| 00:04 | <timeless> | MikeSmith: still here? |
| 00:04 | <timeless> | the message where he wrote "to text/plain" was when he stopped using rich text email messages |
| 00:05 | <timeless> | i'm not sure what mail client you're using |
| 00:05 | <timeless> | the archives seem to only show the plain text version of messages (kinda good, somewhat problematic when referencing rich versions) |
| 00:08 | <MikeSmith> | ah |
| 00:08 | <MikeSmith> | I see |
| 00:09 | <MikeSmith> | I don't read the archives, so didn't see the context |
| 00:09 | <MikeSmith> | and I read all my mail in mutt, so it's all plain text to me anyway :) |
| 00:14 | MikeSmith | finishes reading scrollback |
| 00:15 | <MikeSmith> | Hixie: yeah, wasn't expecting that you'd touch the ones not assigned to you |
| 00:16 | <MikeSmith> | thanks for checking those |
| 00:16 | <MikeSmith> | oh |
| 00:17 | <MikeSmith> | I guess you haven't actually gotten to them yet |
| 00:17 | <MikeSmith> | so, then, thanks in advance :) |
| 00:19 | <timeless> | MikeSmith: do you really want me to reply to www-archive? |
| 00:20 | <MikeSmith> | timeless: no |
| 00:20 | <MikeSmith> | no need |
| 00:20 | <MikeSmith> | it's clear now |
| 00:20 | <timeless> | well, it's clear |
| 00:20 | <timeless> | but i still need someone to smack the rest of the googlers |
| 00:20 | <timeless> | can you do that for me? |
| 00:20 | <timeless> | please? |
| 00:20 | <MikeSmith> | nope |
| 00:20 | <timeless> | (that includes the chromium.org one) |
| 00:20 | <MikeSmith> | because I think the solution to that problem is for you and everybody else to read your mail in better mail client |
| 00:21 | <MikeSmith> | as I do |
| 00:21 | <MikeSmith> | it's like, if there's somebody annoying with you in the same room, and you close your eyes, that annoying person goes away |
| 00:23 | timeless | sighs |
| 00:23 | <timeless> | unhelpful |
| 00:24 | <MikeSmith> | well, seriously, I wish people would not send mail that way too |
| 00:24 | <MikeSmith> | but I do not want to become the netiquette police |
| 00:24 | <MikeSmith> | I think the message you sent to the list was great |
| 00:25 | <timeless> | yeah, i tried to provide a nice set of examples |
| 00:25 | <timeless> | sadly i screwed up and message 4 went to the wrong address and thus appeared out of order |
| 00:25 | <MikeSmith> | oh |
| 00:25 | <timeless> | i really did send it before 5 |
| 00:26 | <MikeSmith> | anyway, that message of yours of course has a URL now so people on the list can point others to it when needed |
| 00:26 | <MikeSmith> | dude |
| 00:26 | <timeless> | yeah, i've already used it at least once |
| 00:26 | <timeless> | i still need to get used to that MID Thing |
| 00:26 | <timeless> | is it possible to have the mailing list web ui software provide MID links? |
| 00:26 | <MikeSmith> | you seem to be excessively worried about some of this kind of stuff lately |
| 00:27 | <MikeSmith> | timeless: yeah it's possible of course |
| 00:27 | <timeless> | could you ask the systeam for me? |
| 00:27 | timeless | wonders if that's the right team |
| 00:27 | <timeless> | oh, and whom was i supposed to send my magic script that figure out speakers and such? was that sys also? |
| 00:27 | <MikeSmith> | would be better if you asked yourself. they will listen to that more than me requesting it |
| 00:27 | <MikeSmith> | yeah |
| 00:28 | <timeless> | ok, mailto:sysreq@w3 ? |
| 00:28 | <MikeSmith> | yep |
| 00:28 | <MikeSmith> | but about the order you sent your messages in, I don't think anybody else would find that confusing and many probably would not even notice |
| 00:29 | <MikeSmith> | I think you need to try to do some things more randomly in your life |
| 00:29 | <MikeSmith> | make some arbitrary decisions |
| 00:29 | <MikeSmith> | or flip a coin |
| 00:29 | <MikeSmith> | or sometimes do the opposite of what you'd normally be inclined to do |
| 00:29 | <MikeSmith> | now and then |
| 00:34 | <AryehGregor> | I am entirely unsurprised that that's a course of action you would endorse. |
| 00:34 | <rniwa> | AryehGregor: hi |
| 00:34 | <AryehGregor> | rniwa, hi. |
| 00:35 | <rniwa> | AryehGregor: did you see the webkit bug I just cc-ed you on? |
| 00:35 | <AryehGregor> | rniwa, it's in my work inbox. I'll look at it tomorrow. |
| 00:35 | <rniwa> | AryehGregor: ok |
| 00:36 | <rniwa> | AryehGregor: I think your editing API spec should match the behavior |
| 00:36 | <AryehGregor> | I'll get back to you on that. |
