| 02:06 | <Hixie> | gsnedders: i have a feature request to replace the earlier feature request |
| 02:07 | <Hixie> | gsnedders: instead of doing cross references across specs, i would now instead like to have a spec gen script that takes the same input style as now, but outputs an RFC-style text file |
| 02:07 | <Hixie> | there are tools that might help here: http://www.rfc-editor.org/formatting.html |
| 02:07 | <Hixie> | documentation of the format is here: http://www.ietf.org/ietf/1id-guidelines.html |
| 02:08 | <Hixie> | specifically http://www.rfc-editor.org/rfc-editor/instructions2authors.txt |
| 02:17 | <Hixie> | gsnedders: also, a template would be useful for new people using anolis |
| 02:19 | <Hixie> | shepazu: yt? |
| 02:36 | <Hixie> | annevk: yt? wanted to ask about whether you had registered or were planning to register the 'Origin' HTTP header |
| 04:33 | <BenMillard> | krijnh, everything in #wai-aria and #webapps for the past 8 days is status messages, including all lines marked important: http://krijnhoetmer.nl/irc-logs/ |
| 04:35 | <BenMillard> | krijnh, almost everything in #microformats is status messages |
| 04:36 | <BenMillard> | maybe the holidays last a longer than I think they do |
| 08:58 | <annevk> | Hixie, it is registered |
| 08:58 | <annevk> | Hixie, see the provisional header registry |
| 09:05 | <Hixie> | annevk: k, thanks |
| 09:16 | <annevk> | as for the content sniffing person at Opera, maybe hallvors would be a good place to start, but otherwise just contact me or zcorpan directly and we'll figure it out |
| 09:20 | <Hixie> | k |
| 10:02 | <annevk> | Hixie, you there still? |
| 10:02 | <annevk> | Hixie, I was wondering whether the "script context" thingy is stable now |
| 10:02 | <annevk> | Hixie, as that is what I need to use in XMLHttpRequest, right? |
| 10:03 | <annevk> | that is, I somehow need to define that window.XMLHttpRequest has an associated script context from where you can get the base URL, etc. blah blah |
| 10:30 | <Hixie> | annevk: completely unstable, both opera and webkit have said they want it changed |
| 10:32 | <annevk> | nicely nicely |
| 10:33 | <annevk> | well, will wait a bit with XHR then |
| 11:10 | Hixie | wonders what idiom "nicely nicely" translates to in dutch :-) |
| 11:13 | Hixie | replies to mail from november 2004 |
| 11:13 | annevk | isn't sure, got it from Ideal (BBC show) |
| 11:14 | <Hixie> | odd |
| 11:14 | <Hixie> | what context do they use it in? |
| 11:14 | <Hixie> | i've never heard it before |
| 11:14 | <Hixie> | google doesn't seem to help |
| 11:14 | <annevk> | the main character says it when he's pleased |
| 11:14 | <Hixie> | aah |
| 11:14 | <Hixie> | odd |
| 11:14 | <Hixie> | the british, really, you never know what they'll say next |
| 11:15 | <Lachy> | LOL |
| 11:15 | <Hixie> | ooh that reminds me, i have a top gear downloading |
| 11:15 | <Hixie> | i get to watch it tomorrow evening! |
| 11:15 | <Hixie> | yay |
| 11:15 | <Hixie> | i was so happy to see top gear on itunes |
| 11:16 | <Hixie> | back on topic, i wonder how to convert <datagrid> to an async API |
| 11:17 | <Hixie> | nessy: ping |
| 11:18 | <Lachy> | Hixie, is that the original or current BBC series, or the US version of Top Gear? |
| 11:18 | <Hixie> | current bbc series |
| 11:19 | <Lachy> | ok. I've not heard of it. Just reading about it in wikipedia |
| 11:19 | <Hixie> | you haven't heard of top gear? oh my |
| 11:19 | <Hixie> | it's awesome |
| 11:19 | <Hixie> | and i don't even remotely like cars |
| 11:20 | <Lachy> | apparently there's a new Australian version that started last year |
| 11:20 | <Lachy> | though it's on SBS, and so even if I were in Australia, I'd be unlikeliy to bother watching it |
| 11:21 | <Philip`> | It seems more like a TV show about people who are making a TV show about cars, rather than actually being a TV show about cars |
| 11:21 | <Hixie> | the reason to watch it isn't the show concept so much as the execution, and i doubt that other teams would get quite the same result |
| 11:21 | <Hixie> | Philip`: right |
| 11:24 | <Lachy> | so is it suitable for people like myself with no interest in cars and/or racing, except for the annual Bathurst 1000? |
| 11:25 | <takkaria> | yes |
| 11:25 | <takkaria> | very much so |
| 11:25 | <nessy> | Hixie: pong |
| 11:25 | <takkaria> | it's suitable for just about anyone :) |
| 11:25 | <Lachy> | alright. I'll download a few episodes later and check it out, if I remember |
| 11:26 | <Philip`> | (One particular example is that they had a segment filmed in America, and introduced it by saying they only had visas for making documentaries, and the last time they were in America they were accused of having made an entertainment film instead, and there wasn't time to get new visas, so this time they would have to be very careful not to be entertaining) |
| 11:26 | <Philip`> | (which is a bit meta) |
| 11:26 | <annevk> | yeah, Top Gear is funny, didn't know it was still going on |
| 11:27 | <Hixie> | nessy: according to ken, what i was saying yesterday is indeed google's opinion; he thinks you may have confused his interest in captioning standards for interest in annotation standards |
