| 00:31 | <tfh> | anyone know if "origin" values, which make an appearance in the Web Sockets draft, are fully defined by http://tools.ietf.org/html/draft-abarth-origin-02, or elsewhere? |
| 00:42 | <Hixie> | yeah that's where they're supposed to be defined |
| 00:42 | <Hixie> | tfh: if there's a problem either with websockets or abarth's origin draft, here's a good place to ask about it before mailing the lists |
| 01:28 | <tfh> | Hixie: thanks |
| 02:49 | <Darxus> | How do you feel about me doing two column page layouts using the table element, where each column is a language? With a thead containing <img src="languageflag.svg" alt="language" title="language" /> ? |
| 02:49 | <Darxus> | Is that sufficiently tabular? |
| 02:50 | <Darxus> | To qualify as HTML5 conformant? |
| 02:54 | <Darxus> | Seems like it would probably be easier to read with a screen reader than a bunch of <div style="float: left;">'s. |
| 03:54 | <jwalden> | wait, I'm older than annevk? what the heck, why did I think otherwise? :-\ |
| 04:01 | <Darxus> | /clear |
| 04:54 | <Hixie> | man i've really no idea what's going on in the htmlwg these days |
| 04:54 | <Hixie> | i feel like i missed something about 2 weeks ago and i've been confused ever since |
| 05:01 | <tantek> | Hixie - I believe that was around the time I joined as an invited expert, or maybe that was only a week ago. Probably just a coincidence. |
| 05:02 | <Hixie> | hah |
| 05:11 | <tantek> | apologies for any disturbances I may have inadvertently caused. |
| 05:31 | <othermaciej> | things I learned today: Sam has no tolerance for procedural games |
| 05:33 | <MikeSmith> | tantek has a caused a disturbance in the Force |
| 05:34 | <MikeSmith> | I think to balance it out will require some human sacrifice |
| 05:35 | <othermaciej> | MikeSmith: do you personally think your document is ready to be published as WD by the way? |
| 05:36 | <MikeSmith> | othermaciej: no, I don't. I suppose I should send a message to the list to make that clear. |
| 05:36 | <othermaciej> | MikeSmith: probably! Sam did ask |
| 05:37 | <MikeSmith> | OK |
| 05:59 | <othermaciej> | I'm really curious if Manu is actually planning to propose DOM changes as part of his HTML5+RDF thing |
| 05:59 | <othermaciej> | that would certainly raise the stakes |
| 06:02 | <MikeSmith> | yeah, definitely would |
| 06:03 | <MikeSmith> | othermaciej: so I just sent a reply to the list, cc'ed you and Sam |
| 06:03 | <MikeSmith> | I doesn't seem to have shown up on the list yet, afaict |
| 06:04 | <othermaciej> | MikeSmith: I saw it (maybe from the Cc) |
| 06:04 | <MikeSmith> | OK |
| 06:06 | <othermaciej> | I still don't like the basic multiple drafts idea, but Sam seems to be administering it in a fairly ad-hoc fashion |
| 09:01 | <annevk> | thanks gsnedders, ezyang :) |
| 09:11 | <GPHemsley> | what happened to the width attribute on <col>? |
| 09:11 | <annevk> | CSS |
| 09:11 | <Hixie> | http://www.whatwg.org/specs/web-apps/current-work/#attr-col-width |
| 09:13 | GPHemsley | cires at one-page links ;_; |
| 09:14 | <Hixie> | you can just add multipage/ into the URL :-) |
| 09:14 | <Hixie> | it'll autoredirect you to the right place |
| 09:14 | <GPHemsley> | oh, that's nice of it :) |
| 09:15 | <GPHemsley> | so is style="width: N%;" the same? |
| 09:17 | <Hixie> | style="" is allowed in HTML5, but generally i would recommend using an external style sheet |
