| 00:33 | <Hixie> | TabAtkins: cool |
| 00:33 | <Hixie> | TabAtkins: (where? i thought E was still the canonical source for that stuff... 2.1 isn't "living" anymore is it?) |
| 00:34 | <TabAtkins> | Not really. We *sometimes* update 2.1, but not always. |
| 00:34 | <TabAtkins> | Also: the update for this goes in the Positioning spec, which is semi-dead at the moment. (Rossen from IE is picking it back up now.) |
| 00:43 | <MikeSmith> | Hixie: about the inject-ARIA-by-script-instead thing, yeah sure I realize that doesn't make the documents an less invalid if the result has conformance errors. I just wish we had a different culture around validation, where people did have "I must eliminate all errors and get a 'You passed!'" as a goal of using the validator |
| 00:44 | <MikeSmith> | http://validator.w3.org/nu/ no longer emits a message saying whether a document is valid or not |
| 00:44 | <MikeSmith> | it just emits no errors |
| 00:45 | <MikeSmith> | and then says "Document checking completed." |
| 00:45 | <MikeSmith> | and if it's not valid, it emits the errors and then at the end still just says, "Document checking completed." |
| 00:46 | <MikeSmith> | (valid or invalidate as far as the subset of requirements it's actually checking) |
| 00:47 | <a-ja> | MikeSmith: <aside>fwiw, noticed that w3c validator doesn't give link for validating css unless there's no errors in html</aside> |
| 00:47 | <MikeSmith> | does anybody know if I can have the microformats wiki e-mail me a notification any time a page on my watchlist is changes? |
| 00:47 | <MikeSmith> | a-ja: I dunno what you mean by "give link for validating css" |
| 00:48 | <MikeSmith> | I don't work on the main w3c validator anyway |
| 00:48 | <MikeSmith> | nobody does any longer |
| 00:48 | <MikeSmith> | it's essentially unmaintained |
| 00:48 | <MikeSmith> | I only work on the validator.nu backend and http://validator.w3.org/nu/ |
| 00:49 | <a-ja> | MikeSmith: on no errors, results page gives you a link to validate your css, too |
| 00:49 | <MikeSmith> | I see |
| 00:49 | <MikeSmith> | a-ja: anyway you really shouldn't be using http://validator.w3.org/ |
| 00:49 | <MikeSmith> | I wish you wouldn't |
| 00:50 | <MikeSmith> | please just http://validator.w3.org/nu/ instead |
| 00:50 | <MikeSmith> | there's no good reason to use http://validator.w3.org/, ever |
| 00:51 | <a-ja> | only use at all cuz it's easy from "referrer" links |
| 00:57 | <a-ja> | MikeSmith: btw, as of sometime today, w3c's no longer gives errors on details role=group and summary role=button |
| 00:57 | <a-ja> | just the warning about details not being ready for primetime yet |
| 00:58 | <a-ja> | so _someone_ must be doing _some_ maintenance |
| 00:59 | <MikeSmith> | a-ja: I made those changes |
| 00:59 | <a-ja> | ah :) |
| 01:00 | <MikeSmith> | for HTML5 validation http://validator.w3.org/ uses the same backend from http://validator.w3.org/nu/ |
| 01:00 | <MikeSmith> | poorly |
| 01:04 | <Hixie> | MikeSmith: yeah... i'm just saying that if the goal is no validation errors, then moving stuff to script does nothing useful, and if the goal is no validator messages, then there's a much simpler way of doing it. |
| 01:05 | <MikeSmith> | Hixie: sure |
| 01:07 | <Hixie> | we really should find some way to do live validatior checking in these debugging tools |
| 01:07 | <a-ja> | i'd note that some org's have "no errors from a validator" requirement for WCAG compliance |
| 01:08 | <a-ja> | mostly govt |
| 01:08 | <MikeSmith> | a-ja: those orgs are broken |
| 01:08 | <MikeSmith> | a-ja: see http://validator.w3.org/nu/about.html |
| 01:09 | <Hixie> | "e broken |
| 01:09 | <Hixie> | uh |
| 01:09 | <MikeSmith> | ーThe Nu Markup Checker should not be used as a means to attempt to unilaterally enforce pass/fail conformance of documents to any particular specifications; it is intended solely as a checker, not as a pass/fail certification mechanism. |