| 11:27 | <Hixie> | nessy: in particular, we're still very much playing around with annotation so we wouldn't really know what we wanted standardised yet |
| 11:27 | <nessy> | ah |
| 11:27 | <nessy> | hmm |
| 11:28 | <nessy> | I wonder what I confused then |
| 11:28 | hallvors | isn't interested in cars and doesn't watch TV anymore either.. :-p |
| 11:29 | <nessy> | anyway - we all know we need to solve the simple case first |
| 11:29 | <Hixie> | hallvors: TV is much more entertaining once you remove all ads, trailers, previews, overlayed images, etc, and make it on-demand and limit it to only quality programming. :-) |
| 11:29 | nessy | gets back to it :) |
| 11:30 | <Hixie> | nessy: :-) |
| 11:30 | <Lachy> | Hixie, if you're referring to the video annotation "feature" that YouTube has implemented, I really despise that. It has ruined so many otherwise good videos, because people abuse it way too much |
| 11:30 | <nessy> | no harm in discussing the future though! |
| 11:31 | Philip` | likes that Top Gear is an hour-long programme and because it's on the BBC it's actually really an hour, not 30 minutes plus adverts :-) |
| 11:31 | hsivonen | wonders how <video> rendering in WebKit is threaded |
| 11:32 | <Hixie> | Lachy: http://www.blogsdna.com/288/turn-off-youtube-video-annotations.htm |
| 11:33 | <Lachy> | Hixie, I always turn them off manually when I see them. |
| 11:33 | <Hixie> | Philip`: i was utterly shocked the first time i was exposed to US TV -- if you think (non-bbc) british tv is bad, you should see tv in the us, holy crap |
| 11:33 | <hsivonen> | I've seriously considered stopping paying for the TV permit and getting rid of the tuner accordingly |
| 11:33 | <Hixie> | Philip`: they play ads between the teaser and the first act, and between the last act and the credits! |
| 11:34 | <Lachy> | Hixie, unfortunately, there's no way to have them off by default |
| 11:34 | <Hixie> | Lachy: yeah it does seem that way, that's unfortunate |
| 11:34 | <hsivonen> | too bad availability of online movie rentals sucks in Finland |
| 11:34 | <Hixie> | i don't understand why media companies are still region-limiting availability |
| 11:35 | <Hixie> | that makes zero sense to me |
| 11:35 | <nessy> | hsivonen: get a digital tuner card and set up mythtv |
| 11:35 | <Lachy> | hsivonen, I suggest using Giganews and Newzbin for your TV and film needs |
| 11:35 | <Hixie> | why would you ever explicitly _restrict_ your audience? |
| 11:35 | <hsivonen> | nessy: I have a USB tuner and EyeTV |
| 11:35 | <Hixie> | then again, they also do drm and stuff, they clearly don't understand the digital age yet |
| 11:35 | <nessy> | and you have to pay for that in Finland? |
| 11:35 | <hsivonen> | nessy: yes |
| 11:36 | <nessy> | :( |
| 11:36 | <hsivonen> | nessy: like in the UK |
| 11:36 | <nessy> | not here |
| 11:36 | <nessy> | fortunately |
| 11:36 | <Lachy> | Hixie, it makes some sense for the BBC, since ok UK citizens are paying their BBC TV licences, and other people would just be free loading off them |
| 11:36 | <Philip`> | Hixie: If you sell something to a wider audience now, you can't sell it to them next year for more money |
| 11:36 | <Lachy> | s/since ok/since only/ |
| 11:36 | <Philip`> | so you lose the opportunity for price discrimination |
| 11:37 | <Hixie> | Lachy: it makes sense taht the bbc doesn't give away their content internationally, but they don't even sell most of it! top gear is only on iTunes in the US because it is shown on BBC America and BBC America has a separate deal with iTunes! |
| 11:37 | <nessy> | I heard this weird argument that audiences outside the target country would need to get different ads and so some sites restrict their region |
| 11:37 | <Hixie> | it's moronic |
| 11:37 | <Hixie> | Philip`: why can't yet sell it straight away for more money? by next year, half the audience will have bittorrented it |
| 11:38 | <nessy> | we still cannot get Hulu down here |
| 11:38 | <Philip`> | Hixie: They sell it on DVD, but I don't know if they sell it on US-regioned DVDs so maybe that doesn't help |
| 11:38 | <Lachy> | Hixie, I agree that there should be no restrictions on selling TV shows, films and music worldwide |
| 11:38 | <hsivonen> | it's also moronic that iTunes rentals are available only in a couple of English-language locales |
| 11:38 | <Philip`> | Hixie: I think you overestimate how many people use BitTorrent :-) |
| 11:38 | <hsivonen> | which suggests that someone at Apple thinks that they couldn't offer untranslated content elsewhere |
| 11:39 | <Hixie> | hsivonen: i doubt it's apple's fault frankly |
| 11:39 | <nessy> | it's usually the content owners (traditional distributors) |
| 11:39 | <hsivonen> | do you mean content owners prohibit untranslated distribution? |
| 11:39 | <nessy> | they are not entering a new market without knowing how to commercialise it properly |