| 00:36 | <rniwa> | k |
| 00:44 | <MikeSmith> | rniwa: please be nice to the command api |
| 00:44 | <MikeSmith> | he needs your help |
| 00:44 | <rniwa> | MikeSmith: yeah |
| 00:45 | <rniwa> | MikeSmith: I want them |
| 00:45 | <rniwa> | MikeSmith: it's just that I don't feel like there have been enough discussions |
| 00:45 | <MikeSmith> | yeah, understood |
| 00:45 | <rniwa> | MikeSmith: e.g. most of webkit developers, etc... have never heard of command apis :( |
| 00:45 | <MikeSmith> | definitely needs some more attention |
| 00:45 | <rniwa> | yeah |
| 00:45 | <rniwa> | MikeSmith: I think they have a lot of good ideas |
| 00:45 | <rniwa> | MikeSmith: just needs some polishing |
| 00:46 | <rniwa> | MikeSmith: and to do that, it's important to address use cases, etc... |
| 00:46 | <MikeSmith> | yup |
| 00:46 | <rniwa> | I know some people (including myself) have a tendency to implement whatever spec says without doing the sanity check of the spec itself |
| 00:47 | <MikeSmith> | hmm, yeah, that's definitely not the way it should work |
| 00:47 | <rniwa> | MikeSmith: so there's a very interesting implication on accessibilty |
| 00:47 | <rniwa> | MikeSmith: good one actually |
| 00:47 | <MikeSmith> | yeah? |
| 00:47 | <MikeSmith> | how so? |
| 00:47 | <rniwa> | MikeSmith: if we expose "commands" |
| 00:47 | <rniwa> | MikeSmith: then it could be listed in context menu, etc... |
| 00:47 | <MikeSmith> | yeah |
| 00:47 | <rniwa> | MikeSmith: and assistant technology can provide a higher-level semantics of what an app does |
| 00:48 | <rniwa> | which is a superb solution to listing buttons and letting people figure out what those buttons do |
| 00:48 | <MikeSmith> | so the spec already does some of that by tying it in with accesskey |
| 00:48 | <rniwa> | also, you could imagine that if command has some human readable level such as "create a new note" |
| 00:49 | <rniwa> | then voice recognition software might be able to automatically execute that command |
| 00:49 | <rniwa> | when the user instructs it to do so |
| 00:49 | <rniwa> | MikeSmith: yeah, typing it to accessKey is nice |
| 00:50 | <rniwa> | MikeSmith: but we aren't so happy about how command's states are tied with the element that defines it |
| 00:50 | <MikeSmith> | yeah, there's a lot of utility that could be had from this feature, so it would be nice to see it getting some more scrutiny and refinement |
| 00:50 | <MikeSmith> | ah |
| 00:50 | <MikeSmith> | yeah |
| 00:50 | <rniwa> | MikeSmith: e.g. we had an idea to tie the command with execCommand's commands |
| 00:50 | <rniwa> | MikeSmith: and with tthat |
| 00:50 | <rniwa> | MikeSmith: you can also tie it to radio box, checkbox, etc... states |
| 00:51 | <MikeSmith> | OK |
| 00:51 | <rniwa> | MikeSmith: for example, you can imagine that if you have <input type="checkbox" command="bold"> |
| 00:51 | <rniwa> | MikeSmith: then UA can automatically check this box when the command state is bold |
| 00:51 | <rniwa> | I mean when bold command's state is true |
| 00:51 | <MikeSmith> | yeah |
| 00:51 | <rniwa> | and uncheck when the state is false |
| 00:52 | <rniwa> | MikeSmith: we can apply similar mechanism to select, radio, etc... |
| 00:52 | <rniwa> | MikeSmith: from that perspective, the current design of ting the command state to the element |
| 00:52 | <rniwa> | and furthermore having inputelement-like checked state isnt' ideal |
| 00:53 | <rniwa> | because you may want to have multiple UI widgets referring to the same command |
| 00:53 | <rniwa> | e.g. <select command="color"><option>blue<option>green |
| 00:53 | <rniwa> | < |
| 00:53 | <rniwa> | </select> |
| 00:54 | <rniwa> | in this case, you want blue or green to be selected when color command's state is blue or green |
| 00:54 | <rniwa> | but in addition to this select element |
| 00:54 | <rniwa> | the same app may have another UI widget that has <input type="color"> |
| 00:54 | <rniwa> | MikeSmith: and this one should automatically reflect whatever the current command value is |
| 00:55 | <MikeSmith> | OK |
| 00:55 | <rniwa> | MikeSmith: furthermore, we probably want to let authors to define and implement their own custom UI component |
| 00:55 | <rniwa> | MikeSmith: and those component should probably be able to be notified of command state's change |
| 00:55 | <rniwa> | MikeSmith: so we need some sort of registry mechanism |
| 00:55 | <rniwa> | MikeSmith: also we need to be able to scope commands |
