| 00:02 | <dbaron> | Hixie, I think the <ruby> edit you just made is wrong; see my reply to whatwg⊙wo |
| 00:04 | <Hixie> | dbaron: isn't the <rt> a block container? |
| 00:05 | <dbaron> | I sure hope not |
| 00:05 | <Hixie> | huh |
| 00:05 | <dbaron> | ruby elements should all be basically inline |
| 00:05 | <dbaron> | behaving a lot like display:inline |
| 00:05 | <Hixie> | sure but the annotation is its own inline block like thing, no? |
| 00:06 | <Hixie> | i guess we can't really say one way or the other since there's no real spec for this currnetly |
| 00:07 | <dbaron> | it shouldn't support 'width', and it participates in inline layout and breaking for its container, so I think it has to be inline |
| 00:07 | <dbaron> | maybe run it by fantasai, though |
| 00:07 | <Hixie> | fair enough |
| 00:09 | <Hixie> | dbaron: btw you asked a while back for specific notification of changes relating to CSS, my plan is to make CSS a subscribable topic in the system you used to get that notification e-mail |
| 00:09 | <dbaron> | Hixie, ok, sounds great |
| 00:09 | <Hixie> | dbaron: haven't made any CSS feature changes (as opposed to rendering changes, which are already covered by the Rendering topic) since, though, hence it not being done yet |
| 00:16 | <Hixie> | woops |
| 00:16 | Hixie | checked in an unrelated change with that revert |
| 00:19 | <a-ja> | dbaron: is there bug filed for the recent sections/headings additions? |
| 00:20 | <dbaron> | a-ja, eh? |
| 00:20 | <a-ja> | um...lemme find it |
| 00:24 | <a-ja> | dbaron: new stuff in 2nd figure at http://www.whatwg.org/specs/web-apps/current-work/multipage/rendering.html#sections-and-headings |
| 09:13 | <zcorpan> | should we make SVGMatrix available in workers (with a new constructor)? |
| 11:17 | <annevk> | SimonSapin: http://annevankesteren.nl/2013/03/zero |
| 11:31 | <SimonSapin> | annevk, zcorpan: some people apparently freak out as soon as you say "public domain". How about trying and tweaking the W3C license to allow specific use cases we care about? |
| 11:31 | <jgraham> | SimonSapin: We tried that |
| 11:31 | <jgraham> | Or rather |
| 11:31 | <jgraham> | We tried the MIT license |
| 11:32 | <jgraham> | Which IMHO is a fine choice |
| 11:32 | <jgraham> | W3C freaked out |
| 11:32 | <SimonSapin> | jgraham: how did it go? |
| 11:32 | <annevk> | I could live with MIT, but I much prefer CC0 |
| 11:32 | <annevk> | attribution is such a nuisance |
| 11:32 | <annevk> | SimonSapin: just say CC0 |
| 11:33 | <jgraham> | I guess karlcow will complain at me later if I don't point out that the *Members* freaked out |
| 11:34 | <SimonSapin> | annevk: we can try that, but I’m not sure it’ll be much better |
| 11:35 | <jgraham> | I dont' really see why CC0 is better than MIT |
| 11:36 | <jgraham> | But of course I am happy with either |
| 11:36 | <annevk> | jgraham: MIT requires attribution |
| 11:37 | <jgraham> | I don't think requiring attribution is a big deal |
| 11:37 | <SimonSapin> | I wonder if I should make use of being in the AC / representing a Member while I still am, to push this kind of stuff. |
| 11:37 | <SimonSapin> | (I haven’t "announced" it yet but it’s not really secret anymore at this point. I’m leaving Kozea and joining Mozilla in May.) |
| 11:38 | <annevk> | SimonSapin: CC0 addresses the concerns raised on your private list |
| 11:39 | <annevk> | jgraham: I didn't say it was a big deal |
| 11:39 | <SimonSapin> | annevk: I agree CC0 would be great, but given the resistance we’ve seen yesterday I don’t see it happening :/ |
| 11:40 | <annevk> | If you think that's the resistance you haven't really seen much yet :-) |
| 11:40 | <annevk> | Also, their concern is that it's not okay per W3C rules, not that it's not okay per se... |