| 11:39 | <Philip`> | Hixie: They probably want to negotiate with broadcasters in foreign regions to do the first showing (and thus get millions of viewers and lots of money from advertisers), which takes months or years to sort out and schedule, before distributing it online |
| 11:40 | <nessy> | I agree |
| 11:40 | <hsivonen> | after all, once the content is available in the U.S. and U.K. letting others pay for it would be free money |
| 11:40 | <Lachy> | unfortunately, the way the rights organisations like the RIAA and MPAA, and their international equivalents are structured, retailers have to negotiate rights in each territory individually, which sucks so much. And they also have to consider release schedules on local TV networks, so they don't compete directly with their old and failing business model |
| 11:41 | <hsivonen> | so is it just a coincidence that the U.K. was the next territory after the U.S.? |
| 11:41 | <Hixie> | Philip`: i would be surprised if they got more money from ads on syndication than they did from iTunes sales to an equal audience. And if it's not an equal audience, then one still doesn't preclude the other. |
| 11:41 | <Hixie> | hsivonen: sometimes the UK is first, even (e.g. some babylon 5 and sg-1 episodes aired in the UK before the US) |
| 11:42 | <Hixie> | basically the entire media industry is aniquated |
| 11:42 | <Hixie> | anyway |
| 11:42 | <hsivonen> | anyway, when Steve promised "Internation later this year [2008]", I expected something more than just the U.K. |
| 11:42 | <hsivonen> | *national |
| 11:43 | <Lachy> | hsivonen, Australia got some TV series too |
| 11:43 | <Hixie> | i expect he did too |
| 11:43 | <Philip`> | Hixie: You can't really broadcast a nationwide exclusive of some series if 5% of people have already downloaded it on iTunes and know exactly what's going to happen and give spoilers to everyone else and ruin the excitement |
| 11:44 | <Hixie> | Philip`: yes because national boundaries are really effective at stopping _that_ problem |
| 11:45 | <hsivonen> | I stopped watching series, because the broadcast distribution model just sucks for them--you have to capture the data at a certain time, and the time isn't stable because sports are allowed to go overtime |
| 11:46 | <Hixie> | i recommend interning with some US company, getting a US credit card, and keeping it when you leave |
| 11:46 | <Hixie> | :-) |
| 11:46 | <Lachy> | Philip`, the fact is that refusing to sell content directly to consumers only ignores the fact that Usenet and BitTorrent are already catering to that market well |
| 11:47 | <hallvors> | even my sister uses torrent. :) |
| 11:47 | <Lachy> | I only use torrent when something isn't available on Usenet because torrents can be and are actively tracked by rights organisations, whereas usenet can't be |
| 11:49 | <Hixie> | my neighbours try to use torrent |
| 11:50 | <Hixie> | i block it because they aren't any good at bandwidth management and it takes down my network |
| 11:50 | <Hixie> | (personally i just use HTTP to download tv shows) |
| 11:51 | <Philip`> | Lachy: Rights organisations could get the news servers taken down for hosting illegal content, and then replace them with servers run by themselves which log all accesses |
| 11:52 | <Hixie> | but they don't :-) |
| 11:52 | <Hixie> | (yet, anyway) |
| 11:53 | <Lachy> | Philip`, if they were running the servers, then they're the ones providing access and they can't turn around and accuse users of obtaining illegal access |
| 11:53 | <Philip`> | I guess the main difference is that people get caught/punished for uploading content (which will happen in any P2P system), not for downloading it (which is all that happens in Usenet) |
| 11:53 | <Lachy> | but Usenet servers are protected by the DMCA safe harbour provisions in the US, and similar laws in other countries |
| 11:55 | <Hixie> | so what you're saying is p2p systems should be written in such a way as to be covered by safe-harbour provisions... |
| 11:55 | <Lachy> | Hixie, yes, if that's possible |
| 11:55 | <Hixie> | that i want to see |
| 11:55 | <Hixie> | for the comedic value if nothing else |
| 11:55 | <Lachy> | like a Freenet-like P2P system |
| 11:55 | <Philip`> | I don't think I'd want to rely on a "safe harbour" defence when the service's entire reason for existence is to serve illegal content to people |
| 11:55 | <Lachy> | unfortunately, Freenet is a little too slow for such things |
| 11:55 | <Hixie> | bittorrent is used for far more than illegal content |
| 12:06 | <Philip`> | Grokster served legal content too, but its main usage and reason for existence was to serve illegal content |
| 12:06 | <Philip`> | which seems to be about the same situation as Usenet binaries |
| 12:06 | <hsivonen> | Philip`: Usenet binaries can be used for distributing copies of perl |
| 12:57 | <Lachy> | Philip`, I can't run the latest IE8 right now to test it. Does it recognise unknown elements properly by default, or does it still require the createElement() hack? |
| 12:57 | <Philip`> | Lachy: It requires createElement |
| 12:57 | <Philip`> | (or at least I'm almost certain it does, because they wouldn't make such drastic parser changes this late in the process) |
