| 02:15 | <Hixie> | heycam: Is getting Date into WebIDL on your radar? |
| 02:20 | <heycam> | Hixie, it's on my radar in as much as i still have that mail flagged to reply to :) |
| 02:20 | <heycam> | do you want it so that you can have an ES Date object there? |
| 02:26 | <Hixie> | HTML5 uses Date in one of the interfaces, and someone noticed that that doesn't have a definition anyway |
| 02:27 | <heycam> | is your intention for it to map to an ES Date object? |
| 02:27 | <Hixie> | yes |
| 02:27 | <Hixie> | i was using DOMTimeStamp which was defined that way but it changed :-) |
| 02:27 | <Hixie> | http://lists.w3.org/Archives/Public/public-webapps/2009JanMar/0458.html |
| 02:28 | <heycam> | hmm, there's no entry in http://www.whatwg.org/specs/web-apps/current-work/multipage/infrastructure.html#reflect for Dates |
| 02:29 | <heycam> | would you expect myTimeElement.date.setDate(...) to do something? |
| 02:29 | <Hixie> | there aren't any DOM attributes of type Date that reflect anything as far as i know |
| 02:29 | <heycam> | ones on TimeElement do |
| 02:29 | <heycam> | s/T/HTMLT/ |
| 02:30 | <heycam> | so actually it's just dateTime on HTMLTimeElement that is defined to reflect |
| 02:30 | <Hixie> | not as far as i can tell |
| 02:30 | <Hixie> | yeah |
| 02:30 | <heycam> | ok |
| 02:30 | <heycam> | so myTimeElement.date returns a new Date object each time it's got? |
| 02:32 | <Hixie> | yeah i guess that would be best |
| 02:32 | <Hixie> | that way there's no worry about the object being some weird live thing |
| 02:32 | <heycam> | yeah |
| 02:33 | <heycam> | (but maybe it'd be nice if a Date object could be read only somehow) |
| 02:33 | <heycam> | an ES Date, that is |
| 02:33 | <Hixie> | i think i'd rather not introduce new complexity for browsers to screw up |
| 02:34 | <heycam> | anyway, it's on my radar (though i don't seem to have a note for it; i'll add one now) |
| 02:35 | heycam | goes to get some breakf.. uh, brunch |
| 02:42 | <Hixie> | cool, thanks |
| 02:42 | <Hixie> | later |
| 04:17 | <mpilgrim> | http://twitter.com/fiddlerelf/status/1674755344 |
| 04:53 | <MikeSmith> | about JSONDB, etc., proposals for document-based unstructured client-side storage: in what ways are these alternatives to SQL-based structured storage? |
| 04:54 | <MikeSmith> | it seems more like they are alternatives to simple name/value storage -- localStorage |
| 05:00 | MikeSmith | re-reads http://blog.vlad1.com/2009/04/06/html5-web-storage-and-sql/ |
| 05:24 | MikeSmith | notices that Hixie forgot to set the authors and validators flags in his checkin description for r3054 |
| 05:56 | <hsivonen> | Hixie: Re: getElementsByTagName |
| 05:56 | <hsivonen> | Hixie: the sentence "ames of HTML elements match regardless of case, but elements from other namespaces are treated as in XML (case-sensitively)." is bad |
| 05:56 | <Hixie> | yeah i don't know what non-normative text to put |
| 05:57 | <hsivonen> | Hixie: the next sentence, which is not equivalent, is OK |
| 05:57 | <Hixie> | the next sentence is an implementation conformance criteria |
| 05:57 | <Hixie> | so it's not always visible |
| 05:57 | <Hixie> | which leaves... nothing |
| 05:57 | <Hixie> | for the authors |
| 05:57 | <Hixie> | hence the annoying sentence |
| 05:58 | <hsivonen> | HTML elements match by lower-casing the argument before comparison |
| 05:58 | <Hixie> | MikeSmith: technically, we readded type=tel, it was in an early WF2 draft :-) |
| 05:58 | <MikeSmith> | Hixie: ah |
| 05:58 | <Hixie> | hsivonen: ok |
| 07:12 | <MikeSmith> | hsivonen: how does offline cache relate to parsing |
| 07:13 | <MikeSmith> | (in regard to http://twitter.com/hsivonen/status/1669704043) |
| 07:14 | <Hixie> | manifest attribute is processed in the parser |
| 07:15 | <MikeSmith> | ah |
| 07:30 | <archtech> | has anything from HTML% been adopted in IE8/ |
| 07:30 | <archtech> | Ops, pardon the typos :P HTML5 |
| 07:32 | <Hixie> | offhand, onhashchange, localStorage, sessionStorage, and postMessage() |
| 07:32 | <archtech> | Oh, great, I thought onHashChange was proprietary. |
| 07:32 | <archtech> | Thanks, Hixie. |
| 07:45 | <MikeSmith> | archtech: I think IE8 also supports navigator.online |