| 01:10 | <Hixie> | a-ja: what i was trying to say is, if "no errors from a validator" means "document is conformant", then ok. if it means "this random software doesn't detect our many errors", then it's idiotic. |
| 01:10 | <Hixie> | so i tend to assume it means "has no errors" |
| 01:10 | <Hixie> | if people are interpreting it the other way, they are misinterpreting it |
| 01:11 | <SamB> | a-ja: what's an error? |
| 01:16 | <a-ja> | SamB: good question |
| 01:17 | <a-ja> | i just seem to recall validated conformance being a requirement for AAA |
| 01:19 | <a-ja> | i know of no orgs, even govt, that require AAA, though many require AA or better cuz of govt contracts |
| 01:24 | <SamB> | I guess most orgs don't drive? |
| 01:38 | <a-ja> | marginally related question -- where is style scoped intended to be allowed? |
| 01:39 | <MikeSmith> | a-ja: where the spec says it's allowed? |
| 01:39 | <a-ja> | asking cuz <details><style scoped><summary> is/was choked on |
| 01:39 | <MikeSmith> | a-ja: btw dunno if you're aware but blink is removing their style@scoped implementation |
| 01:40 | <Hixie> | http://www.whatwg.org/specs/web-apps/current-work/#the-style-element explains it pretty clearly: |
| 01:40 | <Hixie> | "If the scoped attribute is present: where flow content is expected, but before any other flow content other than inter-element whitespace and style elements, and not as the child of an element whose content model is transparent." |
| 01:40 | SamB | really doesn't much like @global, it seems to contravene half the point of <style scoped> ... |
| 01:40 | <a-ja> | complained about summary having to be 1st child of details |
| 01:40 | <Hixie> | MikeSmith: technically they only removed their incomplete unreleased implementation |
| 01:40 | <MikeSmith> | if the spec allows <details><style scoped><summary> but the validator doesn't then please file a validator but |
| 01:40 | <MikeSmith> | *bug |
| 01:40 | <MikeSmith> | Hixie: ok |
| 01:40 | <Hixie> | validator seems correct per that quote |
| 01:41 | <MikeSmith> | a-ja: yeah that's not a specific problem with style@scoped, it's exactly what it says it is, as far as I can see |
| 01:41 | <MikeSmith> | you need to read the spec |
| 01:41 | <MikeSmith> | in general |
| 01:42 | <a-ja> | i have...didn't know whether i'd missed it somewhere though |
| 01:42 | <a-ja> | easily worked around with a wrapper of some sort |
| 01:42 | <MikeSmith> | a-ja: you missed the part where the spec says the content model of <details> is "One summary element followed by flow content" |
| 01:43 | <MikeSmith> | ? |
| 01:43 | <a-ja> | nah...saw that |
| 01:44 | <a-ja> | wasn't sure there wasn't insistancy with scoped though |
| 01:45 | <Hixie> | the spec is just what the spec says, so if the spec says "One summary element followed by flow content" for <details> and says <style scoped> is expected "where flow content is expected, but before any other flow content", then that's all it is, no hidden magic |
| 01:45 | <a-ja> | know asking might even be premature, as css styled spec is still new |
| 01:50 | <a-ja> | <details><summary><style scoped> seems this wouldn't affect the details element, hence need for wrapper around details |
| 01:51 | <a-ja> | might make a good example in spec |
| 02:18 | <Hixie> | a-ja: why wouldn't it style the <details>? |
| 02:20 | <a-ja> | styles it's parent, i.e. <summary> |
| 02:21 | <a-ja> | right? |
| 02:24 | <Hixie> | oh i thought you meant <details><summary></summary><style scoped></style>...</details> sorry |
| 02:24 | <Hixie> | why would you put it in the <summary>?? |
| 02:24 | <Hixie> | it's not allowed in <summary>... |
| 02:24 | <Hixie> | since <summary> doesn't accept flow content |
| 02:25 | <a-ja> | maybe details could be altered to allow a style scope before summary....so as not to require wrapper to style details |