| 12:57 | <Philip`> | (though I don't remember explicitly testing it) |
| 12:57 | <Philip`> | Lachy: (https://connect.microsoft.com/IE/feedback/ViewFeedback.aspx?FeedbackID=364356) |
| 12:57 | <annevk> | I'm not sure we he thinks these new elements only address a narrow scope. The whole point was that these are the most common patterns authors are trying to find workarounds for using class names etc. |
| 12:57 | Lachy | responded http://alistapart.com/comments/semanticsinhtml5?page=2#11 |
| 12:57 | <annevk> | He also starts putting href="" on <li> suddenly which is not backwards compatible at all. And if is in fact demonstrating XHTML2 he should have used <nl> (or is that gone?). |
| 12:57 | <hsivonen> | http://twitter.com/adamcoop/statuses/1099359229 |
| 12:57 | <Lachy> | oh, I didn't notice the href on li elements. |
| 12:57 | <Philip`> | annevk: The most common patterns are a narrow scope because it's excluding the long tail, which consists of all the individually uncommon but combinedly wide semantics people want to express |
| 12:57 | <Philip`> | <li href> looks more like a typo than anything intentional |
| 12:57 | <annevk> | Comment number 6: "As said above; a solution already exists; its xhtml with a custom dtd – in fact thats precisely what a dtd is for." |
| 12:58 | <annevk> | Philip`, fair enough |
| 12:59 | <hsivonen> | Lachy: do you happen to have data about Firefox 2 installed base shrinkage rate? |
| 13:00 | <hsivonen> | I wonder how automatically offered updates on Windows&Mac and EOL of certain Linux distros are going to affect Firefox 2 to 3 update rate |
| 13:01 | <Hixie> | nn |
| 13:01 | <hsivonen> | nn |
| 13:14 | <takkaria> | grr at that twitter update |
| 13:14 | <takkaria> | "begs the question" != "asks the question" |
| 13:14 | <takkaria> | or implies the question |
| 13:16 | <Philip`> | Just read it as an abbreviation of "begs us to ask the question", which happens to be abbreviated to the same words as a different phrase, and then you won't have to get angry about it any more :-) |
| 13:16 | <takkaria> | only if you agree to not be pedantic too. :) |
| 13:19 | <hsivonen> | is begging the question ever used properly where "properly" is defined by prescriptive English pedants? |
| 13:19 | <Philip`> | I think I've seen it used properly once |
| 13:23 | <takkaria> | yeah |
| 13:23 | <takkaria> | in philosophy it's used consistently corectly |
| 13:23 | <takkaria> | my degree is philosophy and so I see it being used that way a lot |
| 13:23 | <hsivonen> | eww. it's creepy to see my Google Code login on Blogger |
| 13:26 | <takkaria> | I heard someone say a while ago that if you hear anyone use the word "scientific" you should question what exactly it means because it often means nothing at all these days |
| 13:26 | <takkaria> | I wonder what would happen to the Web discourse if you did that with "semantics" |
| 13:26 | <annevk> | yay, Yngve will explain how keygen works |
| 13:26 | <takkaria> | for example, "We can’t expect any single solution we develop right now to solve all imaginable and unimaginable future semantic needs." |
| 13:31 | <annevk> | http://www.rgraph.org/ is using <canvas> but also a lot of type="text/javascript" |
| 13:32 | <annevk> | habits |
| 14:44 | <takkaria> | I really want got write a reply to that ALA article, but it'd be a waste of time really |
| 15:20 | <Philip`> | http://alistapart.com/comments/semanticsinhtml5?page=2#16 - "isn’t HTML5, at its core, simply replacing one fixed set of tags with another? The new tags might be meaningful now, but will they be in 30 years time?" - in 30 years time, we can have HTML9 with a new fixed set of tags that are meaningful then, so that doesn't seem a problem |
| 15:31 | <annevk> | and more "Boy it’d be great if this “XML” language existed along with support for “DTDs”..." in the comments there |
| 15:32 | <annevk> | I somehow didn't realize the rest of the world hasn't moved on yet |
| 15:34 | <hsivonen> | http://dret.net/netdret/docs/wilde-cacm2008-xml-fever.html#self-description-delusion |
| 15:34 | <hsivonen> | "Recovery from self-description delusion can take a great deal of personal commitment and effort." |
| 15:44 | <gsnedders> | Hixie: I've always erred away from writing such a script, simply because the RFC format is so strict and has some many minor little rules that have to be got correctly |
| 15:46 | <gsnedders> | Hixie: There are also all ready tools that do this. Ask DanC. |
| 15:52 | <gsnedders> | Hixie: And I have exams starting in 8 days, so no. |
| 15:52 | <gsnedders> | (This is the "I'm not your bitch" statement.) |
| 16:00 | <annevk> | hsivonen, lovely |
| 16:01 | <annevk> | hsivonen, the funny thing is that e.g. http://alistapart.com/comments/semanticsinhtml5?page=2#20 does realize it's problematic but is still arguing for it |
| 16:07 | <hsivonen> | also, the usual Cascading Semantic Sheets are proposed: http://alistapart.com/comments/semanticsinhtml5?page=2#16 |
| 16:13 | <annevk> | hsivonen, apparently the RDF community is seriously discussing that |
