn't use the yellow marker of horror to highlight pople saying cynical things |
| 20:17 | <Hixie> | webapps, webrtc, websockets (though that one was IETF, not W3C, i guess) |
| 20:17 | <swarren08> | is they bother you but yet not enough to get rid of? |
| 20:17 | <Ms2ger> | Yeah |
| 20:18 | <Hixie> | swarren08: pretty much |
| 20:18 | <swarren08> | Ah |
| 20:18 | <AryehGregor> | Hixie, what you need to do is lure someone into doing it who doesn't realize how much work it is. You should start talking about how great it would be if someone just put in a little effort, while of course noting that you yourself already have far too many obligations to fit it into your schedule. |
| 20:18 | <swarren08> | I think I am finially understanding |
| 20:18 | <Hixie> | AryehGregor: i've been doing just that :-) |
| 20:18 | <hober> | AryehGregor: that sounds familiar somehow :) |
| 20:18 | <AryehGregor> | The W3C made enough concessions that it's not worth it for us to try too hard to pull out. |
| 20:18 | <Hixie> | AryehGregor: nobody has bitten enough to actually do it yet |
| 20:18 | <karlcow> | free specs? :) https://github.com/kobolabs/Kobo-Reader/blob/master/fickel/main.cpp#L320 |
| 20:19 | <jgraham> | The idea of spending lots of time locked in a room wih lawyers is much less appealing than the idea of spending that same time thinking of a new cool feature for the web, or implementing part of the spec |
| 20:19 | <jgraham> | So guess which we do |
| 20:19 | <AryehGregor> | jgraham, that depends. I like law, it's fun! |
| 20:19 | <Hixie> | swarren08: the other thing that means we're not hugely motivated to set up a patent policy is that we know we can always do it in the future if the w3c does go off the rails, so there's no urgency |
| 20:19 | <swarren08> | But then it comes back to the point where you guys to the work and W3C publishes the specs |
| 20:19 | <AryehGregor> | Probably related to my being Jewish. |
| 20:19 | <Hixie> | swarren08: most of the important specs are developed in a way that means we could fork them if we had to |
| 20:19 | <swarren08> | You all are playing the waiting game? |
| 20:20 | <jgraham> | No, we're doing fun stuff to improve the web |
| 20:20 | <AryehGregor> | swarren08, we could take the specs away from the W3C if we wanted to. We edit them and we or our employers hold the copyright. All we'd have to do is refuse to license new changes under the W3C's license. |
| 20:20 | <wilhelm> | More like the just-get-the-work-done-game. |
| 20:20 | <Hixie> | what wilhelm said |
| 20:20 | <swarren08> | oh |
| 20:20 | <AryehGregor> | But all this secession stuff would cause even more hassle. |
| 20:20 | <swarren08> | I see now |
| 20:20 | <Hixie> | heycam|away: ok so there's one thing i still need [Supplemental] for -- please ping me when you're around |
| 20:20 | <swarren08> | its just easier to let them play as they say |
| 20:21 | <Hixie> | it would cause a giant shitstorm if we were to actually fork |
| 20:21 | <AryehGregor> | Plus it would be perceived by the general public as "evil browser implementers try to take control of specs away from beloved W3C", since the W3C has a very positive reputation in most circles. |
| 20:21 | <Hixie> | dunno about "most" anymore |
| 20:21 | <Hixie> | maybe "many" |
| 20:21 | <jarek_> | is there an API that would allow me to detect how hard the user has pressed the stylus? |
| 20:21 | <swarren08> | I can only imagine Hixie |
| 20:21 | <AryehGregor> | There's some idea that it's open and inclusive and whatever, and not controlled by implementers, which is a joke, but there you have it. |
| 20:22 | <jarek_> | e.g. on tablet on touch sensitive device |
| 20:22 | <jarek_> | s/on/or |
| 20:22 | <Hixie> | i feel bad, i'm marking Ms2ger's huge patch as "editorial" for the tracker |
| 20:22 | <Ms2ger> | Hah |
| 20:22 | <AryehGregor> | What patch is this? |
| 20:22 | <Ms2ger> | That's fine |