| 07:45 | <archtech> | MikeSmith, great :) |
| 08:20 | <Hixie> | MikeSmith: yeah but that one was added to html5 because they implemented it and we needed something similar :-) |
| 08:22 | <archtech> | Hixie, that's a great approach to standards as well :) it has to work both ways. |
| 08:23 | <MikeSmith> | Hixie: huh, didn't know that came from IE originally |
| 08:23 | <MikeSmith> | drag-and-drop from IE originally also? |
| 08:39 | <MikeSmith> | 'Frankly, I'm finding this whole "rock star on the edge" thing is getting old.' |
| 08:40 | <MikeSmith> | I think it's actually the real rock starts on the edge who are getting old. |
| 08:40 | <MikeSmith> | or who have gotten old |
| 08:40 | <MikeSmith> | but I don't know who the rock star is in this particular case |
| 08:41 | virtuelv | goes to read mrlastweek |
| 08:41 | <MikeSmith> | virtuelv: that ain't MLW |
| 08:41 | <MikeSmith> | MLW differs in terms of humor |
| 08:42 | <MikeSmith> | a sense of humor, that is |
| 08:42 | <MikeSmith> | that is, in having one |
| 08:42 | <MikeSmith> | sort of |
| 08:43 | <MikeSmith> | anyway, I guess the rock star on the edge in this case is mpilgrim |
| 08:44 | <MikeSmith> | mpilgrim: your petty bickering risks undermining the credibility of the entire effort |
| 08:44 | <MikeSmith> | so please stop that |
| 08:44 | <MikeSmith> | the petty bickering, at least |
| 08:44 | <MikeSmith> | please do keep up the rock star part though |
| 08:49 | <MikeSmith> | "An end result of such shenanigans is that web page validity no longer means what it used to mean." |
| 08:49 | <MikeSmith> | yes. good. |
| 08:50 | <MikeSmith> | though we prefer the term "jackassery" |
| 08:51 | <MikeSmith> | " In the future, to get a "valid" stamp means that we have to adhere to a small, controlling group of people's interpretation of what the web should be,..." |
| 08:53 | <MikeSmith> | here's to hoping that more people do adopt that attitude that "this page is valid <whatever>" stamps are totally worthless |
| 08:54 | <MikeSmith> | regardless of what their reasons for adopting that attitude are |
| 08:54 | <MikeSmith> | I do like the "This page is valid in crappy biased markup" proposal, though. |
| 08:54 | <MikeSmith> | that one is a keeper |
| 09:06 | <Lachy> | Hixie, the new <hgroup> element seems completely useless. Why should anyone bother using it around heading elements? What's the difference between: |
| 09:06 | <Lachy> | <header><hgroup><h1>Foo</h1><h2>Bar</h2></hgroup> <p>more header content...</header> and |
| 09:06 | <Lachy> | <header><h1>Foo</h1><h2>Bar</h2> <p>more header content...</header> |
| 09:07 | <annevk5> | counts as a single heading in an outline prolly |
| 09:08 | <Lachy> | <header> should do that by itself |
| 09:09 | <annevk5> | <header> doesn't influence outline at all |
| 09:09 | <Lachy> | oh, he made <header> not be heading content :-( |
| 09:10 | <Lachy> | That really sucks |
| 09:13 | <billyjackass> | Lachy: your anti-hgroup vehemence is refreshing |
| 09:14 | <MikeSmith> | but it's too late |
| 09:14 | <MikeSmith> | I've already made a patch that adds <hgroup> to validator.nu |
| 09:15 | <MikeSmith> | so it's in the validator, and therefore now the Word of God |
| 09:16 | <MikeSmith> | and hsivonen has finally come to his senses about the "no badges" nonsense |
| 09:16 | <MikeSmith> | and agreed to start issuing badges |
| 09:16 | annevk5 | wonders if http://lists.w3.org/Archives/Public/public-device-apis/2009May/0003.html was ever public knowledge |
| 09:16 | <MikeSmith> | the v.nu-issued badge will say, "This page is valid in crappy biased markup" |
| 09:17 | <MikeSmith> | but it will be a badge nonetheless |
| 09:17 | <annevk5> | "This is good to hear. Geo APIs was booted out of WebApps because of IP threats from Nokia; has that now changed?" |
| 09:18 | <MikeSmith> | annevk5: that's not the reason it was "booted out of WebApps", so it doesn't really matter |
| 09:18 | <MikeSmith> | it was never "in" WebApps to begin with, for one thing |
| 09:20 | <MikeSmith> | and given that the Geolocation work has been very successful by any rational measure that anybody'd care to use (e.g., multiple implementations shipped already, actual people benefitting from it already).. |