| 16:13 | <annevk> | hsivonen, forgot where though |
| 16:14 | <annevk> | slightly sad that I said on my blog I'd lost interest in pointing out who's wrong on the internet; I prolly haven't, just moved the ranting to IRC :) |
| 16:16 | <hsivonen> | ranting on IRC or twitter is easier than writing coherent articles |
| 16:25 | <olliej> | annevk: heheh |
| 16:29 | <annevk> | http://twitter.com/clintonskakun/statuses/1099669102 |
| 16:50 | BenMillard | is now reading the ALA article. |
| 16:50 | BenMillard | thought nothing interesting had been written on ALA for about 5 years. |
| 16:51 | <BenMillard> | "We have simply run out of HTML elements and attributes with which to mark up more richly semantic documents." Actually, I think the problem is people forgetting how many there are and not being imaginative enough about how they can be used and combined. |
| 16:51 | <BenMillard> | "It’s pseudo semantic markup at best." that's all it needs to be... |
| 16:52 | <BenMillard> | ROFL @ linking to some random article about "HTML semantics" instead of the actual index pages for HTML 4.01 |
| 16:53 | <BenMillard> | "We need mechanisms in HTML that clearly and unambiguously enable developers to add richer, more meaningful semantics—not pseudo semantics—to their markup." for what purpose? |
| 16:56 | <BenMillard> | "Given the half-life of Internet Explorer, we can predict that most users will be using IE6 or IE7 even several years from now." Nice way to describe it, in terms of a "half-life". |
| 16:58 | <BenMillard> | "[...] the need for greater semantic capability in HTML [...] more semantic goodness [...] we still won’t have solved the problem." So, erm, just what is that problem? |
| 16:58 | <Dashiva> | The semantic web |
| 16:59 | <BenMillard> | I thought the semantic web amounted to publishing databases on the web so people could merge them? |
| 16:59 | <BenMillard> | and that the document web we have today would not be affected |
| 17:01 | <BenMillard> | "[...] the extremely important task of providing a mechanism for semantic richness in HTML [...]" But it just said semantic richness doesn't solve the problem? |
| 17:02 | <BenMillard> | "[...] we’ve been using class and id attributes as mechanisms to extend the semantics of HTML." and yet the article claims several times there's no mechanism to extend HTML |
| 17:03 | <BenMillard> | "Let’s invent a new attribute. [...] we can add a class value to the element for styling." So much for HTML being efficient! |
| 17:04 | <BenMillard> | "Instead of new elements, HTML 5 should adopt a number of new attributes." Is this part requesting for the data-* attributes which are already present? |
| 17:05 | <Philip`> | BenMillard: (I think it's requesting attributes where there is a global shared understanding of what their values mean, so that's not data-*) |
| 17:05 | <Philip`> | (though I don't know where that shared understanding is going to come from) |
| 17:05 | <BenMillard> | ah |
| 17:05 | <Philip`> | (Oh, I know, it comes from DTDs!) |
| 17:05 | <BenMillard> | :P |
| 17:06 | <BenMillard> | oh, the author owns microformatique.com so those are just vanity links |
| 17:07 | <BenMillard> | Philip`, he likens these new attributes to class so maybe their meaning is not expected to be understood by UAs? Perhaps it's just to give authors a warm, fuzzy feeling when they View Source? |
| 17:07 | <Dashiva> | Like <invisible></invisible> to hide source? |
| 17:07 | <Philip`> | BenMillard: I can't really tell, given that his only proposed usage of the attributes is to make divs with a defined structure have bold text |
| 17:10 | <Philip`> | The only real problem with adding new elements to HTML is the block/inline parsing differences, so what HTML5 should say is that if the MD5 of an unknown tag name is an even number then it's parsed as a block element, else it's inline |
| 17:11 | <Philip`> | Then, when HTML6 is adding new elements, it can just be careful to choose element names that have the right MD5 |
| 17:11 | <BenMillard> | "[...] could be used to markup the rhetorical nature of a document." hmm, I wonder why that would be a useful expenditure of time+money at the web scale |
| 17:11 | <Philip`> | which isn't really a problem because there's plenty of synonyms |
| 17:11 | <BenMillard> | Philip`, I kind of hoped unknown elements would work like <ins> and <del>, adapting to their context, but it seems that won't happen |
| 17:12 | <Dashiva> | Philip`: Have you checked if that works for all existing elements? |
| 17:14 | <Philip`> | Dashiva: It works perfectly for all existing elements, because the new rules will only apply to unknown elements |
| 17:14 | <BenMillard> | the article brings up the issue of timestamps in title attributes as a reason to introduce a new attribute, when HTML5 already introduced <time datetime>: http://www.whatwg.org/specs/web-apps/current-work/multipage/text-level-semantics.html#the-time-element |
| 17:16 | <Dashiva> | Philip`: Then it's a silly rule, because it restricts choosing the most awesome tag name for a temporary short-term gain |