| 00:56 | <rniwa> | e.g. many websites include WYSIWYG editors but commands available inside a editor are vastly different from those available outside the editor |
| 00:57 | <rniwa> | MikeSmith: but there's a lot of "data-binding" aspect here though |
| 00:57 | <MikeSmith> | yeah, so that gets into the components stuff |
| 00:57 | <rniwa> | right |
| 00:58 | <MikeSmith> | and maybe it's good to wait on the command things until components is further along |
| 00:58 | <rniwa> | MikeSmith: could be |
| 00:58 | <rniwa> | MikeSmith: but I think we can start the discussion |
| 00:58 | <rniwa> | MikeSmith: and get more attentions |
| 00:58 | <MikeSmith> | yeah |
| 00:58 | <rniwa> | MikeSmith: because we definitely need a good API to define toolbars, menus, and semantically connect those components |
| 00:59 | <rniwa> | MikeSmith: I was hoping to get the discussion started by now but it seems like I've caught up by other stuff :( |
| 00:59 | <rniwa> | so maybe I'll do that in Jan |
| 00:59 | <MikeSmith> | yeah |
| 01:00 | <MikeSmith> | I think we should also plan to have another face to face meeting in the valley next spring |
| 01:00 | <rniwa> | MikeSmith: but this is definitely something I'll be interested in |
| 01:00 | <rniwa> | MikeSmith: oh that'll be nice :) |
| 01:00 | <rniwa> | though i'm sure google doesn't mind flying me over to anywhere if we can be productive like we were in this TPAC |
| 01:01 | <MikeSmith> | yeah, we need to do it for sure |
| 01:08 | <rniwa> | MikeSmith: anyway, I'm planning to work on command API in the coming months |
| 01:09 | <rniwa> | MikeSmith: it's just that i've been swamped by other work |
| 01:09 | <rniwa> | also I really want to get my undomanager ready for early adaptation by the end of the year |
| 01:19 | <zewt> | heh |
| 01:19 | <zewt> | gui text editors with broken undo are so much worse than a simple textbox editor that i almost don't know why gmail bothers |
| 01:20 | <zewt> | if i undo one more time in gmail, and have it revert a change a mile away without showing me what... |
| 01:20 | <MikeSmith> | yeah, please |
| 01:20 | <zewt> | then i have to redo, and (since it still doesn't show what changed) i just have to cross my fingers and hope |
| 02:19 | <MikeSmith> | so David Flanagan has his JS-based HTML parser working with node.js now |
| 02:19 | <MikeSmith> | https://twitter.com/#!/__DavidFlanagan/status/139512471949017088 |
| 02:19 | <MikeSmith> | https://github.com/andreasgal/dom.js |
| 10:22 | <annevk> | MikeSmith: can I share your Google+ post on the WHATWG page? |
| 10:22 | <MikeSmith> | sure |
| 10:22 | <MikeSmith> | of course |
| 10:22 | <annevk> | MikeSmith: it's currently semi-private, but it seems just public info |
| 10:22 | <MikeSmith> | oh |
| 10:23 | <MikeSmith> | yeah, pretty much the only things I make private like that are when I don't think it's necessarily of interest to everybody in my circles |
| 10:23 | <MikeSmith> | so I have a "interested in browser stuff" circle that I post that too |
| 10:23 | <MikeSmith> | shit |
| 10:23 | <annevk> | hmm it seems I cannot share it |
| 10:23 | <MikeSmith> | big earthquake somewhere |
| 10:23 | <annevk> | I guess WHATWG is not in that circle |
| 10:23 | <annevk> | oh |
| 10:23 | <annevk> | :( |
| 10:24 | <MikeSmith> | lemme add it |
| 10:24 | <MikeSmith> | I thought it was |
| 10:24 | <MikeSmith> | hokkaido |
| 10:24 | <MikeSmith> | http://quake.twiple.jp/quake/view/20111124192542 |
| 10:25 | annevk | I have a workaround I think |
| 10:25 | <MikeSmith> | hmm, I don't know how I can change a post to public after I posted it.. |
| 10:26 | <annevk> | no workaround did not work |
| 10:26 | <MikeSmith> | ah, I see |
| 10:26 | <annevk> | I shared it with +WHATWG |
| 10:26 | <annevk> | but then +WHATWG cannot share it with the public |
| 10:27 | <MikeSmith> | hmm, I still don't see how I can change the original post to being public |
| 10:27 | <MikeSmith> | I guess I can't |
| 10:27 | <Workshiva> | You can't |
| 10:28 | <annevk> | it's all lost :) |
| 10:30 | <annevk> | MikeSmith: where do you normally put redirects for specs? |
| 10:30 | <annevk> | MikeSmith: in their parent directory? |
| 10:30 | <jgraham> | In which we discover the fundamental limitation of circles. They are great for solving the "I only want to share drunken photos with my friends" problem but totally useless for the "I publish on multiple topics but the audience for each topic is not the same" |