| 20:22 | <Ms2ger> | Supplemental -> partial |
| 20:22 | <Hixie> | http://www.w3.org/Bugs/Public/show_bug.cgi?id=13069 |
| 20:23 | <AryehGregor> | Ah. |
| 20:23 | <Hixie> | Ms2ger: thanks a ton for this btw |
| 20:23 | <smaug____> | jarek_: touch.force |
| 20:23 | <swarren08> | Wow |
| 20:23 | <smaug____> | and some vendor specific things |
| 20:23 | <smaug____> | like mozPressure |
| 20:23 | <Ms2ger> | Happy to do something useful for once :) |
| 20:24 | <Hixie> | :-D |
| 20:24 | <jarek_> | smaug____: thanks, http://www.w3.org/TR/touch-events/ is exactly what I was thinking about |
| 20:24 | <Ms2ger> | Now I believe you owe me a fix for those DOM Core bugs? ;;) |
| 20:24 | <Hixie> | Ms2ger: aw man, i knew that was coming |
| 20:25 | <Ms2ger> | jarek_, http://dvcs.w3.org/hg/webevents/raw-file/default/touchevents.html, you mean |
| 20:25 | <Hixie> | Ms2ger: ok what do we need to do for this |
| 20:25 | <Ms2ger> | Hixie, I should form a WG to prioritize your work ;) |
| 20:25 | <jarek_> | I was recently playing with http://mrdoob.com/projects/harmony/#sketchy, this could be a killer app if it could support pressure sensitivity |
| 20:26 | <Hixie> | Ms2ger: is there some way we can diff the relevant parts of the specs to make sure we're not regressing anything? |
| 20:26 | <Hixie> | Ms2ger: or maybe let's just do it one small bit at a time? |
| 20:26 | <Ms2ger> | That's fine too |
| 20:26 | <Hixie> | http://dvcs.w3.org/hg/domcore/raw-file/tip/Overview.html is the latest? |
| 20:27 | <Ms2ger> | Yes |
| 20:27 | <swarren08> | I noticed someone didnt like webapps, why is that? |
| 20:27 | <Hixie> | jgraham: i'm getting 504 gateway timeouts on pms |
| 20:27 | <Hixie> | jgraham: not even doing the annotations |
| 20:27 | <Hixie> | swarren08: the chair is pushing me to work on things for bureaucractic reasons rather than because they make the web better |
| 20:27 | <Ms2ger> | swarren08, webapps mean the Web Applications Working Group in this context, btw |
| 20:28 | <Hixie> | Ms2ger: just stuff in Terminology and Exceptions? |
| 20:28 | <Ms2ger> | Sounds good |
| 20:30 | <jgraham> | Hixie: OK I will look |
| 20:30 | <Hixie> | jgraham: it fixed itself |
| 20:30 | <Hixie> | jgraham: dunno what that was about |
| 20:31 | <jgraham> | Oh. Good |
| 20:31 | <jgraham> | I imagine it is just my hosting being rubbish or something |
| 20:32 | <Hixie> | Ms2ger: hmm, there's less here than i realised |
| 20:32 | <Hixie> | Ms2ger: what exactly are you looking for? |
| 20:32 | <Hixie> | Ms2ger: just the string stuff and the microsyntaxes stuff, and domexception? |
| 20:33 | <Hixie> | Ms2ger: or even things like "preferred mime type"? |
| 20:33 | <Ms2ger> | Dunno about mime type |
| 20:34 | <Ms2ger> | The rest was what I had in mind, I think |
| 20:34 | <Hixie> | s/codepoint/code point/ |
| 20:34 | <Ms2ger> | It's been a while :) |
| 20:34 | <Hixie> | ("codepoint" isn't english) |
| 20:34 | <jgraham> | It is if you choose it o be |
| 20:34 | <ParadoX-> | Hey everyone |
| 20:34 | <jgraham> | *to |
| 20:34 | hsivonen | didn't know "HTML Standards" was an "engineering discipline" |
| 20:35 | <hsivonen> | (from Maciej's Wikipedia page) |
| 20:35 | <Hixie> | Ms2ger: also the dom core spec is missing "compatibility caseless" |
| 20:37 | <Hixie> | Ms2ger: the microsyntax section also isn't an exact match |
| 20:37 | <jgraham> | Is there some law that all WHATWG people with a wikipedia page have to have a dreadful photo? |
| 20:37 | <Hixie> | Ms2ger: which makes sense the specs have different requirements from these algorithms |
| 20:37 | <Ms2ger> | We copied just what we needed, I think |
| 20:37 | <Hixie> | Ms2ger: maybe it would make more sense for us to share some material and merge it in at publication time |
| 20:37 | <Hixie> | Ms2ger: rather than reference each other |
| 20:38 | <Hixie> | Ms2ger: i think it would be lame, e.g., for HTML to reference DOM Core for "skip whitespace" but not for "skip White_Space" |