| 17:19 | <Philip`> | Dashiva: How dare you call it a silly rule :-( |
| 17:19 | <Philip`> | It seems perfectly sensible to me |
| 17:21 | <Dashiva> | Philip`: You should create a new hashing function that works on all existing elements, that way new UAs can implement only the hashing and leave out arbitrary specific block/inline designations |
| 17:23 | <Philip`> | Dashiva: That should be easy - just use something like gperf |
| 17:23 | <Dashiva> | I want it on my desk tomorrow |
| 17:25 | <Philip`> | The problem with that is that I'm too lazy |
| 17:34 | <gsnedders> | Opinion needed: I have an English dissertation first draft due Thursday, and physics and computing prelims next Wednesday. What school work do I do now? |
| 17:34 | <jcranmer> | gsnedders: obviously, HTML 5 work |
| 17:34 | <Philip`> | I suggest you do none of it, and chat on IRC/Twitter/etc instead |
| 17:34 | gsnedders | was somewhat tempted by those two answers |
| 17:35 | <Philip`> | By the way, you should not take any advice on this channel seriously |
| 17:35 | <gsnedders> | But also by working on a spec I started on yesterday. |
| 17:35 | gsnedders | doesn't take that statement by Philip` seriously :)| |
| 17:36 | <jcranmer> | gsnedders: just say the computer crashed and destroyed your work |
| 17:36 | <jcranmer> | and you had to spend the next five hours stopping it from crashing ever again |
| 17:36 | <BenMillard> | gsnedders, what are the prelims? |
| 17:36 | <BenMillard> | gsnedders, I suggest dropping any expendable work and getting the essential work done first. |
| 17:36 | <gsnedders> | BenMillard: http://en.wikipedia.org/wiki/Prelims |
| 17:37 | <jcranmer> | since it's in this channel |
| 17:37 | <jcranmer> | BenMillard can't be giving serious advice |
| 17:37 | <jcranmer> | so do the opposite of what he says |
| 17:37 | <gsnedders> | BenMillard: All of this is essential. |
| 17:37 | <gsnedders> | BenMillard: See the "Singapore and Scotland" section of that article |
| 17:38 | <BenMillard> | gsnedders, looks like prelims are the more essential thing, since a "first draft" implies there will be further opportunities to submit the work. |
| 17:38 | <BenMillard> | alternative, just do as much as possible instead of debating it here :) |
| 17:38 | <gsnedders> | But the English is worth 40% of the final grade. |
| 17:39 | <gsnedders> | BenMillard: Nah. |
| 17:39 | <BenMillard> | well, the English is due much sooner so you could work your arse off on that now |
| 17:40 | gsnedders | notes he has actually done more of it than most of his class |
| 17:40 | <BenMillard> | then work your arse off preparing for prelims after this Thrusday |
| 17:40 | <BenMillard> | but the important thing is to do the actual work instead of talking about doing the actual work...you aren't W3C! |
| 17:40 | <Philip`> | gsnedders: If the only result of failing to hand in the English dissertation is that you get told off and have to hand it in later, whereas failing to prepare for practice exams means that you will have forever wasted that opportunity for practice, then it sounds sensible to concentrate on the latter |
| 17:41 | <gsnedders> | Philip`: It would however mean I have less time to do the English, which in the long run is more important than prelims whose results will never (unless I choose to) leave the school |
| 17:42 | <Philip`> | gsnedders: That depends on how far past the deadline you can work on the English |
| 17:42 | gsnedders | sighs |
| 17:42 | <gsnedders> | Life is fun. |
| 17:43 | <BenMillard> | gsnedders, if you can't make up your mind them flip a coin once and do that. |
| 17:43 | <BenMillard> | *then |
| 17:43 | <gsnedders> | That seems too arbitrary :P |
| 17:43 | gsnedders | is now reading RFC2616 |
| 17:44 | <BenMillard> | procrastinating means you get nothing done at all |
| 17:44 | Philip` | is most productive while procrastinating |
| 17:44 | <Philip`> | (Not necessarily productive on the thing I'm meant to be working on, but it's better than nothing) |
| 17:44 | <jcranmer> | gsnedders: rand(1)*2 = 1.9150 |
| 17:44 | <BenMillard> | since both are "essential", making progress on either is worthwhile, so a coin toss as a tie-breaker seems quite reasonable |
| 17:44 | <jcranmer> | do with that as you will |
| 17:45 | <Philip`> | jcranmer: Why not use rand(2)? :-) |
| 17:45 | <jcranmer> | Philip`: that does a 2-by-2 matrix |
| 17:45 | <Philip`> | jcranmer: Oh, crazy language |
| 17:45 | <jcranmer> | Philip`: it's MATLAB :-) |
| 17:45 | <Philip`> | jcranmer: I thought it was Perl, but I can see now how that's clearly a stupid thought |
| 17:49 | <rubys> | if they both are essential, do the shorter task first |
| 17:52 | <BenMillard> | back to timestamps in the title attribute and the ALA suggestion for a new attribute...I just found a topic where I suggest custom attributes for this back in August 2007: http://www.accessifyforum.com/viewtopic.php?p=54804#54804 |
| 17:52 | <BenMillard> | it was shot down by peers of the ALA author...nice to see how times change :) |