| 09:21 | <MikeSmith> | well, I'd hope people would find something better to worry about |
| 09:21 | <MikeSmith> | like our petty bickering here |
| 09:22 | <MikeSmith> | which is undermining the credibility of the entire effort |
| 09:28 | <annevk5> | we have credibility? :p |
| 09:30 | <Lachy> | MikeSmith, it's never too late to reverse a mistake introduced into the spec, and hgroup is one of those things that won't last long |
| 09:30 | <MikeSmith> | oh, it will last. It has a chapter and verse number already. |
| 09:31 | <MikeSmith> | Lachy: are you suggesting we remove verses from the Bible? |
| 09:31 | <MikeSmith> | I don't God would like that. |
| 09:31 | <MikeSmith> | I mean, he wrote that book. |
| 09:31 | <MikeSmith> | indirectly, at least |
| 09:31 | <Hixie> | hsivonen is ok with badges now? |
| 09:31 | <Hixie> | have you been giving him some of your drugs |
| 09:32 | <MikeSmith> | I sense the hand of God in the HTML5 spec. Either that or the hand of Satan. can't tell which quite yet. |
| 09:32 | <Lachy> | MikeSmith, sure. Stuff has been added and removed from the bible many times over the past 2000 years (although not so much recently) |
| 09:33 | <MikeSmith> | Hixie: I was speaking on hsivonen's behalf. Have not checked with him yet, but I am reasonably confident that he will see the wisdom in the "This page is valid in crappy biased markup" badge. |
| 09:33 | <Hixie> | Lachy: <header> has to be allowed to contain <nav>, that's the way people want to use it. Given that, it can't be part of the outlining algorithm, Given that, unless we want every spec on W3C space to have one subsection called "Working Draft - 2 April 2009" or whatever, we need an hx grouping element for subheadings. |
| 09:34 | <Hixie> | MikeSmith: uh huh :-P |
| 09:35 | <Philip`> | I don't like badges |
| 09:35 | <Philip`> | but stickers are great |
| 09:35 | <Philip`> | We should have more stickers |
| 09:36 | <Lachy> | Hixie, The previous content model for it already did allow <nav> as a descendant. |
| 09:37 | <MikeSmith> | Hixie: btw, we are going to use the following at the v.nu badge - |
| 09:37 | <MikeSmith> | http://blog.zugara.com/wp-content/uploads/2009/04/leprechaun.jpg |
| 09:37 | <Hixie> | MikeSmith: (maybe you could convince him to use "I Check My Markup With validator.nu!" without saying it's valid) |
| 09:38 | <Hixie> | Lachy: no it didn't |
| 09:38 | <Hixie> | Lachy: it disallowed sectioning content descendants |
| 09:38 | <Hixie> | Lachy: nothing can be heading and contain sectioning at the sam e time |
| 09:38 | <MikeSmith> | Hixie: yeah (seriously) that would be good. would be good also if W3C validator stopped issuing the badges it does now, and issues an "I Check My Markup..." thing instead |
| 09:38 | <Philip`> | That wouldn't fit on the button |
| 09:39 | <MikeSmith> | Philip`: I'm working on that |
| 09:39 | <Philip`> | Maybe use an animated GIF with marquee text |
| 09:39 | <MikeSmith> | heh |
| 09:39 | <Hixie> | parents around here have "My Student Is An Honour Student At Bla Bla High School" as bumper stickers |
| 09:39 | <Hixie> | maybe we can have "My Web Page Was Checked With validator.nu!" as bumper stickers |
| 09:39 | <Hixie> | give them out at conferences |
| 09:40 | <Lachy> | oh, I misread it. |
| 09:41 | <Lachy> | also, the 3rd example for <header> now looks a little crazy. Why are the "Important News" and "Games" headings included within the header? |
| 09:41 | <MikeSmith> | Lachy: Get thee behind me. Nothing in the Bible has ever been added or removed. Some things in particular *serializations* of the Bible have been added or removed. But the |
| 09:42 | <MikeSmith> | But the abstract model of the Bible has never been altered, and is unalterable. |
| 09:42 | <Lachy> | haha |
| 09:43 | <Lachy> | MikeSmith, stories were written and added to the bible by hundreds of men over many centuries |
| 09:43 | <Hixie> | Lachy: yeah the examples could be improved... send mail or file a bug if you have any better ideas |
| 09:44 | <Philip`> | Lachy: But those hundreds of men were just expressing the fundamental truth laid down by God, so it is as if the book came from a single hand |