| 11:42 | <SimonSapin> | well I got private responses too, and I also say this given it’s been tried before |
| 11:43 | <annevk> | Well, "tried"... I don't think the W3C ever gave it a fair shot |
| 11:43 | <annevk> | Some of W3C legal are very much against |
| 11:54 | <SimonSapin> | So, to approach this I’d say: start with practical issues to solve (as you did in today’s blog post, annevk), emphasize this is compatible with the patent policy (at least I hope it is), emphasize this is following the rules in order to try and change them, only then propose CC0 or MIT as a possible solution. |
| 11:54 | <SimonSapin> | The idea is to preemptively defuse responses like "Not gonna happen because XYZ." |
| 11:55 | <Ms2ger> | SimonSapin, welcome already :) |
| 11:57 | <karlcow> | annevk: "Standards should be resuable" |
| 11:57 | Ms2ger | loves re-suing |
| 11:59 | <annevk> | karlcow: fixed |
| 11:59 | <Ms2ger> | On another note |
| 12:00 | <Ms2ger> | darobin++ |
| 12:01 | <annevk> | SimonSapin: sounds good |
| 12:01 | <darobin> | what did I do wrong this time? |
| 12:02 | <Ms2ger> | darobin, the repo move |
| 12:02 | <karlcow> | I wonder how it went http://projects.opensource.org/pipermail/license-review/2012-February/000092.html |
| 12:02 | <darobin> | jgraham: it would be fairer to say that a few members freaked out |
| 12:02 | <karlcow> | long thread http://projects.opensource.org/pipermail/license-review/2012-February/thread.html#92 |
| 12:02 | <darobin> | Ms2ger: oh, that's yesterday's news :) |
| 12:02 | <Ms2ger> | Yeah |
| 12:03 | <jgraham> | darobin: Enough members freaked out |
| 12:03 | <SimonSapin> | http://opensource.org/faq#cc-zero says "not recommend" |
| 12:03 | <Ms2ger> | SimonSapin, do they list the W3C document license? ;) |
| 12:04 | <SimonSapin> | Ms2ger: yes: http://opensource.org/licenses/W3C |
| 12:04 | <SimonSapin> | whatever that means |
| 12:04 | <jgraham> | SimonSapin: Notice it says "software" |
| 12:04 | <SimonSapin> | oh yeah |
| 12:04 | <SimonSapin> | that’s a different license from the Document one, right? |
| 12:04 | <Ms2ger> | Yep |
| 12:05 | <SimonSapin> | looks like OSI is about software? |
| 12:05 | <marcosc> | hmm... |
| 12:05 | <jgraham> | (I actulaly menat in the CC0 FAQ entry, but yeah, that too) |
| 12:05 | <karlcow> | http://www.w3.org/Consortium/Legal/2002/ipr-notice-20021231#Copyright |
| 12:05 | <marcosc> | I didn't know about the fallback |
| 12:05 | <karlcow> | Documents: http://www.w3.org/Consortium/Legal/2002/copyright-documents-20021231 |
| 12:05 | <SimonSapin> | jgraham: indeed |
| 12:05 | <karlcow> | Softwares: http://www.w3.org/Consortium/Legal/2002/copyright-software-20021231 |
| 12:06 | <darobin> | SimonSapin: if you want to reopen discussion on the open license with the AC, I encourage you to |
| 12:06 | <darobin> | and don't hesitate to ping me first to coordinate — there is support to be found |
| 12:06 | <SimonSapin> | darobin: is ac-forum the right place? |
| 12:06 | <Ms2ger> | darobin, oh dear, what has the guy done to you? |
| 12:07 | <darobin> | SimonSapin: yes, but it might be best if we sync up first |
| 12:07 | <SimonSapin> | sure |
| 12:07 | <darobin> | Ms2ger: the Force is strong with this one |
| 12:07 | <Ms2ger> | darobin, strong enough? |
| 12:07 | <SimonSapin> | well, I’m not sure I really want to get into this, but … |
| 12:07 | <darobin> | Ms2ger: find out we must |
| 12:08 | <karlcow> | never give up the fight |
| 12:08 | <Ms2ger> | SimonSapin, http://images.cryhavok.org/d/3385-2/its-dangerous-to-go-alone-take-this.jpg |
| 12:12 | <annevk> | SimonSapin: fwiw, in addition to CC0 the specs disclaim patent stuff to cover the party they don't like |
| 12:12 | <annevk> | s/party/part/ |
| 12:16 | <annevk> | I don't understand one bit of how HTTP auth works. |