| 17:53 | <Philip`> | The computer science approach is to make use of the understanding of real-time OS task scheduling algorithms, so e.g. if it's possible to satisfy your tasks with no more than 100% utilisation then you can use http://en.wikipedia.org/wiki/Earliest_deadline_first_scheduling |
| 17:55 | <BenMillard> | (zcorpan has been suggesting data-dtstart=timestamp and suchlike for about as long, IIRC) |
| 17:56 | <BenMillard> | ALA article says: "[...] a significant amount of work needs to be done to really develop a workable solution." That's already been done and it led to data-*, right? |
| 17:58 | <Philip`> | BenMillard: That's a solution for associating private script-accessible data with elements, not for defining the semantics of elements |
| 17:59 | <Philip`> | e.g. you can't write '<span data-datetime="2009-01-01">a few days ago</span>' and have a browser extension that lets you click dates to add them to your calendar |
| 18:00 | <Philip`> | because that extended browser would be non-conforming to HTML5 |
| 18:00 | <Philip`> | so it's not a solution to that problem |
| 18:01 | <BenMillard> | Philip`, yes, that restriction would need to be lifted or softened. |
| 18:02 | <Philip`> | BenMillard: That restriction is there for a reason, and it can't be lifted without addressing the reason for it |
| 18:02 | <Philip`> | (The reason being that there's no mechanism for people to agree on the meanings of data-* attributes) |
| 18:04 | <BenMillard> | Philip`, I imagine the mechanism would be the same as class values? So things like Microformats would move their class values to the part after the hyphen in data-* and UAs could choose to interpret them natively, or via add-ons. |
| 18:05 | <BenMillard> | they could be managed like rel values using a WHATWG wiki page, too |
| 18:05 | <BenMillard> | (For the logs, data-* attributes are here in the HTML5 spec: http://www.whatwg.org/specs/web-apps/current-work/multipage/dom.html#embedding-custom-non-visible-data) |
| 18:06 | <Philip`> | BenMillard: HTML5 had a mechanism for providing specific meaning to certain class values, and that didn't work well and got ripped out, so it's not a good precedent :-) |
| 18:08 | <BenMillard> | Philip`, I guess something ad-hoc and communally agreed, like the Microformats approach, will be more successful than a central registry, then. |
| 18:08 | <Philip`> | But I suppose what's likely to happen is than a Microformats-like process will emerge to define common meaning to some data-* values, and they'll ignore that it's abusing the spec and that there's no reason to think it's not going to conflict with other people using the same name |
| 18:08 | <BenMillard> | indeed :) |
| 18:08 | <Philip`> | because the validator won't complain and it'll achieve their goals, so it's futile to try to stop them |
| 18:19 | <BenMillard> | ooh, ALA's live preview is nice, except it doesn't do linkification |
| 18:20 | <BenMillard> | hmm, it seems you can't make clickable links in ALA comments :| oh well |
| 18:21 | <Philip`> | Maybe they haven't grasped the concept of "web" and hyperlinking yet |
| 18:26 | <BenMillard> | hmm, commenting takes me to a slightly different URL than clicking the permalink for my comment |
| 18:26 | <BenMillard> | anyway, my comment is here: http://alistapart.com/comments/semanticsinhtml5?page=3#26 |
| 18:26 | <BenMillard> | time for dinner |
| 19:37 | <annevk> | "The problem with namespaces though (and THE reason why they haven’t been picked up) is that DTD is not namespace aware" |
| 19:37 | <annevk> | the ALA comments crack me up |
| 19:40 | <olliej> | annevk: hmm? |
| 19:41 | <olliej> | annevk: (i'm not sufficiently well versed in xml, dtds, or namespaces to know the problem with that comment) |
| 19:43 | <annevk> | olliej, DTDs are dead, namespaces are too complex, XHTML is not supported by IE, the authoritarian tone of the author |
| 19:43 | <annevk> | olliej, it's several layers of awesome :) |
| 19:44 | <annevk> | also things like |
| 19:44 | <annevk> | "I think the main reason why “(X)HTML with a custom DTD” has never appeared to be appealing is because of the DTD switch – custom DTDs (AFAIK) run in quirks mode. This alone is enough reason why no one would use a custom DTD." |
| 19:44 | <olliej> | hehe |
| 19:46 | <hsivonen> | yay for truthiness and resident experts |
| 21:04 | gsnedders | needs something like unrealism which is a real world :) |
| 21:04 | <gsnedders> | *word |
| 21:13 | <annevk> | comment 27 gets really funny at the end |
| 21:13 | <annevk> | Philip` might appreciate reading it |
| 21:14 | <annevk> | (the comment from Vasil Rangelov that is, in case they remove earlier comments) |
| 21:23 | <hsivonen> | annevk: the same post also talks about XML Schema and XHTML 1.2... |
| 21:23 | <annevk> | yeah :) |
| 21:25 | <annevk> | it's something I could have written when I didn't really comprehend how Web browsers and the Web evolve |
| 21:27 | <annevk> | oh, seems the Apple keynote started |
| 21:29 | <hsivonen> | annevk: ended, too |