| 09:45 | <MikeSmith> | Philip`: Amen. |
| 09:45 | <MikeSmith> | Hallelujah |
| 09:45 | <Lachy> | yeah, and all those men who made copies of it by hand and introduced their own "corretions" in the process were just acting on behalf of god too... |
| 09:46 | <MikeSmith> | the Holy Spirit chooses Philip` to speak through now and then |
| 09:46 | <Lachy> | :-D |
| 09:49 | <Philip`> | Lachy: They were just fixing typos |
| 09:50 | <Philip`> | or, uh, scribos |
| 09:50 | <Philip`> | because maybe the people writing down the original were a bit drunk at one point and started making up a few extra sections, and so the later people had to fix those mistakes |
| 09:51 | <Philip`> | Or maybe they made changes as a way of detecting copyright infringement, like how map makers include fake roads so they can see who's copied their maps |
| 09:51 | <Lachy> | yeah, and I'm sure it had nothing to do with needing the bible to suit the needs of the religious leaders at the time, wanting to control the public with it. |
| 09:52 | <Lachy> | (copyright didn't exist until the 18th century) |
| 09:56 | <Philip`> | "One of the earliest copyright disputes reputedly took place in 557 A.D. between Abbot Finnian of Moville and St. Columba over St. Columba's copying of a Psalter belonging to an Abbot. The dispute over ownership of the copy led to the Battle of Cúl Dreimhne (also known as the Battle of Cooldrumman), in which 3,000 men were killed." |
| 09:58 | <Lachy> | sure, but there were no copyright laws at the time. |
| 10:00 | <Philip`> | The concept of "if you copy my stuff, I'll send my army to attack your army" seems much more powerful than any copyright law |
| 10:04 | <Lachy> | oddly enough, that's similar to the technique being employed today by the RIAA/MPAA, only using lawyers instead of soldiers and slightly less killing |
| 10:05 | <Philip`> | So, apart from being completely different, it's similar? |
| 10:05 | <annevk5> | why is it impossible to find the anolis spec generator? |
| 10:05 | annevk5 | wonders what keyword he's missing |
| 10:05 | <Lachy> | http://anolis.gsnedders.com/ |
| 10:05 | <annevk5> | that's just the source code |
| 10:05 | <Philip`> | gsnedders needs more Google juice |
| 10:06 | <Lachy> | http://pimpmyspec.net/ |
| 10:06 | <Philip`> | http://wiki.whatwg.org/wiki/Developing_HTML5_derivatives points at http://pimpmyspec.net/ if that's what you want |
| 10:09 | <annevk5> | ta |
| 10:28 | <annevk42> | Hixie, why the inconsistency in the event handler attribute tables? first heading lowercase, second heading starts with a capital |
| 11:22 | <virtuelv> | WHile being completely off topic: Could someone bash the twitter-folks over the head with the largest cluebat you can find? |
| 11:24 | <virtuelv> | Whatever was wrong with loading actual pages using actual URLs, instead of pretending everything is pointing to a document fragment |
| 11:30 | <annevk42> | looks a lot better than before actually, imo |
| 11:31 | <annevk42> | and it's not like you need to share the URLs, they're all bound to your login details anyway |
| 11:41 | <virtuelv> | yes, but it actually feels much slower than before |
| 11:43 | <virtuelv> | and I see no particular good reason for abolishing loading pages |
| 11:45 | <annevk42> | making things look better :) |
| 11:48 | <gsnedders> | And when it fails you have to reload everything and start from the newest all over again |
| 12:48 | gsnedders | wonders about automatically getting titles for photos based upon geo data |
| 12:49 | <gsnedders> | (i.e., writing code to allow me to be lazy in the future) |
| 13:56 | <Philip`> | How exciting |
| 13:56 | <Philip`> | Mr & Mrs Last Week posted a link to http://209.85.229.132/search?q=cache:(somewhere you probably don't want to bother looking)&cd=1&hl=en&ct=clnk&gl=uk&client=firefox-a |
| 13:56 | <Philip`> | which suggests they're either a Firefox user from the UK, or else copied a Google cache link from someone who was |
| 13:57 | <gsnedders> | Damnit! And I always thought they were IE users! |
| 13:57 | <Philip`> | (Well, probably the 'view PDF as HTML' link rather than the cache) |