| 02:26 | <Hixie> | why do you need a wrapped |
| 02:26 | <Hixie> | just put it in the <details> |
| 02:26 | <Hixie> | like the spec says |
| 02:26 | Hixie | doesn't understand the confusion here |
| 02:27 | <a-ja> | details content model says summary followed by flow |
| 02:27 | <Hixie> | right... |
| 02:28 | <a-ja> | details doesn't allow scope style before summary |
| 02:28 | <Hixie> | i don't understand why that matters |
| 02:28 | <Hixie> | how do you get from "details content model says summary followed by flow" to there being a problem with style not being before summary? |
| 02:28 | <a-ja> | matters if you want to style the details element |
| 02:29 | <Hixie> | you still put the style element as a child of the details |
| 02:29 | <Hixie> | you just put it where it's valid |
| 02:29 | <Hixie> | instead of before the summary, where it's not valid... |
| 02:29 | <a-ja> | doh |
| 02:29 | a-ja | was putting it inside summary |
| 02:30 | <a-ja> | no first-child requirement for scope style, i take it |
| 02:31 | <Hixie> | you don't have to assume |
| 02:31 | <Hixie> | read the spec |
| 02:31 | <Hixie> | "If the scoped attribute is present and the element has a parent element, then the style element must precede any flow content in its parent element other than inter-element whitespace and other style elements, and the parent element's content model must not have a transparent component." |
| 02:34 | <a-ja> | and summary is phrasing or heading....not flow. ok |
| 02:36 | <a-ja> | guess now i fear polyfills toggling style element's display...heh |
| 02:37 | <a-ja> | i'll test a few and file bugs on em if a prob |
| 10:02 | <asmodai> | Behaviour-wise I am not sure I understand something wrt JavaScript. Is there a reason why a global map is undefined in an event handler, whereas a global array isn't? |
| 10:04 | <anchnk> | asmodai ###javascript |
| 10:05 | <anchnk> | oups #javascript |
| 10:08 | <jgraham> | asmodai: Got an example? |
| 10:10 | <jgraham> | Uh, what happened to the /topic |
| 10:16 | <Ms2ger> | Huh |
| 10:17 | <asmodai> | jgraham: would need to distill it, just something at work we encountered. |
| 10:20 | <asmodai> | jgraham: Guess topic might have been wiped due to netsplits/netjoins. |
| 10:24 | <annevk> | I cannot set the topic... |
| 10:25 | <annevk> | http://www.whatwg.org/ — logs: http://krijnhoetmer.nl/irc-logs/ & http://logbot.glob.com.au/ — stats: http://gavinsharp.com/irc/whatwg.html — Please leave your sense of logic at the door, thanks! |
| 10:25 | <annevk> | is what it should be, fwiw |
| 10:26 | <Ms2ger> | Dammit ChanServ |
| 10:40 | <annevk> | Ms2ger: can you give me +o so I can fix it? |
| 10:41 | <Ms2ger> | The topic is fixed, I just can't give everyone access to change it |
| 10:45 | <jgraham> | (isn't the convention that you de-op at this point) |
| 10:45 | <annevk> | Ms2ger: it's not using proper dashes |
| 10:46 | <Ms2ger> | I copied yours! |
| 10:46 | <annevk> | Ms2ger: your IRC client is broken it seems |
| 10:46 | <Ms2ger> | Plausible |
| 10:46 | <Ms2ger> | Huh |
| 10:47 | <Ms2ger> | You try? :) |
| 10:47 | annevk | wins |
| 10:48 | <Ms2ger> | Yep, looks better, thanks |
| 10:48 | <annevk> | topic finally uses Unicode em dash rather than hyphens :-) |
| 10:49 | <a-ja> | em dash and spaces? |
| 10:49 | <a-ja> | tsk tsk |
| 10:50 | <annevk> | a-ja: hah, yes! |
| 10:50 | <a-ja> | not in *my* styleguide |
| 10:51 | <Ms2ger> | Not in mine either! |
| 10:51 | <a-ja> | spaces around en dash, but not around em dash |
| 10:52 | <jgraham> | Well in this case it's not being used as a parenthetical—like this—but as a seperator. OTOH, I also think that looks silly without spaces |
| 11:23 | <annevk> | mathiasbynens: yo |
| 11:23 | <annevk> | mathiasbynens: see http://unicode.org/repository/*checkout*/draft/trunk/reports/tr46/tr46.html#ToASCII |