| 12:16 | <annevk> | I have this in a .htaccess: |
| 12:16 | <annevk> | RewriteEngine on |
| 12:16 | <annevk> | RewriteRule test.php test.php [E=REMOTE_USER:%{HTTP:Authorization},L] |
| 12:17 | <annevk> | in the same directory I have an HTML file for which I get a 403 when I try to open it |
| 12:17 | <annevk> | how does that even make sense? |
| 12:18 | <darobin> | oh, Apache |
| 12:19 | <darobin> | Apache never makes sense |
| 12:19 | <annevk> | oh permissions |
| 12:20 | <annevk> | my fault this time |
| 12:23 | <SimonSapin> | annevk: sorry, what about patents? |
| 12:24 | <annevk> | zewt: so IE/Chrome/Firefox/Safari prompt if the request originates from XHR |
| 12:25 | <annevk> | zewt: the same happens for <img> |
| 12:26 | <annevk> | SimonSapin: see header of specs |
| 12:27 | <annevk> | even <script> |
| 12:28 | <annevk> | browsers are crazy |
| 12:28 | <Ms2ger> | You must be new here |
| 12:28 | <annevk> | here longer than you, punk |
| 12:31 | <annevk> | I guess the next thing is cross-origin auth |
| 12:47 | <annevk> | Okay, so fetch happens, the user agent evaluates the response, prompts the user synchronously and then either issues a new request including the Authorization header or continues fetching? |
| 12:47 | <annevk> | shit is weird |
| 12:48 | <annevk> | I wonder if you post a large blob if it's posted twice |
| 12:48 | <annevk> | prolly is |
| 13:33 | <zewt> | annevk: post requests by definition are never supposed to be sent twice, right? |
| 13:34 | <annevk> | zewt: if the response is a 307 I don't think that's true |
| 13:35 | <zewt> | that's what 100 Continue is for, iirc |
| 13:35 | <annevk> | no |
| 13:35 | <zewt> | been a while since i've messed with that though |
| 13:35 | <annevk> | I think that rule only applies if the server responded with a 2xx code |
| 13:37 | <zewt> | i guess the client has to explicitly say that it wants a 100 (the Expect header), and since you don't know in advance if there will be an auth error, nobody will want to use it (since it adds a round-trip delay) |
| 13:38 | <zewt> | confused at people talking about "Future" on the event handler thread as if it's a thing that exists |
| 13:38 | zcorpan | learns about step_func_done |
| 13:39 | <Ms2ger> | zewt, it will in the future, apparently |
| 13:39 | Ms2ger | ducks |
| 13:39 | <zewt> | hopefully that's recursive and it'll always be in the future |
| 13:39 | <zewt> | because it's so ugly and out of place and nothing like any API on the platform |
| 13:41 | <zewt> | annevk: fwiw, the user only needs to be prompted synchronously if the fetch is synchronous ... i wonder if browsers do that prompting if the fetch is in a worker |
| 13:44 | <annevk> | hmm, if XHR defines a timeout, will the prompt just go away? |
| 13:44 | <zcorpan> | MikeSmith: something broke? https://dvcs.w3.org/hg/webapps/rev/747f5eb0c352 |
| 13:44 | <zewt> | worse, in a shared worker, which isn't associated with a specific window (or, once shared workers support cross-origin use, even the same origin) |
| 13:44 | <annevk> | yeah if there's no browsing context HTTP auth should just not work |
| 13:45 | <zewt> | that sounds implementation-specific (re timeout) |
| 13:45 | <annevk> | of course that's not actually defined and importScript might very well do something |
| 13:45 | <annevk> | it shouldn't be |
| 13:45 | <annevk> | wow, HTTP auth is so full of hurt |
| 13:45 | <zewt> | whether the UI is closed should be (what happens if they do decide to do that shouldn't be) |
| 13:45 | <zcorpan> | why is importScripts under "also has obsolete members" in the spec? |
| 13:46 | <annevk> | zcorpan: is it? |
| 13:46 | <zcorpan> | i got 500. but now it seems to work |
| 13:47 | <zewt> | doesn't seem to be |