| 14:14 | <Philip`> | The 'pv' tool is great |
| 14:15 | <Philip`> | If you're doing command-line processing of (as a hypothetical example) a 2GB gzipped file with some decompression and greps and perls and suchlike, you can just stick pv into the pipeline and it'll tell you how fast the processing is going and give you an ETA |
| 14:17 | Philip` | gets a fairly consistent 11 MB/sec, and is unsure whether that's good or not |
| 14:19 | <Dashiva> | Disk IO speed would be one possible baseline for comparison |
| 14:20 | <Philip`> | cat >/dev/null goes at 722 MB/sec |
| 14:20 | <Dashiva> | That seems a bit... high |
| 14:20 | <Philip`> | md5sum goes at 228 MB/sec |
| 14:21 | <Dashiva> | I suppose if you have the 3Gb SATA |
| 14:21 | <Philip`> | It would be a bit high if it was actually reading from physical disk, not from the file cache |
| 14:21 | <Dashiva> | No, even then... |
| 14:21 | <Philip`> | since it's really just reading from RAM |
| 14:21 | <Dashiva> | Yeah, true |
| 14:22 | <Philip`> | gzip -cd|md5sum goes at 11 MB/sec |
| 14:22 | <Philip`> | so I blame gzip |
| 14:23 | Philip` | assumes the disks are SATA, but doesn't really know |
| 14:24 | <Dashiva> | Tried different compression levels? |
| 14:24 | <Philip`> | and I don't know how to find out how fast they are |
| 14:28 | <Philip`> | The output from gzip -cd goes at ~185 MB/s if I compress with -1 |
| 14:28 | <Philip`> | The output from gzip -cd goes at ~205 MB/s if I compress with -9 |
| 14:29 | <Philip`> | and there's a roughly linear increase between those points |
| 14:29 | <Philip`> | (The input to gzip is faster with -1, but there's a lot more of it) |
| 14:34 | <Philip`> | But I'm way too lazy to wait for it to recompress 35GB of data with gzip -9 |
| 14:34 | <Dashiva> | Hmm |
| 14:34 | <Dashiva> | But you blamed gzip for 11 MB/sec earlier |
| 14:34 | <Philip`> | (even though I'm insufficiently lazy that I'll spend two minutes counting that it's 35GB) |
| 14:34 | <Dashiva> | Now you're saying it's 180-200 |
| 14:34 | <Philip`> | 11 MB/sec input |
| 14:35 | <Philip`> | 200 MB/sec output |
| 14:35 | <Dashiva> | ah |
| 14:35 | Philip` | should probably be clearer about such things |
| 14:37 | <Dashiva> | Well, it makes sense |
| 14:40 | <Philip`> | Oops, the files I've been looking at aren't all text/html |
| 14:40 | <Philip`> | I'm sure they were when I last looked, but they certainly haven't changed since then... |
| 14:43 | <Philip`> | Hmm, I've got 1500 more content-type headers than pages |
| 14:45 | <Philip`> | http://www.agriaffaires.de/gebrauchte/andere-landwirtschaftliche-anhaenger/698421/krampe-big-body-600-premium.html |
| 14:45 | <Philip`> | *13* content-type headers?! |
| 14:49 | <gsnedders> | Why not? I mean, it's hardly if browsers about any but the last. |
| 15:56 | <Dashiva> | In the future, to get a "valid" stamp means that we have to adhere to a small, controlling group of people's interpretation of what the web should be, and I'm just not willing to go there. |
| 15:56 | <Dashiva> | Isn't this how it has been all along? |
| 15:58 | <jgraham> | Philip`: You written any useful scripts to work with the dotnetdotorg data? |
| 15:58 | <Philip`> | Dashiva: No |
| 15:59 | <Philip`> | Dashiva: It's been a *different* small, controlling group of people |
| 16:00 | <Dashiva> | Right, my bad |
| 16:02 | <Dashiva> | I also like the part where it goes "I guess the browser vendors want to cooperate [...] Look forward to lots of browser incompatability" |
| 16:04 | <Philip`> | Dashiva: Where does it say that? |
| 16:04 | <Philip`> | Also, what "it" is it? |
| 16:04 | <Dashiva> | http://realtech.burningbird.net/semantic-web/semantic-web-issues-and-practices/going-non-standard |
| 16:05 | <Philip`> | Dashiva: I can't see any mentions of e.g. looking forward to incompatibility in there |
| 16:05 | <Philip`> | Oh, wait, maybe it sort of says that in the last-but-one paragraph |
| 16:05 | <Philip`> | though not with your words or your spelling error |
| 16:06 | <Dashiva> | Yeah, I always typo that |
| 16:07 | <Dashiva> | I've started getting ominous right, though! |
| 16:07 | <Philip`> | Oh my nous! |