| 09:18 | <GPHemsley> | I'm questioning whether that CSS will give me the same output |
| 09:18 | <annevk> | did you try? |
| 09:18 | <GPHemsley> | no :) |
| 09:20 | <GPHemsley> | could someone remind me what the current standing is with regard to closing empty elements with the trailing slash? |
| 09:21 | <GPHemsley> | (i.e., e.g. <img />) |
| 09:21 | <othermaciej_> | GPHemsley: I think the only difference would be cascade order |
| 09:22 | <GPHemsley> | ? |
| 09:22 | <othermaciej_> | (between using the width="" attribute and either inline style or a stylesheet rule) |
| 09:22 | <GPHemsley> | ah |
| 09:23 | <othermaciej> | for void elements, it's allowed but not required to use the trailing slash syntax |
| 09:23 | <othermaciej> | you can't use that syntax for elements that aren't labeled as "void elements" though, even if they happen to be empty |
| 09:23 | <othermaciej> | so <div /> for example is not kosher, in the text/html serialization |
| 09:23 | <GPHemsley> | am I gonna get complaints about NET-enabled start tags, though? |
| 09:23 | <GPHemsley> | or whatever that dumb error is |
| 09:24 | <othermaciej> | not in HTML5 |
| 09:24 | <GPHemsley> | ok |
| 09:24 | <GPHemsley> | so logic prevails for a change :) |
| 09:25 | Hixie | points at the topic |
| 09:25 | <GPHemsley> | hah |
| 09:25 | <othermaciej> | the original illogic was allowing sending of XHTML with text/thml MIME type |
| 09:27 | <othermaciej> | or alternately you could say it was illogical to keep specifying HTML as an SGML application |
| 09:28 | <GPHemsley> | but I can keep text/html and <void />, right? |
| 09:28 | <GPHemsley> | in HTML5 |
| 09:30 | <othermaciej> | yes |
| 09:30 | <GPHemsley> | OK, then I'm happy :) |
| 09:30 | <GPHemsley> | <3 Hixie for fixing the web |
| 09:30 | <othermaciej> | I don't think Hixie is terribly proud of this particular detail in the spec :-) |
| 09:31 | <Hixie> | i hate the /> crap :-) |
| 09:31 | <Hixie> | i wish people wouldn't ever use /> in text/html :-) |
| 09:31 | <GPHemsley> | oh... but that's because you follow the topic ;) |
| 09:31 | <GPHemsley> | I think it makes much more sense to note the beginning and end of every tag |
| 09:31 | <GPHemsley> | then nothing can run away |
| 09:32 | <Hixie> | unfortunately, unlike what some people on public-html seem to think, i don't get my way when i'm wrong :-( |
| 09:32 | <GPHemsley> | heh |
| 09:33 | <GPHemsley> | for the record, it does appear that the col width and CSS width are the same, thankfully :) |
| 09:38 | <GPHemsley> | well... the index is rather useless ATM... |
| 09:40 | <GPHemsley> | good to see ' defined, too :) |
| 09:41 | <GPHemsley> | hmm... comments still aren't allowed to have -- ? |
| 09:43 | <zcorpan> | GPHemsley: right. it has weird behavior in old browsers that implement sgml-style comments, and also you wouldn't be able to convert it to xml |
| 09:43 | <GPHemsley> | ah, OK |
| 09:43 | <zcorpan> | annevk: grattis |
| 09:44 | GPHemsley | wonders why the numbered lists don't use different numbering systems, to avoid phrases like "step 2 in these inner steps" |
| 09:48 | <Hixie> | GPHemsley: these days i name the steps |
| 09:48 | <Hixie> | GPHemsley: any remaining cases of "jump to step 2" or stuff like that is just old text i haven't updated recently |
| 09:49 | <GPHemsley> | Hixie: http://www.whatwg.org/specs/web-apps/current-work/multipage/syntax.html#determining-the-character-encoding |