| 10:30 | <MikeSmith> | yeah, usually |
| 10:30 | <MikeSmith> | jgraham: yup |
| 10:31 | <annevk> | MikeSmith: might be easier for you actually |
| 10:31 | <MikeSmith> | OK |
| 10:32 | <MikeSmith> | XHR spec? |
| 10:32 | <annevk> | to redirect http://dev.w3.org/2006/webapi/XMLHttpRequest-2/ and http://dev.w3.org/2006/webapi/XMLHttpRequest/ to http://dvcs.w3.org/hg/xhr/raw-file/tip/Overview.html |
| 10:32 | <annevk> | yup |
| 10:38 | <Workshiva> | jgraham: Or rather, circles solve sender ACLs, not receiver filtering |
| 10:39 | <MikeSmith> | annevk: OK, done |
| 10:40 | <jgraham> | Workshiva: If you want to be all fancy about it :p |
| 10:42 | <Workshiva> | Maybe one day you'll be able to filter by hashtags |
| 11:00 | <hsivonen> | yay. HTML in XHR broke Wolfram Alpha |
| 11:02 | <hsivonen> | alos, having the HTML parser run uselessly on Gmail is not good |
| 11:02 | <hsivonen> | I'm leaning more and more towards supporting HTML only if responseType == "document" |
| 11:02 | <hsivonen> | to avoid opting all legacy in |
| 11:11 | <MikeSmith> | hsivonen: in what way does it run uselessly on Gmail? |
| 11:11 | <hsivonen> | MikeSmith: Gmail does XHR to a text/html resource and has been written before HTML in XHR existed |
| 11:12 | <annevk> | http://dvcs.w3.org/hg/xhr/raw-file/tip/Overview.html#specification-history |
| 11:12 | <hsivonen> | MikeSmith: so either they have written code that knows how to handle HTML in XHR in advance |
| 11:13 | <MikeSmith> | ah |
| 11:13 | <hsivonen> | MikeSmith: or HTML parsing happens uselessly and they use responseText anyway |
| 11:13 | <annevk> | lets make it "document" only |
| 11:13 | <MikeSmith> | I see |
| 11:13 | <hsivonen> | annevk: sold! |
| 11:14 | <annevk> | that way we don't need the weird async difference either |
| 11:14 | <smaug____> | "document" only sounds ok to me |
| 11:16 | <jgraham> | Oooh a bandwagon! I'll jump on! |
| 11:18 | <hsivonen> | smaug____: is it intentional that we don't throw for "document" in the sync mode yet? |
| 11:19 | <smaug____> | hsivonen: hmm, is there some patch waiting for landing.. |
| 11:21 | <smaug____> | seems like so |
| 11:30 | <annevk> | MikeSmith: can you create a notifications repo that jgraham can access? |
| 11:30 | <MikeSmith> | sure |
| 11:30 | <annevk> | MikeSmith: "notifications" as a name seems fine |
| 11:31 | <annevk> | sweet |
| 11:43 | <MikeSmith> | http://dvcs.w3.org/hg/notifications/ |
| 11:43 | <MikeSmith> | annevk: ↑ |
| 11:43 | <MikeSmith> | with that I step out to get some food |
| 11:43 | <annevk> | thanks man |
| 11:46 | <jgraham> | MikeSmith: Cool, thanks |
| 14:34 | <annevk> | hsivonen: you around? |
| 14:35 | <annevk> | hsivonen: I wondered whether for http://dvcs.w3.org/hg/xhr/raw-file/tip/Overview.html#text-response-entity-body the only part that should be restricted to legacy is the XML handling |
| 14:44 | <annevk> | I will read your emails again instead of this lazy approach |
| 14:55 | <hsivonen> | annevk: I think step #4 should only apply for responseType == "" |
| 14:55 | <hsivonen> | annevk: I think it shouldn't apply to responseType == "text" |
| 14:55 | <hsivonen> | or chunked-text |
| 14:55 | <annevk> | right |
| 14:55 | <annevk> | I just made that change |
| 14:55 | <annevk> | (not pushed yet) |
| 14:55 | <hsivonen> | ok |
| 14:56 | <annevk> | now for document I will change that text/html only works for "document" |
| 14:56 | <hsivonen> | in other news, detecting HTML in XHR support is complicated: https://developer.mozilla.org/En/HTML_in_XMLHttpRequest#Feature_Detection |
| 14:56 | <hsivonen> | annevk: thanks |
| 14:56 | <annevk> | and then update the bug filed on HTML |
| 14:56 | <annevk> | or can you update that bug? |
| 14:56 | <annevk> | you probably know better what the HTML spec has to say |
| 14:57 | <hsivonen> | ok |
| 15:01 | <annevk> | I was afraid these changes would be much harder |
| 15:03 | <annevk> | hsivonen: the spec is updated, I just need add the more specific invocation for the HTML parser spec |
| 15:05 | <hsivonen> | I commented on the bug |
| 15:05 | <hsivonen> | annevk: thank you |
| 15:05 | <annevk> | thank you too! |
| 15:05 | <annevk> | :) |
| 15:06 | <annevk> | hsivonen: is the parser run with scripting disabled? |
| 15:07 | <hsivonen> | annevk: yes (as required by the spec draft I read!) |
| 15:07 | <annevk> | hsivonen: okay, I'll add that to the bug to make sure hixie considers that as an option |