| 16:08 | <Dashiva> | I still think omnious sounds more ominous. But it's a lost battle. |
| 16:08 | <Philip`> | I find it hard to understand the comment "You know who really makes me mad? The browser companies. They want obfuscation. They want imprecision. They want gaps. They want things in a state of confusion, so they can do what they want, and not have to worry about "standards"." |
| 16:09 | <Philip`> | since as far as I'm aware the browser companies formed the WHATWG specifically in response to imprecision and confusion and gaps in the W3C's specification of web technologies |
| 16:09 | <Dashiva> | I reckon it's just the usual misunderstanding of how things work |
| 16:09 | <Philip`> | and HTML5 defines everything precisely and tries to fill in all the gaps |
| 16:09 | <jgraham> | I find it really hard to understand Shelly in general. I really don't understand why she's so angry all the time |
| 16:09 | <Philip`> | (though it may still be obfuscated and confusing) |
| 16:09 | <Dashiva> | jgraham: Maybe the owls are out to get her |
| 16:10 | <jgraham> | Dashiva: Owls deliver post |
| 16:10 | <Dashiva> | They also come in the night and tear your legs apart |
| 16:10 | <Philip`> | Can owls deliver get too? |
| 16:11 | <jgraham> | Philip`: I don't recall Owls doing anything that resembles a get |
| 16:11 | <Dashiva> | They get field mice |
| 16:11 | <jgraham> | So HTTP over Owl would be non-trivial |
| 16:11 | <Dashiva> | Or I suppose that's more of a delete |
| 16:11 | <Philip`> | Or a head, depending on how hard they bite |
| 16:11 | <Dashiva> | Head is supposed to be safe, isn't it? |
| 16:12 | <jgraham> | Idempotent |
| 16:12 | <Dashiva> | Okay, then it's okay I guess |
| 16:12 | <jgraham> | Maybe if we could train them to only attach starfish and other creatures that can regrow parts of their body |
| 16:13 | <jgraham> | *attack |
| 16:13 | <Philip`> | jgraham: I have a Perl script that splits the file into more-easily-parseable parallelisable chunks, and a couple of organically grown (i.e. horridly written) Java programs to do grep and to run various analyses on the parsed content |
| 16:13 | <Dashiva> | But isn't idempotent defined as repeated applications being the same as one application? |
| 16:13 | <Dashiva> | Without saying anything about initial state vs one application |
| 16:15 | <Philip`> | jgraham: How about axolotls? |
| 16:16 | <Philip`> | (which apparently can regrow limbs) |
| 16:16 | <Philip`> | (which I learned via the ever-informative blog at http://fuckyoupenguin.blogspot.com/2009/03/no.html ) |
| 16:17 | <Philip`> | Dashiva: In that case, DELETE would be considered idempotent |
| 16:17 | <Philip`> | since deleting something twice has the same consequence as deleting it once |
| 16:18 | <Philip`> | so the term 'safe' seems better than 'idempotent' |
| 16:18 | <Dashiva> | Yeah, that's why I said safe :) |
| 16:20 | <jgraham> | Ah yes, head is safe + idempotent. Delete is jsut idempotent |
| 16:20 | <Philip`> | Of course it's not actually idempotent in an asynchronous environment |
| 16:21 | Philip` | wonders if HTTP has a more subtle definition of 'idempotent' |
| 16:27 | <Philip`> | jgraham: (All my other code for working with the data is one-off bash commands, usually some combination of sort and uniq and perl, and I don't have them saved anywhere) |
| 17:27 | gsnedders | isn't entirely happy with his photos from the ball |
| 17:32 | <Philip`> | gsnedders: Photoshop |
| 17:32 | <gsnedders> | … is where I've spent far too much time trying to make them less shit :) |
| 17:33 | <Philip`> | (I suppose I could say "gsnedders: Gimp" to promote OS software, but that might be easy to misinterpret) |
| 18:24 | <gsnedders> | http://www.flickr.com/photos/gsnedders/sets/72157617511428181/ — My best attempt. |
| 18:41 | <fearphage> | any of you ever used XHR to send an element? what is the use case for that? |
| 18:41 | <fearphage> | just investigating |
| 18:47 | <Philip`> | http://www.span-raps.com/English2/index.php?option=com_content&task=view&id=51&Itemid=3 - <time w:st="on" hour="13" minute="00"><u><span lang="EN-CA" style="mso-ansi-language: EN-CA">1:00 PM</span></u></time><u><span lang="EN-CA" style="mso-ansi-language: EN-CA">, Saturday, May 15<sup>th</sup>, 2004<p></p></span></u> |