| 11:23 | <annevk> | mathiasbynens: given those definitions, what do you want the methods to do? |
| 11:23 | <annevk> | mathiasbynens: I feel like we should also add domainToUI() and maybe urlToUnicode() and urlToUI() |
| 11:31 | <mathiasbynens> | what does `ToUI` mean? |
| 11:31 | <annevk> | User Interface |
| 11:32 | <mathiasbynens> | i know what UI means but I still don’t get it |
| 11:33 | <mathiasbynens> | what would it do? |
| 11:33 | <annevk> | It would match the algorithm UAs have for whether they show Unicode or Punycode |
| 11:33 | <mathiasbynens> | ah, weird |
| 11:34 | <mathiasbynens> | didn’t we all agree Safari showing Unicode in the address bar at all times is a security issue? |
| 11:34 | <annevk> | See links from http://wiki.whatwg.org/wiki/URL#UI |
| 11:34 | <annevk> | mathiasbynens: that's why you want toUI |
| 11:35 | <annevk> | mathiasbynens: it's up to the UA to decide what it returns |
| 11:36 | <smaug____> | haha, those Attr.specified comments in blink-dev are fun. |
| 11:36 | <smaug____> | jQuery is really missing some testing there |
| 11:37 | <annevk> | yes |
| 11:38 | <jgraham> | smaug____: Internet thought police detected offence "dissing jQuery" you will be deported to MySpace forthwith |
| 11:41 | <smaug____> | oh dear, I must have lived in MySpace for a long time then |
| 11:45 | <jgraham> | Although I guess that MySpace isn't a good analouge for a prison in that it failed because it made it too easy for people to leave. Which kind of suggests that the right analogy is some MySpace successors that are heavily focused on lock-in |
| 11:46 | <mathiasbynens> | annevk: what use cases did you have in mind? |
| 11:46 | <smaug____> | jgraham: Facebook? |
| 11:46 | <annevk> | mathiasbynens: e.g. search results |
| 11:47 | <smaug____> | ahaa, acid3 required certain .specified handling, https://bugzilla.mozilla.org/show_bug.cgi?id=199959#c6 |
| 11:53 | <annevk> | smaug____: did you review https://dvcs.w3.org/hg/innerhtml/raw-file/tip/index.html#dfn-concept-serialize-xml |
| 11:54 | <Ms2ger> | I reviewed parts of it |
| 12:00 | <annevk> | It seems if namespace is '"' that algorithm might produce garbage |
| 12:00 | <Ms2ger> | Plausible |
| 12:00 | <annevk> | Oh, data:text/xml,<s xmlns='"'/> is a parse error in Chrome but not Firefox |
| 12:02 | <annevk> | Same for data:text/xml,<s xmlns="<"/> |
| 12:03 | <smaug____> | I didn't review that, if that is the thing travis changed recently |
| 12:39 | <JakeA> | annevk: Any reasons XHR send() couldn't return a promise? |
| 12:40 | <Ms2ger> | Legacy? |
| 12:41 | <jgraham> | Depends on whether anyone depends on it returning undefined |
| 12:41 | <jgraham> | Stranger things have happened |
| 12:59 | <JakeA> | Is there any precedent here? Has switching from return undefined to return obj been an issue before? |
| 13:20 | <Ms2ger> | JakeA, the precedent is that any change that's been made has caused issues :) |
| 13:23 | <JakeA> | Anything particular to undefined returns becoming objects? |
| 13:24 | <Ms2ger> | I don't remember a case |
| 13:25 | <JakeA> | The alternative is adding an extra .ready() method on there, of course. But legacy aside, .send() is the right place for it |
| 13:26 | <Ms2ger> | Legacy aside, XHR is crap :) |
| 13:27 | <JakeA> | This is why ServiceWorker has fetch() |
| 13:36 | <annevk> | JakeA: I don't understand why we'd give XHR a promise if we're going to have fetch() |
| 13:37 | <JakeA> | That's fair |
| 13:39 | <annevk> | Having said that, I have added other features to XMLHttpRequest recently |
| 13:49 | <annevk> | JakeA: so what is the state of things? |
| 13:50 | <annevk> | JakeA: you and Jungkee are writing the spec? |
| 13:50 | <JakeA> | & slightlyoff, yeah |
| 13:50 | <annevk> | JakeA: does it have a nice URL yet? |
| 13:51 | <JakeA> | annevk: Nope https://github.com/slightlyoff/ServiceWorker/issues/197 |