| 09:55 | <GPHemsley> | Hixie: Also, any word on this? http://www.w3.org/Bugs/Public/show_bug.cgi?id=7150 |
| 09:56 | <Hixie> | when i get to it, i'll comment on the bug :-) |
| 09:57 | <GPHemsley> | heh, OK :P |
| 10:08 | <zcorpan> | GPHemsley: hixie has said in email before that examples intentionally vary as to not give a perception that one coding style is discouraged or anything |
| 10:08 | <GPHemsley> | oh, really? |
| 10:08 | <GPHemsley> | that's too bad... |
| 10:09 | <zcorpan> | some examples use unquoted attributes, some omit tags, etc |
| 10:09 | <zcorpan> | why is it too bad? |
| 10:10 | <GPHemsley> | because I prefer consistency and unambiguity (e.g. lowercase elements and quoted attributes) |
| 10:13 | <zcorpan> | other people may prefer typing less markup that isn't needed |
| 10:13 | <Hixie> | uppercase elements and unquoted attributes aren't ambiguous |
| 10:14 | <Hixie> | and consistency gives the perception that other coding styles are discouraged |
| 10:14 | <zcorpan> | or holding down the shift button for all three keystrokes in "<P>" |
| 10:14 | <GPHemsley> | Hixie: well, that would be my intention ;) |
| 10:15 | <zcorpan> | um, i meant "</P>" |
| 10:15 | <Hixie> | well, if we're going to be consistent, i vote for uppercase element names, no />s ever, no quotes around attributes unless absolutely necessary, and all omittable tags always omitted |
| 10:15 | <zcorpan> | on swedish keyboard layout you have to hold shift for the slash and > |
| 10:15 | <GPHemsley> | ah |
| 10:15 | <krijnh> | Holding down shift for </P> creates a PI! We need PIs in HTML5! |
| 10:15 | <krijnh> | OMG, why aren't PIs allowed, grmbl grmbl |
| 10:16 | <Hixie> | what would a PI be allowed for |
| 10:16 | <zcorpan> | php! |
| 10:16 | <krijnh> | For PHP, I want to validate my PHP as HTML! |
| 10:16 | <Hixie> | wouldn't PHP use... PHP, rather than HTML? |
| 10:16 | <GPHemsley> | Hey, I'm just talking about my personal opinion here. |
| 10:16 | <krijnh> | You guys are taking away my ability to validate my PHP, you suck! |
| 10:16 | <Hixie> | GPHemsley: right, but where the spec allows multiple things, the spec doesn't have an opinion :-) |
| 10:16 | <krijnh> | Okay, enough for today :) |
| 10:17 | <Hixie> | i think Turing took that ability away |
| 10:17 | <Hixie> | but anyway |
| 10:17 | <GPHemsley> | Hixie: Right. I get that. I'm content to drop the issue. :) |
| 10:17 | <krijnh> | Heh |
| 10:17 | <krijnh> | (Do people mailing this stuff to the list know they are wasting so much of other peoples time?) |
| 10:18 | <takkaria> | no |
| 10:18 | <GPHemsley> | As long as the spec lets me do it my way, I'm good. |
| 10:18 | <takkaria> | because leif wants to validate his HTML as PHP. :) |
| 10:18 | <krijnh> | (When I'm reading this stuff on my phone I'm really not that happy with being subscribed to the list :() |
| 10:18 | <Hixie> | krijnh: i'm sure they're trying to do the right thing |
| 10:18 | <takkaria> | er, his PHP as HTML |
| 10:18 | <GPHemsley> | And, in case it's not already obvious, I don't read the mailing list. :) |
| 10:18 | <Hixie> | can't say i blame you |
| 10:18 | <GPHemsley> | heh |
| 10:19 | <krijnh> | I wish I wasn't as addicted to it :) |
| 10:19 | zcorpan | would like a filter that marks all non-technical emails as read |
| 10:20 | <zcorpan> | i have developed such a filter in my brain, but it takes more time than i'd like |
| 10:20 | <zcorpan> | and sometimes it doesn't work and i read silly emails anyway |