| 13:47 | <zcorpan> | except my last change doesn't seem to be synced on w3c-test.org. http://w3c-test.org/webapps/Workers/tests/submissions/Opera/constructors/Worker/ctor-undefined.html should have one test |
| 13:47 | <annevk> | zcorpan: oh it is, I think Hixie introduced that as a mistake |
| 13:47 | <annevk> | zcorpan: by mistake* |
| 13:47 | <annevk> | zcorpan: when he changed it into a partial interface |
| 13:47 | <zcorpan> | annevk: is it known or should i file a bug? |
| 13:47 | <annevk> | zcorpan: file |
| 13:48 | <zewt> | where's that? the idl in 9.3 doesn't say that |
| 13:48 | <zcorpan> | http://www.whatwg.org/specs/web-apps/current-work/multipage/workers.html#workerglobalscope does |
| 13:49 | <zewt> | oh that |
| 13:52 | <zcorpan> | annevk: does "unsupported:" as a url resolve or not per spec? |
| 13:55 | <zcorpan> | annevk: scheme data state doesn't say what to do when it gets EOF, i think |
| 13:56 | <annevk> | zcorpan: it avoids handling EOF and therefore the loop ends at the next turn |
| 13:56 | <annevk> | zcorpan: seems covered |
| 13:56 | <zcorpan> | ok |
| 14:03 | <zcorpan> | ironic that a non-breaking space breaks the test (when interpreted with the wrong encoding) |
| 14:13 | <darobin> | if something goes to the trouble of calling itself "non-breaking" you know it's up to no good |
| 14:20 | zcorpan | made option+space insert a space instead of non-breaking space with keyremap4macbook |
| 14:25 | <Ms2ger> | annevk, HTML doesn't use MIT, does it? |
| 14:26 | <annevk> | ta |
| 14:31 | <zewt> | zcorpan: usually the way i end up with weird literal nbsps in code is due to copying from mdn, heh |
| 14:31 | <zewt> | browsers really should fold those to regular spaces on copy |
| 14:31 | <zcorpan> | i thought they did |
| 14:32 | <zewt> | maybe some do, at least one of firefox and chrome doesn't though |
| 14:33 | <zewt> | speaking of which, i should see if there's a firefox option to disable copying as html, that bites me constantly when pasting into gmail... |
| 14:36 | <zewt> | i guess this is one of those ungooglable things, since everything related to disabling copying html is people asking how to restrict users from copying, heh |
| 14:44 | <zcorpan> | can't you change gmail to plain text only? |
| 14:44 | <zewt> | i don't want plain text only, i just don't want weird formatting every time i paste text, heh |
| 14:45 | <zewt> | apparently control-shift-v in firefox pastes without formatting (would prefer to just never have it copied in the first place, heh) |
| 14:50 | <zcorpan> | http://w3c-test.org/webapps/Workers/tests/submissions/Opera/interfaces/WorkerUtils/importScripts/004.html says e is not defined. i don't understand why not |
| 14:51 | <Ms2ger> | } catch(ex) { |
| 14:51 | <Ms2ger> | if (e instanceof SyntaxError) |
| 14:51 | <Ms2ger> | This one? |
| 14:51 | <jgraham> | That's what I was going to say :) |
| 14:52 | <annevk> | hmm so cross-origin XHR actually does something else than <img> or <script> for HTTP auth |
| 14:52 | <annevk> | I wish I'd seen the implications back then |
| 14:52 | <annevk> | well actually, mode = CORS just means no HTTP auth |
| 14:53 | <annevk> | well, unless the request is same-origin to begin with of course |
| 14:53 | <zcorpan> | ah, thanks |
| 15:07 | <MikeSmith> | zcorpan: looking at that link now |
| 15:08 | <MikeSmith> | not seeing any breakage |
| 15:08 | <MikeSmith> | maybe darobin restarted the server already? |
| 15:08 | <zcorpan> | MikeSmith: i get a 500 response sometimes when pushing |
| 15:08 | <MikeSmith> | ok |
| 15:08 | <zcorpan> | but it goes through anyway so maybe it's fine |
| 15:08 | <MikeSmith> | I'll just go ahead and restart the server |
| 15:09 | <SimonSapin> | \o/ First publication! http://www.w3.org/TR/2013/WD-css3-page-20130314/ |
| 15:10 | <MikeSmith> | nice |