| 13:51 | <JakeA> | I've been working in the ts file & https://github.com/slightlyoff/ServiceWorker/blob/master/spec/service_worker/algorithms.md |
| 13:52 | <annevk> | JakeA: is there's an output file somewhere? |
| 13:52 | <JakeA> | annevk: Lifecycle is almost there. I've taken a swing at caching. It's been declared "not insane" by some Chromium cache guys, but I want us to look at it in detail at our next meetup |
| 13:53 | <annevk> | JakeA: is 7-8 still on? I have booked flights fwiw |
| 13:53 | <JakeA> | annevk: The spec is at https://github.com/slightlyoff/ServiceWorker/blob/master/spec/service_worker/index.html - Jungkee has been doing that, but I want it moved to gh-pages |
| 13:53 | <JakeA> | annevk: Yeah, 7-8 is go |
| 13:53 | <annevk> | JakeA: great |
| 13:54 | <annevk> | JakeA: marcosc might be joining us as I understand things |
| 13:54 | <JakeA> | annevk: Cool! |
| 13:54 | <JakeA> | annevk: I want to take the opportunity to look at HTML imports render-blocking too |
| 13:54 | <annevk> | JakeA: also, I starting around then I should have time to help out with the specification |
| 13:54 | <JakeA> | annevk: Maybe not on 7-8 though |
| 13:55 | <JakeA> | annevk: Oh brilliant! |
| 13:55 | <Ms2ger> | annevk, really? We're not giving you enough to do? :) |
| 13:55 | <JakeA> | annevk: Any idea if Mozilla plans to make HTML imports render-blocking? |
| 13:57 | <annevk> | Ms2ger: well, we want this to happen |
| 13:57 | <annevk> | JakeA: no, forgot who is implementing that too... |
| 13:58 | annevk | registered for BlinkOn 2 |
| 15:04 | <annevk> | SimonSapin: http://ln.hixie.ch/?start=1140242962&count=1 |
| 15:47 | <dglazkov> | good morning, Whatwg! |
| 15:51 | <SamB> | so, whose idea was it to name an engine after the dumbest tag ever anyway? |
| 15:58 | <jgraham> | Hahaha |
| 15:58 | <jgraham> | If you think <blink> is the worst HTML tag you need to get out more |
| 15:59 | <jgraham> | Or stay in more |
| 15:59 | <jgraham> | Probably the latter |
| 15:59 | <SamB> | I didn't say "worst" |
| 15:59 | <jgraham> | Well |
| 15:59 | <jgraham> | <span> is the dumbest tag ever since it doesn't do or mean anything |
| 15:59 | <SamB> | hehehe |
| 16:00 | <SamB> | you mean, sort of like how git is "the stupid content tracker"? |
| 16:02 | SamB | guesses he's just still bitter about how <blink> renders the contents invisible for so long at a time |
| 16:03 | <jgraham> | That's accidentially good because it means it's far outside the range that triggers photosensitive epilepsy |
| 16:04 | <SamB> | I gues I would have preferred if it blinked between different colors instead or something |
| 16:34 | <annevk> | SimonSapin: it seems the only somewhat complicated rewinding case to rewrite to prepend would be utf-16 |
| 16:34 | <annevk> | SimonSapin: when you hit an unpaired surrogate |
| 16:34 | <SimonSapin> | annevk: don’t you keep the first surrogate around? |
| 16:35 | <annevk> | SimonSapin: yeah, complicated as in that you have to convert it back to bytes |
| 16:35 | <annevk> | SimonSapin: although I guess I can just use http://encoding.spec.whatwg.org/#convert-a-code-unit-to-bytes |
| 16:35 | <annevk> | SimonSapin: not too hard after all... |
| 16:36 | <SimonSapin> | I was gonna say something like that, but if you already have a hook for it it’s even easier |
| 16:40 | <annevk> | So public-webapps reached agreement it would be great to split a specification without actually having an editor for anything? |
| 16:40 | <annevk> | Are all these people managers? |
| 16:41 | <jgraham> | Well none of them are editors |
| 16:42 | <jgraham> | Clearly |
| 16:47 | <Domenic_> | i kind of tuned out after they said the only purpose of this split was to advance it through the w3c process |
| 16:54 | <Hixie> | which spec? |
| 16:55 | <Hixie> | selection? |
| 16:55 | <Hixie> | isn't that just part of aryeh's spec? |