| 10:20 | <krijnh> | At least you get payed for it :) |
| 10:21 | <GPHemsley> | BTW, I got flack from people on the Wordpress forum for attempting to nest <em> tags, as allowed by HTML5. |
| 10:22 | <GPHemsley> | Wordpress didn't like that, and decided to close the first <em> tag before the second one |
| 10:22 | <GPHemsley> | And it bugs me when people (well, machines) mess with my HTML without my permission |
| 10:26 | <krijnh> | http://www.sitepoint.com/blogs/2009/07/31/html-4-considered-harmful/ oh noes! |
| 10:30 | <Philip`> | http://blog.whatwg.org/ - <p style="display: none"> <a href="http://childrenscenter.boisestate.edu/CFIDE/gettingstarted/refusion.php?p=sildenafil-rezeptfrei">viagra rezeptfrei</a> |
| 10:30 | <Philip`> | I don't think cleaning up the templates every few days is a great way to go |
| 10:30 | <krijnh> | Philip`: that's Microdata :) |
| 10:30 | <Hixie> | i thought lachy had fixed it |
| 10:30 | <Philip`> | He had |
| 10:31 | <Philip`> | but presumably nobody has fixed the actual vulnerability yet |
| 10:31 | <Hixie> | oh |
| 10:31 | <Hixie> | i thought that's what he'd fixed |
| 10:31 | <Philip`> | or the server's more deeply infected, or something |
| 10:31 | <Philip`> | Evidently it's not fixed :-) |
| 10:32 | <Hixie> | clearly |
| 10:36 | <Philip`> | It's lucky that these people never want to actually cause any direct harm and e.g. delete all the site's content, they just want to spam search engines |
| 10:39 | <annevk> | zcorpan, Swedish? |
| 10:39 | <annevk> | Google says yes; bedankt :) |
| 10:39 | <Hixie> | Philip`: i have backups of the database, but yes |
| 10:39 | <krijnh> | 23 pas, jonkie |
| 10:40 | <annevk> | mag nog steeds ongestraft flamen :p |
| 10:40 | <krijnh> | Ja, dat je daar mee weg komt, wow :) |
| 10:41 | <Hixie> | i thought lachy was going to set the perms so nothing was writable or something |
| 10:45 | <GPHemsley> | what's the deal with the accesskey attribute? |
| 10:47 | <Hixie> | is there a deal? |
| 10:48 | <GPHemsley> | http://dev.w3.org/html5/html4-differences/ says it's absent |
| 10:48 | <Hixie> | then that document is wrong :-) |
| 10:48 | <GPHemsley> | that's what I thought |
| 10:49 | <krijnh> | Somebody should submit a new document, so we can have a poll about it |
| 10:50 | <krijnh> | Okay, I will write some text. |
| 10:50 | <annevk> | GPHemsley, refresh |
| 10:50 | <GPHemsley> | annevk: :) |
| 10:51 | <annevk> | would "Gordon Hemsley" be ok for the acknowledgments section? |
| 10:51 | <GPHemsley> | I prefer Gordon P. Hemsley |
| 10:51 | <annevk> | k |
| 10:52 | <annevk> | refresh |
| 10:52 | <krijnh> | I formally object to this refreshing stuff! |
| 10:52 | <GPHemsley> | ^_^ |
| 10:54 | <GPHemsley> | annevk: Happy Birthday, BTW. :) |
| 11:01 | <annevk> | Hixie, btw, when you changed property to itemprop you did not change the order of global attributes so they're no longer a-z now |
| 11:01 | <annevk> | GPHemsley, ta |
| 11:04 | <gsnedders_> | Hixie: Lachy did set the perms like that… |
| 11:06 | <Hixie> | annevk: file a bug using the new box :-) |
| 11:07 | <Philip`> | Did Lachy set the permissions in such a way that the attacker can't simply switch them back? |
| 11:08 | Philip` | hopes the attacker only has complete access to the machine under the web server's username |
| 11:09 | <annevk> | Hixie, where should I file bugs to suggest that the form is cleared after a bug has been filed? |