| 15:10 | <MikeSmith> | SimonSapin: congrats |
| 15:10 | <darobin> | huwah? |
| 15:10 | <darobin> | oh, dvcs |
| 15:10 | <darobin> | MikeSmith: go back to vacation man! |
| 15:11 | <MikeSmith> | SimonSapin: oh I see you've aligned yourself with the Haakon paged-media faction. Good man. |
| 15:11 | <darobin> | wait, why is zcorpan putting tests in dvcs? |
| 15:11 | <darobin> | zcorpan! go put your tests in the shiny new github repo for everyone! |
| 15:11 | <zcorpan> | darobin: i thought webapps was still using it |
| 15:12 | <SimonSapin> | MikeSmith: faction? |
| 15:12 | <MikeSmith> | darobin: I will! morning here and my girlfriend is getting ready so I got a short window of opportunity for following up on IRC pings before we head out for the day |
| 15:12 | <darobin> | doesn't webapps do whatever we're doing? |
| 15:12 | <MikeSmith> | SimonSapin: Haakon vs the party leaders |
| 15:12 | <darobin> | MikeSmith: ok, say hi to her then :) |
| 15:13 | <darobin> | zcorpan: let me go bug Art about that |
| 15:13 | <MikeSmith> | SimonSapin: just joking. Was alluding to the exchange between Haakon on glazou and Peter a couple weeks back |
| 15:13 | <MikeSmith> | darobin: will do |
| 15:14 | <darobin> | "the exchange" |
| 15:14 | <SimonSapin> | MikeSmith: well, Håkon is officially an editor but he has not touched css3-page since before I knew what Working Group is. As to GCPM, I think most of it should be dropped and the rest needs a lot of work |
| 15:14 | <darobin> | (of variegated ordnance) |
| 15:24 | <zcorpan> | http://w3c-test.org/webapps/Workers/tests/submissions/Opera/semantics/navigation/001.html fails in opera but the old version passes. :-/ |
| 15:27 | <jgraham> | Do you have the arguments to assert_greater_than the right way around? |
| 15:27 | <annevk> | FWIW, all browsers prompt for importScripts running into HTTP auth |
| 15:27 | <annevk> | or new Worker() running into HTTP auth |
| 15:29 | <zcorpan> | jgraham: yes |
| 15:30 | <annevk> | SharedWorker() prompts too in Chrome |
| 15:30 | <annevk> | and Safari |
| 15:33 | <zcorpan> | hmm. increasing the inner timeout makes it pass |
| 15:37 | <annevk> | hmm, if you set www-authenticate as a header, does PHP really do magic to override the status code? |
| 15:38 | <annevk> | man |
| 15:42 | <jgraham> | Well |
| 15:42 | <zcorpan> | yay, workers done |
| 15:42 | <jgraham> | You are using PHP |
| 15:42 | <jgraham> | zcorpan: Nice |
| 15:43 | <annevk> | jgraham: sure, but this kind of low-level stuff is equally annoying in any language |
| 15:43 | <annevk> | jgraham: maybe not node.js so much, but that doesn't run by default on DreamHost |
| 15:44 | <jgraham> | annevk: Hmm? Other langauges typically don't have HTTP baked in at the language layer |
| 15:46 | <annevk> | there's a way to override it btw using header(headernameandvalue, true, status) |
| 15:48 | <annevk> | jgraham: Python has the same thing if you do the simple CGI thing |
| 15:58 | <gsnedders> | annevk: For HTTP stuff I basically concluded the only sane way was to have a custom server that sent exactly the response you gave it. |
| 15:58 | <gsnedders> | Response line and all. |
| 15:58 | <jgraham> | You could use mod_asis and apache ont he w3c server, perhaps |
| 16:00 | <zcorpan> | annevk: i'm having trouble finding a cssom commit from you that has cc0 |
| 16:01 | <annevk> | zcorpan: I suspect my stuff is in http://dev.w3.org/cvsweb/csswg/cssom/Attic/ |
| 16:02 | <annevk> | gsnedders: yeah, pretty much, though I think given what the inspectors gave me think I got away with PHP this time |
| 16:04 | <zcorpan> | annevk: still... |
| 16:05 | <darobin> | annevk: you can run node on heroku and other things like that |
| 16:05 | <darobin> | it's a smart thing actually that php sets the status if you use an auth header, unless you're writing tests it's what you want |