| 16:57 | <jgraham> | Hixie: Hence "to split a specification" |
| 16:58 | <Hixie> | well if splitting it gets it an editor, i guess that's a win? |
| 16:58 | <Hixie> | at least they're not forking an actively maintained spec for once |
| 16:58 | <hober> | my impression was that rniwa was volunteering to edit the selection spec |
| 16:59 | <SamB> | yay for selections |
| 17:02 | <Hixie> | yeah looks like rniwa and aryeh are talking, so i'm happy |
| 17:02 | <SamB> | oh darn, opacity affects the selection :-( |
| 17:04 | <annevk> | Hixie: I missed rniwa volunteering |
| 17:04 | <annevk> | Sorry for the noise everyone :/ |
| 17:31 | <SamB> | darn |
| 17:32 | <SamB> | I hit the "submit" button and then notice I screwed up the terminology ... |
| 17:42 | <SamB> | so, um, I just reported <https://www.w3.org/Bugs/Public/show_bug.cgi?id=25145>; and I'm wondering where, say, :active's semantics are assigned ... |
| 17:43 | <TabAtkins> | A combination of HTML and Selectors. |
| 17:45 | <SamB> | yes, that much I see on https://developer.mozilla.org/en-US/docs/Web/CSS/Pseudo-classes |
| 17:46 | <SamB> | oh, hmm, I guess I could go to the MDN page for it ... |
| 17:47 | <SamB> | ... but that doesn't mention HTML5 or the Living Standard at all |
| 17:47 | <Hixie> | it's a wiki :-) |
| 17:48 | <SamB> | well I guess I could try googling ... |
| 17:50 | <SamB> | gah, punctuation blindness strikes again :-( |
| 18:12 | <SamB> | there isn't a pseudo-class for things that are (at least partially) selected, is there? |
| 18:14 | <TabAtkins> | No. |
| 18:17 | <annevk> | ::selection is dead? |
| 18:17 | <annevk> | I guess it is |
| 18:17 | <annevk> | Note that the hit testing part of :active is undefined |
| 18:20 | <TabAtkins> | annevk: Yup, dead. We'll revive, but we still haven't settled on what it actually *means*. dbaron had a decent model, at least 3 years ago, but hasn't written anything down. :/ |
| 18:38 | <dbaron> | annevk, TabAtkins, my post was http://lists.w3.org/Archives/Public/www-style/2008Oct/0268.html -- not so much a model as a list of things to consider when making one |
| 19:27 | <SamB> | annevk: ::selection doesn't do what I want, anyway; I want something more like :hover/:active, that would select existing nodes that contain some part of the selection (including all of the parent elements) |
| 19:28 | <SamB> | mostly to cancel an "opacity: 0" so that it's possible to SEE the selection |
| 19:30 | <TabAtkins> | SamB: If you're trying to do a spoiler element, just use <details>. It's what it's for. |
| 19:31 | SamB | thinks he'd have a tough time marketing that to Stack Exchange ... |
| 19:34 | <TabAtkins> | Why? |
| 19:39 | SamB | may simply be ignorant of <details>' vintage |
| 19:41 | <TabAtkins> | I don't understand. |
| 19:43 | <TabAtkins> | Hm, I thought you could use unicode in JS variable names. Am I wrong? |
| 19:55 | <jgraham> | TabAtkins: Define "unicode" :) |
| 19:56 | <jgraham> | (I think it should be possible, excluding certain character classes for an implementation-defined version of unicode. But I may have misremembered) |
| 19:56 | <TabAtkins> | Right. Well, you can definitely use "non-ASCII" in variable names, because fóó is valid. But f💩 isn't. |
| 19:56 | <TabAtkins> | Nor are things like f© |
| 20:00 | <jgraham> | TabAtkins: http://www.ecma-international.org/ecma-262/5.1/#sec-7.6 |
| 20:01 | <TabAtkins> | Ah, cool. Thanks. |
| 20:06 | <mathiasbynens> | TabAtkins, jgraham: or http://mathiasbynens.be/notes/javascript-identifiers |
| 20:06 | <mathiasbynens> | …and http://mothereff.in/js-variables |
| 20:07 | <mathiasbynens> | …and http://mathiasbynens.be/demo/javascript-identifier-regex and https://gist.github.com/mathiasbynens/6334847 |
| 22:16 | SamB_ | goes to check http://catb.org/~esr/writings/taoup/html/modularitychapter.html in validator.nu ... |