| 11:09 | <Hixie> | to file that bug, after you file a bug, click on the field and press Ctrl+A |
| 11:09 | <Hixie> | Cmd+A on mac |
| 11:09 | <gsnedders_> | :P |
| 11:10 | <annevk> | mwaha |
| 11:10 | <annevk> | also, it does not seem to accurately track what I'm looking at |
| 11:10 | <Philip`> | Clearing it would prevent accidentally double-clicking on the submit button |
| 11:10 | <annevk> | I first have to click the screen or something... |
| 11:12 | <krijnh> | annevk: html4-difference still mentions the eventsource element |
| 11:12 | <krijnh> | differences even |
| 11:13 | <gsnedders_> | Hixie: Useful answer re the spam in #wordpress: talk to your host |
| 11:13 | <Hixie> | gsnedders_: my host is lachy :-) |
| 11:13 | <gsnedders_> | :P |
| 11:13 | <GPHemsley> | Yeah, those Wordpress people are always very helpful |
| 11:14 | <krijnh> | annevk: also key words is one word, I think |
| 11:15 | <GPHemsley> | krijnh: I agree, but they're using it consistently as two |
| 11:15 | <GPHemsley> | in the main spec, too |
| 11:15 | <krijnh> | HTML 5 itself uses "2.4.3 Keywords and enumerated attributes" |
| 11:15 | <GPHemsley> | ...except there |
| 11:17 | <gsnedders_> | Hixie: Now apparently Apache is at fault |
| 11:18 | gsnedders_ | sighs |
| 11:19 | gsnedders_ | sighs |
| 11:19 | <GPHemsley> | Is the official HTML5 validator w3.org's or validator.nu's? |
| 11:19 | GPHemsley | pats gsnedders_ on the back |
| 11:19 | <Hixie> | there is no official HTML5 validator |
| 11:19 | <Hixie> | and hopefully there will never be one |
| 11:19 | <krijnh> | annevk: <img border> and <script language> are not allowed, according to html5.validator.nu |
| 11:20 | <GPHemsley> | Hixie: recommended, perhaps? |
| 11:20 | <gsnedders_> | Hixie: Could I possibly get SSH access to the blog at least? |
| 11:20 | <Hixie> | gsnedders_: hm? |
| 11:20 | <Hixie> | gsnedders_: it's under lachy's account |
| 11:20 | <gsnedders_> | Hixie: Ah |
| 11:20 | <annevk> | krijnh, bug in validator.nu |
| 11:20 | <GPHemsley> | Hixie: Or, last resort, your preference? :) |
| 11:20 | <gsnedders_> | Hixie: So the answer is ask Lachy? :P |
| 11:20 | <annevk> | krijnh, fixed the other nits |
| 11:20 | <Hixie> | gsnedders_: yes |
| 11:20 | <krijnh> | annevk: ah, okay |
| 11:20 | <gsnedders_> | Lachy: ping |
| 11:20 | <Hixie> | GPHemsley: valkidator.whatwg.org |
| 11:20 | <gsnedders_> | Where has Lachy been anyway? |
| 11:22 | <Hixie> | dunno |
| 11:22 | <Hixie> | but i'm going to bed |
| 11:22 | <Hixie> | nn |
| 11:22 | <gsnedders_> | nighty |
| 11:26 | <krijnh> | annevk: die puntjes zijn geen acknowledgment waard imho.. |
| 11:27 | <zcorpan> | Hixie: have you seen http://projectcerbera.com/do#html ? |
| 11:28 | <annevk> | krijnh, net als Hixie voeg ik een naam toe voor elke wijziging die ik maak dankzij feedback van die persoon |
| 11:29 | <krijnh> | Ow okay |
| 11:29 | <annevk> | hoeft er ook niemand te bepalen wat belangrijk is of niet |
| 11:29 | <krijnh> | True |
| 11:32 | <MikeSmith> | is there a way in CSS to make the list number markers in <ol> lists right-aligned instead of left-aligned? |
| 11:33 | <takkaria> | flip the rtl setting, and then flip it back again for its il children? |
| 11:33 | <takkaria> | *li |
| 11:33 | <takkaria> | though maybe they're ill too, I really have no idea |
| 11:34 | <takkaria> | oh, I think I totally misunderstood you there |
| 11:34 | <takkaria> | nevermind |