| 15:09 | <annevk> | hsivonen: I sometimes wonder whether it should be "act as if scripting was on but do not actually run scripts" so you do not get unexpected <noscript> stuff, but I think it does not matter much and the current solution is arguably cleaner |
| 15:12 | <timeless> | hsivonen: you need to fix your page |
| 15:12 | <timeless> | it talks about XHR2 |
| 15:12 | <timeless> | but annevk just spent time killing XHR2 :) |
| 15:16 | <annevk> | ah sorry about that |
| 15:16 | <annevk> | XHR will not have version numbers anymore |
| 15:16 | <annevk> | if it's up to me |
| 15:17 | <timeless> | annevk: you didn't do anything wrong, hsivonen just needs to fix it |
| 15:17 | <timeless> | although, it's a wiki, you could do it if you like :) |
| 15:17 | timeless | objects to wikis, they require creating more passwords which is annoying |
| 15:23 | <hsivonen> | timeless: good point |
| 15:23 | <hsivonen> | Living XHR |
| 15:24 | hsivonen | lands some XHR code before fixing the docs |
| 15:44 | <hsivonen> | timeless: I fixed the documentation |
| 15:45 | <timeless> | kinda |
| 15:45 | <timeless> | it should say "the X dated version" adds |
| 15:45 | <timeless> | otherwise it sounds tautological |
| 15:45 | <timeless> | The XMLHttpRequest specification adds HTML parsing support to XMLHttpRequest. |
| 15:46 | <timeless> | it doesn't actually need to says <Dated version>, but you roughly need to indicate that this happened during its evolution instead of just saying it added it to what seems like itself |
| 15:47 | <timeless> | and giving a date is moderately helpful for people trying to figure out if some browser not covered by a table at the bottom has a prayer of supporting it |
| 16:00 | <hsivonen> | timeless: I made the first sentence make more sense. Beyond that, it's a wiki. :-) |
| 16:06 | <AryehGregor> | Google Docs' selection handling is unbelievably messed up. |
| 16:06 | <AryehGregor> | Like: position cursor at beginning of line. Shift+Down Arrow. Ctrl-X. Move to beginning of another line. Ctrl-V. |
| 16:06 | <AryehGregor> | Result? It prepends it to the second line instead of adding a new line before the current line. |
| 16:07 | <AryehGregor> | Same as if you did Shift+End instead of Shift+Down Arrow. |
| 16:07 | <AryehGregor> | And with RTL, arrows move in logical direction, not visual. So the right arrow moves left and vice versa. Which frankly might make some sense, but it's not what any other program in the universe does. |
| 16:08 | <AryehGregor> | If you're going to reinvent the wheel, is it too much to ask that you do it properly? How is it that GNOME people are capable of doing all this stuff basically as people expect but Google isn't? |
| 16:08 | AryehGregor | grumbles |
| 16:10 | <annevk> | I omitted the SotD in the editor's draft: http://dvcs.w3.org/hg/xhr/raw-file/tip/Overview.html |
| 16:10 | <annevk> | and included "Participate:" similar to what WHATWG HTML uses |
| 16:11 | <timeless> | hsivonen: alright :) |
| 16:12 | <timeless> | AryehGregor: i was going to say "at least you aren't playing w/ RTL" |
| 16:12 | <annevk> | maybe I should omit abstract too |
| 16:12 | timeless | just filed some bugs about non English UI foolishness |
| 16:12 | <annevk> | yes I should |
| 16:12 | <timeless> | this app which is aware of the input method editor (on screen keyboard) |
| 16:12 | <AryehGregor> | timeless, I was dumping the text of my sheva berachos invitation in Google Docs for safe-keeping. |
| 16:13 | <timeless> | asked the user to enter text in English |
| 16:13 | <timeless> | but the IME only supports Hebrew (and has no way to switch) |
| 16:13 | <timeless> | or the IME only supports Arabic (and has no way to switch) |
| 16:14 | <timeless> | oh, and i haven't found a way to insert BiDi push markers w/ the keyboard |
| 16:14 | <timeless> | so I end up with stupid things like: |
| 16:14 | <AryehGregor> | timeless, the Google Translate app for Android is ingenious like that. Translate from Hebrew to English . . . without being able to enter Hebrew! |
| 16:14 | <timeless> | 4. In foo, select CIBARA or .WERBEH |
| 16:14 | <timeless> | AryehGregor: that makes me feel so much better |
| 16:15 | timeless | likes not being alone |
| 16:15 | <AryehGregor> | (supposedly my version of Android natively supports Hebrew, but it actually doesn't) |
| 16:15 | <timeless> | lol |
| 16:15 | <AryehGregor> | (that will be more interesting when I move to Israel in a few months) |