| 18:47 | <Philip`> | Hmm, that's the only <time> element I see |
| 19:15 | <hsivonen> | interesting accusation on Shelley's blog: "You know who really makes me mad? The browser companies. They want obfuscation. They want imprecision. They want gaps. They want things in a state of confusion, so they can do what they want, and not have to worry about "standards"." |
| 19:15 | <Hixie> | Philip` commented on that earlier |
| 19:43 | <virtuelv> | hsivonen: oh, assume someone has a position, and attack that |
| 19:43 | <virtuelv> | let me look that up in the big book of fallacies |
| 19:45 | <virtuelv> | while I'm not involved in HTML5 as such, but other efforts, I can say that what we _don't_ want is imprecision |
| 19:45 | <Hixie> | yeah precision is somewhat the point :-) |
| 19:52 | <gsnedders> | Let's go back to HTML 4.01! |
| 19:52 | <gsnedders> | That's imprecise! |
| 19:53 | <gsnedders> | Wait, if we want imprecision, why are we revising it? |
| 19:55 | <fearphage> | can anyone help me with a use case/purpose for sending nodes/elements with XHR? or a place i could possibly find an answer for that |
| 20:14 | <Hixie> | i wonder what to do about accesskey="" |
| 20:23 | <Hixie> | webkit uses ctrl+option |
| 20:23 | <Hixie> | firefox uses ctrl |
| 20:23 | <Hixie> | (on mac) |
| 20:23 | <gsnedders> | Delete it! |
| 20:23 | <Hixie> | i never remember what opera's behaviour is |
| 20:23 | <Hixie> | something to do with the escape key? |
| 20:25 | <Philip`> | Shift+esc |
| 20:28 | <Hixie> | interesting |
| 20:28 | <Hixie> | just the menu, no direct shortcuts right? |
| 20:29 | <krijnh> | Jep |
| 20:29 | <Hixie> | interesting, Webkit, Firefox, and Opera all do nothing if the accesskey="" attribute is more than one character long |
| 20:30 | <Hixie> | only webkit supports accesskey=" " |
| 20:30 | <Hixie> | none of them strip leading spaces |
| 20:30 | <Hixie> | hmm |
| 20:31 | <Hixie> | this leaves us with some interesting options |
| 20:32 | <Hixie> | accesskey="" could be an enumerated attribute with values a..z, 0..9, and some keywords like "search", which could be bound to UA-specific standard bindings |
| 20:34 | <krijnh> | What's the benefit over using input type="search" for example? |
| 20:34 | <Hixie> | none really |
| 20:34 | <Hixie> | but we could have some generic keywords, like "f1" .. "f9", which map to a set of keys unique to each browser which are always available |
| 20:34 | <Hixie> | where "f1" is your most important site function |
| 20:35 | <Hixie> | though discoverability would be an issue |
| 20:35 | <krijnh> | Sounds confusing |
| 20:36 | <jgraham> | Hixie: I guess more than just a-z work on keyboards with more characters |
| 20:37 | <jgraham> | So that sounds bad |
| 20:37 | <jgraham> | gsnedders: 50mm f/1.8 It won't make you a better photographer but may be good for taking better portraits |
| 20:38 | <gsnedders> | jgraham: I've been more tempted by the 85mm |
| 20:38 | <jgraham> | gsnedders: Well assuming you are rich and always have rather a lot o space the 85mm is supposed to be very nice indeed |
| 20:39 | <gsnedders> | I mean the f/1.8 85mm, not the f/1.2 |
| 20:39 | <jgraham> | But the 50mm is like 80 GBP or less secondhand |
| 20:39 | <gsnedders> | Though unsurprisingly I would quite like the f/1.2 :P |
| 20:40 | <Hixie> | almost every link on http://www.456bereastreet.com/archive/200601/accesskey_problems_remain_in_xhtml_2/ is now 404 |
| 20:40 | <jgraham> | gsnedders: Still 300+GBP |
| 20:40 | <gsnedders> | Hmm, I remember seeing it for less before. Oh well. |
| 20:40 | <Philip`> | http://philip.html5.org/data/accesskey-values.txt |
| 20:40 | <Philip`> | (Not very interesting) |
| 20:40 | <jgraham> | The fundamental problem with accesskey is that it is a solution for a UI issue that has not been designed by UI designers |
| 20:40 | <Philip`> | (but at least it shows most people use single-character accesskeys) |
| 20:41 | <jgraham> | This is A Bad Idea |
| 20:41 | <krijnh> | Philip`: do the links include titles or something? |
| 20:41 | <Hixie> | Philip`: i like accesskey="10" |