| 11:39 | gsnedders_ | shudders at the thought of that, knowing some browsers don't cope with it well |
| 11:40 | <jmb> | I guess you could do something with ol li::marker { text-align: right; }, assuming anything actually implements that :) |
| 11:41 | <MikeSmith> | takkaria, jmb : thanks |
| 11:41 | <krijnh> | li { direction: rtl; unicode-bidi: embed; text-align: left; } |
| 11:41 | <krijnh> | Works great :) |
| 11:41 | <krijnh> | Even without the unicode-bidi: embed; |
| 11:44 | <Lachy> | gsnedders_, send me a public key |
| 11:45 | <gsnedders_> | Lachy: What's your email? |
| 11:45 | <Lachy> | surely it's in your address book? |
| 11:45 | <krijnh> | Don't you guys work at the same company? |
| 11:45 | <Lachy> | lachlan.hunt⊙lia |
| 11:45 | <gsnedders_> | Lachy: Probably |
| 11:46 | <gsnedders_> | krijnh: Shhh |
| 11:46 | <Lachy> | gsnedders_, after I've added it, log in using: ssh lhunt⊙bwo |
| 11:46 | gsnedders_ | can claim to be Lachy! |
| 11:47 | <gsnedders_> | Lachy: sending |
| 11:47 | <gsnedders_> | Lachy: sent |
| 11:50 | <annevk> | Lachy, I forwarded you some information on a better skin |
| 11:50 | <annevk> | Lachy, let me know if I can help |
| 11:50 | <annevk> | Lachy, getting rid of the default would be great |
| 11:51 | <annevk> | Lachy, I think we should make some changes to the template pages as well; I can do that too if I get access |
| 11:53 | <Lachy> | gsnedders_, done. |
| 11:53 | <Lachy> | annevk, send a public key |
| 11:54 | <annevk> | I've no idea where I stored my public key... |
| 11:55 | <annevk> | maybe this is best done later |
| 11:55 | <gsnedders_> | ~/.ssh? |
| 11:55 | <Lachy> | just send a copy of your authorized_keys file that you have on dreamhost |
| 11:55 | <annevk> | I should clean up a little |
| 11:55 | <Lachy> | or you should be able to extract your public key from your private key file, if you know the right command |
| 11:56 | <Lachy> | I think it's the openssl rsa or rsautl command |
| 11:56 | <gsnedders_> | Why on earth is there a fly flying around here when windows are so wide open? |
| 11:57 | <Lachy> | gsnedders_, having the windows open would explain how it came in |
| 11:58 | <gsnedders_> | Indeed, but it's doing amazingly well at failing to leave |
| 11:58 | gsnedders_ | hears a thud behind him and sees a corpse drop |
| 11:59 | <Lachy> | gsnedders_, did the fly win or lose? |
| 11:59 | <gsnedders_> | Well, I guess it would say it lost, if it were alive to say so |
| 12:03 | <Lachy> | gsnedders_, what did you want access to the blog for anyway? |
| 12:08 | <Lachy> | I really don't understand how the hacker got into the templates again. I definitely changed the file permissions for the template to be -r--r--r-- |
| 12:08 | <Lachy> | but now they're -rw-r--r-- |
| 12:12 | <gsnedders_> | Lachy: Try and see if I can work out how they did again |
| 12:13 | <Lachy> | gsnedders_, ok. I'll just restore a copy of our original templates, which had a few modificiations like showing the author and stuff |
| 12:20 | <Lachy> | what the hell? When I try to delete the hacked template directory, I get permission denied errors, despite me being the owner of the files and having write permission set |
| 12:20 | <gsnedders_> | Lachy: where? |
| 12:22 | <Lachy> | oh, nevermind. The directory itself didn't have write permission set |
| 12:22 | <Lachy> | I thought I'd looked at that ealier and it did. |