| 16:15 | <AryehGregor> | (for now I use a third-party keyboard when necessary) |
| 16:15 | <timeless> | my BlackBerry thankfully has both Hebrew as a UI language and Hebrew for hardware and software keyboards |
| 16:15 | <timeless> | there are just a couple of rough edges |
| 16:15 | <timeless> | which are mostly things you'd *never* see |
| 16:16 | <timeless> | (hence me only spotting them now and not months ago) |
| 16:18 | <AryehGregor> | Yes, Hebrew works fine on my fiancée's Blackberry, although it's not unlocked, so in Israel it will mostly serve as an expensive paperweight. |
| 16:18 | <AryehGregor> | Actually, Hebrew text doesn't render even remotely correctly in my version of Android either. |
| 16:18 | <AryehGregor> | Nexus One. |
| 16:18 | <AryehGregor> | It's incredibly annoying. |
| 16:18 | <AryehGregor> | It's sometimes displayed in reverse order, and nikkud is absolutely hopeless. |
| 16:18 | <AryehGregor> | And apparently my phone is now too old to get updates anymore. |
| 16:19 | <timeless> | cyanogen? |
| 16:19 | <AryehGregor> | Which I got, uh, like a year and a half ago? |
| 16:19 | <AryehGregor> | Amazingly lame. |
| 16:19 | <AryehGregor> | Yes, I might have to resort to Cyanogen. |
| 16:20 | <timeless> | http://theunderstatement.com/post/11982112928/android-orphans-visualizing-a-sad-history-of-support |
| 16:23 | <AryehGregor> | That is horrifying. |
| 16:33 | <jgraham> | I just noticed Brendan Eich accused someone else of writing an overlong bug comment |
| 16:33 | <jgraham> | That is funny |
| 16:33 | <jgraham> | Also, if anyone is writing a dictionary and needs good examples for irony ^ |
| 16:34 | <AryehGregor> | It was probably on one of the WebIDL bugs I filed, right? |
| 16:34 | <jgraham> | Yeah, the legacyconst one |
| 16:42 | <timeless> | AryehGregor: i take it you haven't seen that before? |
| 16:42 | <AryehGregor> | No. |
| 16:42 | <AryehGregor> | I mean, I knew the basic fact, I just didn't see the detailed analysis. |
| 16:43 | timeless | thought it made news everywhere |
| 16:43 | <annevk> | hsivonen: thanks for that tweet btw |
| 16:43 | <annevk> | hsivonen: seems that's sufficient to inform the world the new state of affairs :) |
| 16:43 | <jgraham> | I thought it was generally agreed that the iPhone parts of that figure wre misleading |
| 16:45 | <timeless> | jgraham: no |
| 16:45 | <AryehGregor> | In what way? |
| 16:45 | <timeless> | what was agreed was that people can't read labels |
| 16:45 | <timeless> | (or perhaps "people don't read") |
| 16:47 | <timeless> | AryehGregor: people somehow had trouble understanding that "on current major version" doesn't mean "on current major version <today>" it means "<on current major version at the relative mark time indicated -- based on the original release date>" |
| 16:47 | <AryehGregor> | That was apparent to me. |
| 16:47 | <timeless> | if you look at /. and others, you'll discover that people can't read |
| 16:48 | <timeless> | and/or can't interpret charts |
| 16:48 | <timeless> | it's incredibly depressing |
| 16:51 | <jgraham> | (alternative theory: the visualisation is bad is bad) |
| 16:51 | <AryehGregor> | I think the visualization here was good. It makes it clear how long each phone gets updates. |
| 16:51 | <timeless> | it would be possible to make the graph have a <date> based x axis |
| 16:51 | <AryehGregor> | But then it would be harder to compare meaningfully. |
| 16:51 | <AryehGregor> | You'd know which phones are up-to-date now, but it would be bad for telling how long each one got support. |
| 16:51 | <timeless> | but then you'd have to do a heck of a lot of scanning |
| 16:51 | <AryehGregor> | It is a bit confusing that the chart only goes to three years, so if you don't pay attention you might think the oldest iPhones are still supported. It should be extended to the present in all cases. |
| 16:52 | <AryehGregor> | Although then it's confusing in a different way. |
| 16:52 | <AryehGregor> | Hmm. |
| 16:52 | <AryehGregor> | Visualization is hard. |
| 16:53 | <timeless> | that could be done i suppose |
| 18:07 | <annevk> | are data URLs already same origin with the XMLHttpRequest origin? |
| 18:08 | <annevk> | or does XMLHttpRequest need to say something for that? |
| 18:10 | <timeless> | what do people use for temporary image uploads? |
| 18:11 | <Ms2ger> | I used to use http://imageshack.us/ when I did that |