| 20:38 | <Ms2ger> | I can pull that in as well |
| 20:38 | <Hixie> | (just like many of the split-out versions of WA1 have duplicate terminology) |
| 20:38 | <Hixie> | Ms2ger: yeah but then it's lame that DOM Core has stuff it doesn't use |
| 20:38 | <AryehGregor> | jgraham, apparently. Part of the reason I haven't posted photos of myself anywhere is because I'm horrifyingly unphotogenic. |
| 20:39 | <Ms2ger> | The potential for divergence is also lame :) |
| 20:39 | <Hixie> | Ms2ger: yeah, hence, let's have some sort of common source we can merge together |
| 20:40 | <Hixie> | Ms2ger: the common source would be a superset of what we both need, and we'd just take the bits we need automatically when generating the spec |
| 20:40 | <jgraham> | AryehGregor: But this isn't even about being unphotogenic. It's just about the photos being bad |
| 20:40 | <Hixie> | Ms2ger: if you like we can make the dom core spec be that document, if it has appropriate markers i can just slurp it in |
| 20:40 | <Ms2ger> | Hmm |
| 20:40 | <AryehGregor> | Wikipedia photos are usually bad. They prefer bad photos to no photos, because bad photos encourage the subject to contribute better photos under a free license. |
| 20:41 | <jgraham> | Obviouly th W3C should blow its budget on hiring a pro photographer for TPAC |
| 20:41 | <Hixie> | AryehGregor: i thought it was against the rules to submit your own photo |
| 20:41 | <jgraham> | Who will license the photos under a free license |
| 20:41 | <AryehGregor> | It often comes up for celebrities, where they post some trash that someone took at a public event with their cheap phone and put on Flickr. |
| 20:41 | <AryehGregor> | Then all someone has to do is inform the celebrity's press department and they'll get a freely-licensed photo right away! |
| 20:42 | <Ms2ger> | So I add the definitions we don't have yet and add some magic markers? |
| 20:42 | <Hixie> | Ms2ger: that would work |
| 20:42 | <Hixie> | in the meantime i'm fixing the exceptions stuff |
| 20:42 | <AryehGregor> | Hixie, not at all. You're perfectly entitled to edit your own articles, even, as long as you're careful about it and are making only clear improvements. See <http://en.wikipedia.org/wiki/WP:COI>;. |
| 20:43 | <Hixie> | Ms2ger: you don't even need to add the stuff you don't have |
| 20:43 | <Hixie> | Ms2ger: just put markers around what you do have and i'll do the rest |
| 20:43 | <Hixie> | Ms2ger: i have an infrastructure to import stuff already |
| 20:43 | <AryehGregor> | If you have a better photo that's freely licensed, upload it and replace the existing one yourself if you like. |
| 20:43 | <jcranmer> | good god that's a bad photo |
| 20:44 | <Hixie> | i have a horrible feeling this project is going to involve me learning Yet Another source control versioning system |
| 20:44 | <Ms2ger> | But you'd want to put, say, compatibility caseless matches right in the middle, no? |
| 20:44 | <Hixie> | AryehGregor: that feels... self-indulgent in a way i'm not comfortable with |
| 20:44 | <AryehGregor> | They even have a specific guideline for it: http://en.wikipedia.org/wiki/Wikipedia:COI#Photographs_and_media_files |
| 20:44 | <Hixie> | Ms2ger: yeah but that's easy to do |
| 20:44 | <Ms2ger> | And if you don't know hg yet, yes |
| 20:44 | <AryehGregor> | No it's not, you're only improving Wikipedia. |
| 20:44 | <Ms2ger> | OK |
| 20:44 | <Hixie> | Ms2ger: just put plenty of markers in |
| 20:44 | <AryehGregor> | Uploading high-quality photos of Wikipedia subject matter is encouraged. |
| 20:44 | <Hixie> | Ms2ger: and i can just do regexps to slurp the right bits |
| 20:45 | <Hixie> | well on the plus side, hg is on this machine already |
| 20:45 | <Hixie> | that's lucky |
| 20:45 | <Ms2ger> | I hope you have all of this code backed up :) |
| 20:45 | <Hixie> | code? |