| 12:28 | <Lachy> | I have now restored the original, slightly modified, template and set the permissions to -r--r--r-- and dr-xr-xr-x for everything in there |
| 12:29 | <Lachy> | and I have created a complete backup copy in the themes directory |
| 12:31 | <Lachy> | I wonder if it's possible for the permissions to be changed by some code in PHP, and whether wordpress does that for any reason |
| 12:32 | <Lachy> | if it does, that could be how we got hacked again |
| 13:04 | <Lachy> | actually, it looks like I do have access to the logs in ~/logs/ |
| 13:04 | <Lachy> | didn't realise that was there |
| 15:50 | <Philip`> | "Your PeopleSoft connection has expired. For increased security on this site, connections are expired after 60 minutes of inactivity. Your PeopleSoft session has expired. Close all browser windows before logging in again. If this is your only active PeopleSoft session, click the Sign In link to sign in again. <a href=...>Sign in to PeopleSoft</a>" |
| 15:50 | <Philip`> | Glad to see the PeopleSoft people aren't passing up an opportunity to make me aware of their software |
| 15:52 | <takkaria> | oh, yeah, PeopleSoft's stuff is awesome |
| 15:53 | <takkaria> | manchester uni's instance of their student software has "In the interests of performance and efficiency, this website supports the Western European Character Set (ISO 8859/1)" written by the login box |
| 15:54 | <gsnedders_> | takkaria: But I want to pay in €! |
| 15:54 | <Philip`> | I bet you wouldn't dare entering any non-ASCII characters, though |
| 15:55 | <jcranmer> | no, everyone should pay in ¤! |
| 15:56 | <Dashiva> | Yeah, UTF-8 is such a performance hog |
| 15:56 | <gsnedders_> | Hmm, we should use U+FFFF for our own WHAT currency |
| 15:56 | <Dashiva> | Use U+FFFD instead, easier to "fix" payments then |
| 15:58 | <gsnedders_> | Dashiva: But where's the fun in using a character you can send over XML? |
| 16:00 | <Dashiva> | gsnedders: Fun is sacrificed for profit |
| 16:00 | <Philip`> | I want no part in any regime that sacrifices fun |
| 16:02 | <Dashiva> | You can buy fun with the profits |
| 16:03 | <Philip`> | What's the exchange rate between money and happiness? |
| 16:04 | <Dashiva> | Depends on the currency |
| 16:04 | <Dashiva> | (of the happiness) |
| 16:05 | <Philip`> | Do it in Standard Happiness Units, and then I can easily convert it to any other scale |
| 16:07 | <Dashiva> | Well, if the person accepts happiness in Ham Sandwich Equivalents, the exchange rate is quite favorable |
| 19:17 | <gsnedders_> | If you have an object in ECMAScript and iterate over it with for … in, what happens if you add stuff to the object? |
| 19:17 | <Dashiva> | You mean introduce a new property inside the loop? |
| 19:17 | <gsnedders_> | Yeah |
| 19:18 | <gsnedders_> | Does it iterate over a copy, or does it iterate over the variable, and hence loop infinitely? |
| 19:18 | <takkaria> | most languages say that behaviour is undefined for that case I believe |
| 19:19 | <gsnedders_> | According to ECMAScript 5 it copies |
| 19:19 | <Dashiva> | Yeah, it's up to the implementation |
| 19:20 | <Dashiva> | ES3 says that deleted properties must not be visited, and added properties may be visited. |
| 19:21 | <gsnedders_> | "If new properties are added to the object being enumerated during enumeration, the newly added properties are guaranteed not to be visited in the active enumeration. " |
| 19:21 | <gsnedders_> | (ES5) |
| 19:22 | <Dashiva> | What if you add a property to the object's prototype? |