| 20:41 | <Philip`> | krijnh: Which links? |
| 20:41 | <Hixie> | wish i had a "10" key |
| 20:41 | <gsnedders> | jgraham: I have been thinking about getting the 50mm just as a stop-gap solution until I have money, but I do so little portrait photography anyway, and the bokeh on the Sigma 17–70mm is nicer IMO |
| 20:41 | <krijnh> | Philip`: on which the accesskey attributes are set |
| 20:42 | <krijnh> | Perhaps a combination of accesskey and title can be helpful |
| 20:42 | <Philip`> | krijnh: Don't know, and I don't have an easy way of extracting that data |
| 20:42 | <krijnh> | Ke |
| 20:43 | <jgraham> | gsnedders: What is the maximum apature on the sigma? And are you prepared to live wih the fact tha there is a high chance that your sigma lens will have qualiy conrol issues |
| 20:43 | <jgraham> | ? |
| 20:43 | <jgraham> | *aperture |
| 20:43 | <gsnedders> | jgraham: f/2.8-3.5 only, which does kinda suck. |
| 20:43 | <Hixie> | http://www.w3.org/mid/B2750F31-8ACF-4B46-8D02-44371AEE4C9F⊙ac seems sound |
| 20:43 | <gsnedders> | jgraham: Also, it hasn't given me any problems yet |
| 20:43 | <Philip`> | gsnedders: The photos would be better if the lighting didn't make everybody look like corpses |
| 20:44 | <gsnedders> | Yeah, a lot of them are over-exposed. I've already rectified some of them. |
| 20:44 | <jgraham> | gsnedders: Oh, you already have it? Well I guess there's no point in discouraging you from buying it then |
| 20:44 | <gsnedders> | jgraham: Yes, I have it. That's what I took all those photos on. |
| 20:45 | <gsnedders> | jgraham: I got it to replace the 18–55mm kit lens, and it is certainly a lot better than that |
| 20:45 | <jgraham> | gsnedders: A lot of those would also benefit from some fill flash |
| 20:46 | jgraham | would like the 24-105L to replace his kit lens bu it is a bit expensive |
| 20:46 | <gsnedders> | Indeed. |
| 20:46 | <jgraham> | gsnedders: If you don't already have en external flash, it is well, wel worth getting one |
| 20:47 | <gsnedders> | I have been seriously considering getting one… |
| 20:47 | <jgraham> | The Speedlite 430EX is perfectly fine |
| 20:47 | <jgraham> | As far as I can tell |
| 20:47 | <gsnedders> | jgraham: But I can't decide between an external flash and 70–300mm |
| 20:47 | <jgraham> | (well I would like lower power sometimes, although in principle it is always possible to arrange for that) |
| 20:48 | <jgraham> | gsnedders: What is your use case for the 70-300mm? |
| 20:48 | <gsnedders> | jgraham: Things in the distance? :P |
| 20:48 | <jgraham> | gsnedders: What things in particular? |
| 20:49 | <jgraham> | It makes rather a lot of difference |
| 20:49 | <gsnedders> | jgraham: More seriously, there's quite a few things I'd like to be able to photograph while out walking which I can't really get to (e.g., I'm on a mountain and they aren't) |
| 20:49 | <jgraham> | So landscapes? |
| 20:49 | <gsnedders> | jgraham: yes |
| 20:49 | <gsnedders> | (You think you can get a simple answer from me!? Peh!) |
| 20:50 | <jgraham> | Oh in that case the 70-300mm is probably a reasonable idea (although I haven't used the canon one for more than a few shots) |
| 20:50 | <jgraham> | But external flash opens up whole new worlds of possibility |
| 20:51 | <jgraham> | The problem with phoography is a) there are a lot of good things to buy b) They are all expensive and c) None of hem is an effecive substitute for actual talent |
| 20:52 | gsnedders | lacks the talent part :P |
| 20:52 | jgraham | too |
| 20:52 | gsnedders | thinks jgraham is better than him |
| 20:53 | <jgraham> | Oh, and the money part :) |
| 20:53 | <gsnedders> | Yeah, I lack that too. |
| 20:53 | <gsnedders> | But I guess we both have too much we could buy :) |
| 20:54 | jgraham | would also like the 100-400mm. Or maybe one of the long fixed focal length things. |
| 21:20 | inimino | looks forward to a new era of "cool, oversexed, overscripted Ajax applications" |
| 21:31 | gsnedders | would quite like a Lensbaby, and they're fairly cheap, so he may get one soon |
| 21:31 | gsnedders | thought they were pointless and not overly fun until he got to play with Chris Wilson's at TPAC |