| 18:12 | <timeless> | thanks |
| 18:13 | <timeless> | AryehGregor: http://imageshack.us/f/687/016aandroidorphansbydat.png/ |
| 18:13 | <timeless> | is that really better? |
| 18:18 | <Ms2ger> | Anybody who feels like figuring out what setInterval.length should be? |
| 18:18 | <Ms2ger> | AryehGregor? |
| 18:18 | <timeless> | heh |
| 18:19 | <timeless> | lol, gecko says 0 |
| 18:19 | <Ms2ger> | Yep |
| 18:19 | <Ms2ger> | Bug was filed in 2003 |
| 18:20 | <timeless> | IE says the same fwiw |
| 18:20 | <timeless> | it's probably based on the fact that the first argument type was chaotic |
| 18:20 | <timeless> | so gecko idl was foo setInterval(void); |
| 18:21 | <timeless> | where the c++ impl stole arguments from the js stack |
| 18:21 | <Ms2ger> | Indeed |
| 18:21 | <smaug____> | we should change that |
| 18:21 | <smaug____> | to take jsvals |
| 18:21 | <timeless> | Chrome agrees w/ 0 |
| 18:21 | <smaug____> | as parameters |
| 18:21 | <timeless> | so i think we have all the browsers agreeing on 0 |
| 18:21 | <Ms2ger> | smaug____, we can't do that yet :) |
| 18:22 | <smaug____> | hmm, why not |
| 18:22 | <timeless> | is there a really big reason for changing it? |
| 18:22 | <smaug____> | consistency would be a good reason |
| 18:22 | <timeless> | other than breaking any site that happens to already be using this value? |
| 18:22 | <Ms2ger> | The setInterval(fun, timeout, arg1, ...) form |
| 18:23 | <smaug____> | ah, the ... part |
| 18:24 | Ms2ger | patiently awaits the WebIDL parser |
| 18:35 | <timeless> | Ms2ger: so, you'd rather the value be 2? |
| 18:35 | <Ms2ger> | No, I don't care |
| 18:35 | <Ms2ger> | I want to close bugs |
| 18:41 | <timeless> | oh brother |
| 18:41 | <timeless> | Safari says the answer is 2 |
| 18:42 | timeless | wonders if that's just because it's an older version of webkit |
| 18:42 | <annevk> | either you are playing this naive or you should be pleased because there's several engines that actually agree here! |
| 18:42 | <timeless> | 5.1.1 (7534.51.22) |
| 18:42 | <annevk> | usually it's like 4 browsers, 6 different answers |
| 18:42 | <timeless> | heh |
| 18:44 | <timeless> | opera also says 0 |
| 18:44 | <timeless> | so, Safari is my only odd man out |
| 18:44 | <_bga> | annevk does not count Konqueror heh |
| 18:44 | <timeless> | 4/5 browsers agree the answer is 0 |
| 18:44 | <timeless> | the answer is 0. resolved. |
| 18:44 | <timeless> | i'm assuming that safari would say 0 if i could get a newer webkit backend for it |
| 18:45 | <timeless> | annevk: yes, i naively assumed that both webkit browsers on my computer would agree |
| 18:45 | <annevk> | _bga: maybe I didn't count IE and Konquerer, who knows |
| 18:46 | <timeless> | grr |
| 18:46 | <annevk> | timeless: fwiw, with Web IDL it's also often the case that there is interop, but everyone wants to change to something new instead |
| 18:46 | <timeless> | rim's webkit also says 2 |
| 18:46 | timeless | doesn't remember how to ask what version of webkit this has |
| 18:46 | <annevk> | timeless: I suspect it depends on the ECMAScript implementation or maybe a timer implementation if that is also not in WebKit itself |
| 18:48 | <timeless> | ok, i have webkit 534.11+ if my blackberry ua is to be trusted |
| 18:48 | <timeless> | annevk: yeah, i know |
| 18:48 | <timeless> | re webidl and vendors wanting to change even w/ interop |
| 18:49 | <timeless> | my guess though here is that webkit changed and my safari/torch browsers are just old webkits |
| 18:49 | <timeless> | (i could check, it isn't worth my time right now) |
| 18:50 | <timeless> | anyone know if giuseppe pascale (opera) irc's outside of meetings? |
| 18:55 | <annevk> | sometimes internally |
| 18:55 | <timeless> | i'd like to send him backchannel on irc, it's easier than me writing tiny emails on my phone |
| 18:57 | <annevk> | I'd just ask him |
| 18:58 | <annevk> | (which would require an email :) ) |
| 19:13 | <timeless> | my aren't we helpful :) |
| 19:13 | <timeless> | AryehGregor / jgraham : i'd really be interested in knowing if you found http://imageshack.us/f/687/016aandroidorphansbydat.png/ to be better/more useful |
| 19:40 | <Ms2ger> | annevk, the last example in http://dev.w3.org/csswg/cssom/#examples-0 may be missing a decimal point |
| 20:57 | <timeless> | anyone here speak japanese / happen to recognize jakomobile.com ? |
| 20:57 | <Ms2ger> | MikeSmith, ^ |
| 20:58 | <MikeSmith> | timeless: there's not japanese on that page :) |