| 19:22 | <Dashiva> | And what if you change the prototype chain during enumeration? |
| 19:23 | <gsnedders_> | That has no effect |
| 19:23 | <gsnedders_> | You effectively copy the object at the start of evaluating the loop |
| 19:24 | <inimino> | AIUI the set of properties to be visited is fixed at the start and only deletions will affect it |
| 19:25 | <Dashiva> | So if you change the prototype, you'll end up visiting several properties that don't actually exist? |
| 19:27 | <gsnedders_> | Hmm, I'm not sure |
| 19:27 | gsnedders_ | isn't an ES guy, damnit |
| 19:27 | <inimino> | I'm not sure about that |
| 19:27 | <Dashiva> | Anyone have a link to the latest ES5 draft? |
| 19:28 | <inimino> | mutating the prototype would work the same as mutating the object |
| 19:28 | <gsnedders_> | Dashiva: http://www.ecma-international.org/publications/files/drafts/tc39-2009-025.pdf |
| 19:28 | <gsnedders_> | Dashiva: Page 85/86, section 12.6.4 |
| 19:28 | <gsnedders_> | Dashiva: Search for "for-in" |
| 19:28 | <Dashiva> | Not mutating the prototype, but replacing the prototype object entirely by mutating the [[Prototype]] |
| 19:29 | <inimino> | Dashiva: that's not even allowed for in ES3 |
| 19:29 | <inimino> | (and I don't think that changed in ES5 though I might be wrong) |
| 19:29 | <Dashiva> | But it's implicitly allowed in ES5 |
| 19:29 | <Dashiva> | Not in the language itself, but it nods to implementations allowing it in extensions |
| 19:30 | <inimino> | oh, right, the catchall extension thing |
| 19:32 | <Dashiva> | Wow, nice change in the wording. :) |
| 19:33 | <ttepasse> | Uhm. Why are there ? as bullet points in the ES5 document? |
| 19:33 | <Dashiva> | ES3 says "the newly added properties are not guaranteed to be visited" and ES5 is "the newly added properties are guaranteed not to be visited" |
| 19:33 | <Dashiva> | Just moved the not :) |
| 19:35 | <Dashiva> | ttepasse: Where? I see bullets here |
| 19:37 | <ttepasse> | Every list: http://tepasse.org/tmp/i-heart-es5 |
| 19:39 | <inimino> | heh |
| 19:40 | <inimino> | it looks fine in evince, probably some kind of font issue |
| 19:40 | <ttepasse> | A font issue in a pdf? |
| 19:41 | <inimino> | sure, PDFs can use fonts without embedding them |
| 19:41 | <inimino> | then if the renderer picks the wrong font, you get something like that |
| 19:42 | <inimino> | it's a shame that ES isn't specified in a first-class Web format |
| 19:42 | <Dashiva> | It's better this way than the wiki mess that preceded it :P |
| 19:43 | <inimino> | the wiki was a mess sure |
| 19:44 | <inimino> | but mostly because the process is opaque and most of the relevant information was hidden at any given time |
| 19:44 | <inimino> | like "oh yeah, that stuff on the wiki is totally obsolete, but we can't show the current status of anything because our process is broken" |
| 19:54 | <gsnedders_> | Oh yay. Nothing matches the spec at all for script@type |
| 19:57 | <gsnedders_> | Dashiva: You still have BTS access? |
| 23:44 | <Darxus> | A noscript element is allowed to be a child of a body element, right? |
| 23:45 | <Darxus> | It's phrasing content, which is flow content, and a body contains flow content? |
| 23:45 | <Darxus> | If this is correct, the validator has a bug: http://html5.validator.nu/?doc=http://www.chaosreigns.com/test/noscript.html |
| 23:50 | <Darxus> | If I'm *not* correct, then my validator has a bug. |