| 20:45 | jgraham | wonders what the average number of VCSs developers know these days is |
| 20:45 | <Ms2ger> | Your preprocessor |
| 20:45 | <Hixie> | yes |
| 20:46 | <Hixie> | but why would i need that? :-) |
| 20:46 | <Hixie> | my plan is to just have a script that splits apart your spec into separate files for eahc bit i need |
| 20:46 | <Hixie> | and then use the code i already have for importing files into the spec |
| 20:46 | <Hixie> | the same code i use to take examples in whatwg.org/demos and merge them into the workers section, w.g. |
| 20:46 | <Hixie> | e.g. |
| 20:46 | <Ms2ger> | I'm just afraid of the time lost if anything ever happens to your computer :) |
| 20:47 | <jcranmer> | most of my scripts aren't in VCS |
| 20:47 | <jcranmer> | I have an entire directory of that stuff |
| 20:47 | <jcranmer> | /src/mozilla-tools (which is a bit of a lie) |
| 20:47 | <jcranmer> | it's more of a "this isn't a separate repository, but it's a collection of little things which aren't worth crap by themselves" |
| 20:48 | <Ms2ger> | So, which sections, and what marker? :) |
| 20:49 | <hsivonen> | it seems that http://en.wikipedia.org/wiki/Wikipedia:Article_Feedback_Tool has existed for months, and yet I only learned about it just now |
| 20:50 | <karlcow> | hsivonen: is it bad to lag? |
| 20:51 | <hsivonen> | hsivonen: I don't know if it is *bad*. I just use wikipedia so often that I'm surprised I haven't seen that before. |
| 20:52 | <AryehGregor> | hsivonen, it's being gradually rolled out. |
| 20:52 | <AryehGregor> | First it was only a small subset of articles. |
| 20:52 | <AryehGregor> | IIRC, current plans are to roll it out to all articles soon, if it hasn't happened already. |
| 20:57 | <annevk> | so make stopped working on Lion |
| 20:57 | <annevk> | what the fuck |
| 20:59 | <Ms2ger> | Lion-- |
| 21:01 | <jgraham> | annevk: You need to download the new Xcode too |
| 21:01 | <karlcow> | mwahah |
| 21:01 | <annevk> | ffs |
| 21:01 | <jgraham> | Always the best UX with Apple |
| 21:02 | <karlcow> | bye Google Labs http://googleblog.blogspot.com/2011/07/more-wood-behind-fewer-arrows.html |
| 21:02 | <annevk> | thanks jgraham |
| 21:03 | <annevk> | kind of interseting that macports is still around |
| 21:04 | <annevk> | but I cannot run upgrade |
| 21:04 | <annevk> | oh this is going to take ages too |
| 21:04 | <karlcow> | :) I guess annevk was too impatient to go to his (Lion) prom ball. |
| 21:05 | <annevk> | well it would have taken ages at some point |
| 21:06 | <annevk> | where ages is not really that long |
| 21:06 | <annevk> | downloading Lion takes about 30min |
| 21:06 | <annevk> | same for Xcode |
| 21:06 | <annevk> | installing either takes about half an hour too I guess, but that's still quite long |
| 21:07 | <karlcow> | I guess it depends on which type of lines you are. |
| 21:07 | <karlcow> | I wonder how long it takes in some less broadband fortunate places in the world. |
| 21:08 | <annevk> | actual ages |
| 21:08 | <karlcow> | It seems that they will finally release a USB key in August |
| 21:08 | <annevk> | this is >100mbit theoretically |
| 21:08 | <karlcow> | yup I do not see my parents downloading it through their house communication in the countryside |
| 21:09 | <Hixie> | Ms2ger: my computer here is a machine at dreamhost, which i backup via rsync nightly to a machine that is backed up to both time machine to a redundant drobo and via the cloud to an off-site backup system |
| 21:09 | <Hixie> | Ms2ger: so i think we're good |
| 21:09 | <Hixie> | lunch, bbl |
| 21:09 | <Ms2ger> | Good |
| 21:18 | <annevk> | I am not really convinced that CSS splitting is that great. Simple global things like parsing and its object model are very much under appreciated with the current design. And I am pretty sure there is a lot more. |
| 21:25 | <TabAtkins_> | That's because parsing is BOOOORING. |
| 21:25 | <TabAtkins_> | (Not really, please god let's fix it.) |