| 21:34 | <annevk> | apart from the DRM-free stuff it doesn't seem very interesting |
| 21:35 | <hsivonen> | agreed |
| 21:36 | <hsivonen> | well, iWork.com is interesting implementation-wise, but I most likely won't use it |
| 21:38 | <Hixie> | gsnedders: sadly none of the existing tools convert html to rdf. Maybe anolis could convert html to one of the other formats that is then converted to rfc? :-) |
| 21:38 | <Hixie> | er, s/rdf/rfc/ |
| 21:38 | <gsnedders> | Hixie: DanC definitely pointed me to one beore |
| 21:38 | <gsnedders> | *before |
| 21:38 | <Hixie> | gsnedders: (no rush, btw, please do pass your exams) |
| 21:39 | <gsnedders> | Hixie: I maintain my position of "ask DanC". |
| 21:42 | <Hixie> | k |
| 21:43 | <roc> | what's interesting about the iWork implementation>? |
| 21:43 | <gsnedders> | Hixie: Also, it really ought to be an entirely separate tool. How xref is done for HTML makes no sense for text/plain. |
| 21:44 | olliej | seconds roc's question |
| 21:45 | <hsivonen> | roc: I haven't looked at the impl, but if they advertise same layout as in Pages, I'm intrested in line breaking fidelity and ligatures |
| 21:46 | <roc> | is it Web-standards based or a plugin thing? |
| 21:47 | <hsivonen> | I don't know. I assumed HTML&CSS without evidence |
| 21:47 | <roc> | then I have no idea how they'd achieve that |
| 21:50 | <Hixie> | gsnedders: fair enough |
| 21:51 | <Philip`> | iwork.com claims to support only Firefox 3 and Safari 3 |
| 21:51 | <Hixie> | gsnedders: on an unrelated note, it would be cool to have a template for anolis a bit like the template bert made, that happens to be w3c pubrules-compliant |
| 21:51 | <roc> | doesn't support IE? |
| 21:51 | <roc> | at all? |
| 21:51 | <gsnedders> | Hixie: I also have no interest in writing such a thing, because xml2rfc fulfils my use-case fine, and I'm not your slave :P |
| 21:51 | <Philip`> | "For best performance and to use all iWork.com features, use one of the following supported browsers: Safari 3 or later; Firefox 3 or later" |
| 21:51 | <Philip`> | says http://publish.iwork.com/shareddocuments/?a=x |
| 21:51 | <roc> | cool |
| 21:52 | <gsnedders> | Hixie: What sort of template? Just a basic outline of a W3C spec? |
| 21:52 | <hsivonen> | it's also interesting how exporting to Word can work in any satisfactory way |
| 21:52 | <hsivonen> | last time I tried exporting from Keynote to PowerPoint (versions ago), the result was useless |
| 21:52 | <gsnedders> | Hixie: Also: email me. |
| 21:53 | <Hixie> | gsnedders: address? |
| 21:53 | <Philip`> | ...but they've gone to effort of doing things like <meta http-equiv="X-UA-Compatible" content="IE=EmulateIE7" /> so I guess they care a bit about IE, and maybe it'll still mostly work in there |
| 21:53 | <gsnedders> | Hixie: me⊙gc |
| 21:55 | <Hixie> | done |
| 21:56 | gsnedders | goes back to writing notes on Nabokov's "The Enchanter" |
| 21:56 | <gsnedders> | (in HTML, of course) |
| 21:57 | <gsnedders> | Hixie: bug me on the 22nd if I haven't done anything by then |
| 22:17 | <gsnedders> | Hixie: Should fictional speech be marked up using q? |
| 22:37 | <Hixie> | gsnedders: "Content inside a q element must be quoted from another source, whose address, if it has one, should be cited in the cite attribute. The source may be fictional, as when quoting characters in a novel or screenplay." |
| 22:38 | <gsnedders> | Oh, I missed that bit. |
| 22:38 | gsnedders | is n00b |
| 22:38 | gsnedders | wonders whether to use ISBN URNs in @cite |
| 22:47 | <hsivonen> | gsnedders: if you click an ISBN URN, will a book come down the wire? if not, then I suggest not. |
| 22:50 | Dashiva | wonders how many people will add postal addresses to @cite |
| 22:56 | <gsnedders> | hsivonen: If you have Firefox with an extension installed, it can take you to a site to buy it :P |
| 22:56 | <gsnedders> | (i.e., no) |
| 22:56 | <Hixie> | hsivonen: http://dret.net/netdret/docs/wilde-cacm2008-xml-fever.html#web-blindness is possibly even more apt currently |
| 22:59 | <Hixie> | that entire document is spot on and priceless |
| 23:05 | <Dashiva> | "Things that are priceless just cost more." |
| 23:15 | <takkaria> | is someone accusing HTML5 of not knowing how to refuse things again? |
| 23:15 | <takkaria> | tsk |
| 23:15 | <Dashiva> | I guess we'll have to remove @headers again |
| 23:16 | <Hixie> | who's complaining we're not saying no enough? |
| 23:17 | <Hixie> | i just said no like a dozen times in the last 24 hours |
| 23:18 | <takkaria> | from Preston on public-html |
| 23:23 | <Hixie> | ah |
| 23:25 | <Hixie> | you know, i look forward to october, and going to last call. people are always ragging on about how html5 has a long timeframe, but hell, it's moving faster than, e.g., DOM3 Events, or many parts of CSS, and at least we have a real timetable that we're sticking to. |
| 23:25 | <Hixie> | reminds me of http://ln.hixie.ch/?start=1172653243&count=1 |
| 23:41 | <webben> | gsnedders: A link to the book's entry in http://www.worldcat.org/ might be more practical. |
| 23:41 | <webben> | or to a particular page in Google Books. |