| 20:58 | <MikeSmith> | and I never heard of jakomobile |
| 20:59 | <timeless> | http://pastebin.mozilla.org/1388112 |
| 21:17 | MikeSmith | looks at timeless link |
| 21:17 | <timeless> | thanks |
| 21:17 | timeless | was wondering if MikeSmith died |
| 21:17 | <MikeSmith> | been working on slides I was asked to send to an event organizer a week ago :) |
| 21:18 | <MikeSmith> | for a presentation tomorrow |
| 21:19 | <MikeSmith> | so this is some kind of service that you apparently were signed up for |
| 21:19 | <MikeSmith> | the details about what the service actually is are sparse |
| 21:19 | <MikeSmith> | but maybe a gaming service? |
| 21:19 | <MikeSmith> | something called A-Team? |
| 21:44 | <timeless> | do i want to try to tell them i didn't sign up for it? |
| 21:56 | <MikeSmith> | timeless: dunno |
| 21:56 | <MikeSmith> | it is a legitimate service at least |
| 22:01 | <timeless> | ok, so, according to gmail's translate, i send email to the jakomobile address, is that correct? |
| 22:01 | timeless | thanks MikeSmith for taking the time to look |
| 22:01 | <MikeSmith> | yeah |
| 22:15 | <MikeSmith> | "Institutions will try to preserve the problem to which they are the solution." |
| 22:17 | <MikeSmith> | "pleas to evolve are drowned in the politics of an aging organization resistant to change" |
| 22:17 | <MikeSmith> | "People have a great capacity for change. Those people can and will continue to lead us as our institutions fail and eventually harm us." |
| 22:19 | <Ms2ger> | Context? |
| 22:19 | <MikeSmith> | "still solving a problem projects don't have, source hosting, appears to be beyond the capabilities of the organization" |
| 22:19 | <MikeSmith> | http://www.mikealrogers.com/posts/apache-considered-harmful.html |
| 22:20 | <MikeSmith> | "People and their contributions are as transparent as we can imagine and the direct connection of these people to each other turn social problems back in to social problems rather than political problems." |
| 22:21 | <MikeSmith> | "it is hosted, developed, and maintained by someone but they do not enforce any set of governance or process over the users of the system" |
| 22:21 | <MikeSmith> | "became a very political organism and navigating those politics has come to require more and more institutional knowledge over the years" |
| 22:27 | Ms2ger | likes how http://dowebsitesneedtolookexactlythesameineverybrowser.com/ looks exactly the same in every browser |
| 22:28 | <timeless> | not lynx :) |
| 22:28 | <timeless> | and technically it depends on which of three fonts you have |
| 22:28 | <timeless> | since that controls how NO! is rendered :) |
| 22:29 | <timeless> | http://wayback.archive.org/web/*/dowebsitesneedtolookexactlythesameineverybrowser.com |
| 22:30 | timeless | likes that the site has been around for years and is properly indexed |
| 22:31 | <timeless> | MikeSmith: it's unfortunate that the author of that post didn't use a spell checker |
| 22:31 | timeless | wonders if that's because Apache doesn't have one |
| 22:34 | <MikeSmith> | timeless: man, I think few people would read that post and have their first comment be that it needs a spell checker |
| 22:34 | <MikeSmith> | you're exceptional :) |
| 22:34 | <timeless> | thanks :) |
| 22:34 | <timeless> | actually, what was going through my mind was a parallelism to a mozilla thread |
| 22:34 | <timeless> | involving webdev v. webtools |
| 22:34 | <timeless> | but i didn't really want to share that |
| 22:35 | <timeless> | what bothers me is that having arguments which are probably fairly good |
| 22:35 | <timeless> | but happen to be tarnished by silly easily catchable errors |
| 22:36 | <timeless> | makes it harder to be willing to present them in other venues w/o having their value discarded for lack of ... |
| 22:36 | timeless | tries to find a similar complaint by someone else about an even shorter thing in a different context |
| 22:37 | <timeless> | gah, the fact that lists.mozilla.org and mail.mozilla.org aren't the same is um... |
| 22:38 | <timeless> | > oh, God forbid Mozilla hire fucking editors: "When discussing the list else where, please respect people's privacy and do name participants, individuals or companies. " |
| 22:38 | <timeless> | > https://mail.mozilla.org/listinfo/enterprise |
| 22:39 | <timeless> | n.b. that seems to have been corrected |
| 22:56 | <Hixie> | AryehGregor: you around? |