| 21:25 | <Ms2ger> | Yeah, let TabAtkins fix it :) |
| 21:26 | <TabAtkins_> | If I had another me, I'd do it right now. |
| 21:29 | <annevk> | parsing serializing and the object model are all overlooked in favor of "hey, new features!" |
| 21:29 | <annevk> | and "new features" often get prioritized in weird ways |
| 21:29 | <annevk> | quite often not in line with what developers and browsers are asking for I have the feeling |
| 21:30 | <annevk> | anyway |
| 21:30 | <annevk> | Xcode updated |
| 21:30 | <annevk> | make still fails |
| 21:30 | <annevk> | need to reboot terminal? |
| 21:30 | <annevk> | does not appear to work |
| 21:31 | <AryehGregor> | Argh, I really wish someone would fix CSS serialization. |
| 21:31 | <AryehGregor> | It's a nightmare to deal with. |
| 21:31 | <AryehGregor> | No interop at all. |
| 21:31 | <annevk> | wtf |
| 21:31 | <annevk> | Error: Target org.macports.configure returned: configure failure: shell command failed (see log for details) |
| 21:31 | <annevk> | Log for cvs is at: /opt/local/var/macports/logs/_opt_local_var_macports_sources_rsync.macports.org_release_ports_devel_cvs/main.log |
| 21:31 | <annevk> | Error: Unable to upgrade port: 1 |
| 21:31 | <annevk> | I hate everything about this |
| 21:32 | <annevk> | Maybe I need to reboot Lion |
| 21:32 | <annevk> | hmm |
| 21:32 | <jamesr> | you're trying to develop on lion? ahahahahahahahahahahaha |
| 21:38 | <annevk> | yeah fuck this |
| 21:39 | <annevk> | time to do nothing |
| 22:10 | <heycam> | Hixie, here now |
| 22:16 | <Hixie> | heycam: dedicated workers and shared workers both have a global scope that is the same except it has different members |
| 22:16 | <Hixie> | heycam: the way i do it now is that i have a common ancestor |
| 22:17 | <Hixie> | heycam: and the descendant interfaces are [Supplemental, NoInterfaceObject] |
| 22:17 | <Hixie> | heycam: and then in the prose i just refer to the descendant interfaces |
| 22:17 | <heycam> | Hixie, ok so you want distinct dedicated and shared worker prototype objects, which have some common members, is that right? |
| 22:17 | <Hixie> | yeah |
| 22:17 | <Hixie> | but they have the same name |
| 22:18 | <heycam> | Hixie, the same name? |
| 22:18 | <Hixie> | WorkerGlobalScope |
| 22:21 | <heycam> | Hixie, ok, could you mail public-script-coord about it? I will try my hardest to avoid adding [Supplemental] but you never know ;) |
| 22:21 | <Hixie> | i've no problem solving it another way, i'd just like to avoid awkward prose |
| 22:21 | <Hixie> | will e-mail the list |
| 22:21 | <heycam> | sure |
| 22:21 | <heycam> | thanks |
| 22:22 | heycam | will get back to looking at Web IDL in a couple of weeks |
| 22:57 | <Hixie> | heycam: what happens if an object marked NoInterfaceObject inherits from another and is instantiated? |
| 22:57 | <jwalden> | huh, ie9 doesn't support wav in its audio element? |
| 22:57 | <heycam> | Hixie, it works just as if it didn't have NoInterfaceObject -- the only thing NoInterfaceObject affects is whether the global.InterfaceName property exists |
| 22:59 | <Hixie> | heycam: k |
| 23:00 | <Hixie> | heycam: mail sent |
| 23:00 | <heycam> | thanks |
| 23:11 | jwalden | marks http://krijnhoetmer.nl/irc-logs/whatwg/20110720#l-1335 appropriately |
| 23:35 | <annevk> | did Google Maps get a new UI? |
| 23:35 | <TabAtkins> | Yes. |
| 23:35 | <annevk> | let me the be the first to point it is inconsistent with Calendar and Google+ |
| 23:35 | <TabAtkins> | Vaguely related: I wish we used a distance-preserving projection on Maps. |
| 23:36 | <annevk> | David Bloom refers to this as first world problems |
| 23:36 | <TabAtkins> | Canada looks too damn big. |
| 23:36 | <annevk> | I say ugly |
| 23:37 | <jcranmer> | lol |
| 23:37 | <jcranmer> | zoom out to the second notch |
| 23:37 | <jcranmer> | hover somewhere on the left |
| 23:37 | <jcranmer> | and then zoom in |