| 16:05 | <darobin> | but for testing, you need to work on the metal, so node would certainly be a better choice imho |
| 16:06 | <darobin> | (or something like it) |
| 16:06 | <Ms2ger> | http://25.media.tumblr.com/14b031162e22a6f909bd43337e79260c/tumblr_mjm4rqmxOv1rmq849o1_400.jpg |
| 16:06 | <Ms2ger> | Neat. |
| 16:07 | <annevk> | zcorpan: yeah dunno |
| 16:12 | <annevk> | zcorpan: if glenn removed it, it must have been in there somehow |
| 16:12 | <Ms2ger> | Flawless logic |
| 16:12 | <annevk> | did Anolis generate it maybe? |
| 16:12 | <Ms2ger> | I doubt it |
| 16:13 | <zcorpan> | annevk: https://dvcs.w3.org/hg/cssom/rev/74784effb4e3 |
| 16:13 | <annevk> | so Glenn put it under CC0? |
| 16:13 | <zcorpan> | looks like it |
| 16:14 | <annevk> | wfm |
| 16:14 | <zcorpan> | but doing so isn't allowed by the document license |
| 16:15 | <zcorpan> | so there isn't anything "clean" to start from, i think |
| 16:23 | <dglazkov> | good morning, Whatwg! |
| 16:39 | <SteveF> | dglazkov: forgotten the stuff i wrote about web components + ARIA thanks for the reminder :-) |
| 17:17 | <annevk> | HTTP Authentication: not secure; imposes bogus UI; makes the fetching model complicated |
| 17:17 | <annevk> | ... |
| 17:18 | <annevk> | hahaha |
| 17:18 | <annevk> | for Digest they recommend that the realm includes the host requesting the authentication |
| 17:24 | <dglazkov> | SteveF: np |
| 17:26 | <annevk> | now whenever someone says HTTP Authentication I'll just laugh for a while |
| 17:59 | <Hixie> | annevk: is the "URL standard: Query string parsing; host parsing" thread done, or do you still have to respond to it? |
| 18:33 | jwalden | sees this E4H thread and joins the popcorn-eaters |
| 18:36 | <jwalden> | tc39 sometimes has a tendency to go for big large solutions over smaller ones, I think |
| 18:36 | <jwalden> | e.g. es4 |
| 18:37 | <hober> | abarth is stopping by tc39 today to have this argument in person |
| 18:37 | <hober> | sadly there's no popcorn here |
| 18:44 | <jwalden> | :-) |
| 18:44 | <zcorpan> | TabAtkins: thanks for the support |
| 18:45 | <jwalden> | aside from E4X being a disaster of a spec, and in a few cases doing too much, mostly I think it did about the amount that's desirable |
| 18:45 | <jwalden> | trying to solve everything -- URL syntax checking, event handler checking, arbitrary format checking -- seems a bit much to me |
| 18:45 | zcorpan | wonders why https://twitter.com/w3cmemes/status/312259499224354816 |
| 18:46 | <jwalden> | which is not to say that I like or dislike E4H, haven't looked, although I'm likely super-leery of anything here given previous third-degree burn experience with E4X |
| 18:47 | <hober> | zcorpan: I imagine that is due to http://www.w3.org/QA/2013/03/drm_and_the_open_web.html |
| 19:10 | <TabAtkins> | zcorpan: I've been planning to do my own specs co-published for a while now, but haven't gotten around to it. I'll actually do so now. ^_^ |
| 19:11 | <zcorpan> | TabAtkins: nice |
| 19:13 | <TabAtkins> | jwalden: Agree. Hixie's idea of e4h more or less works, and is really simple. |
| 19:17 | <Ms2ger> | zcorpan, ArtB requesting your attention in W3C/#testing |
| 19:28 | <zcorpan> | cssom is still generated using anolis |
| 19:28 | <zcorpan> | glenn just added a preprocessor and changed the source format to be idl with html in it instead of html with idl in it |
| 20:11 | <hober> | abarth steps in front of the firing squad, err, technical committee |
| 23:57 | <zewt> | wish i could say <img src=sha1.jpg filename=family.jpg> to hint browsers to save-as with a different filename (without having to fight with servers to set content-disposition) |
| 23:57 | <zewt> | but i couldn't even get people to believe i wanted that for <a> |
