| 00:08 | <astearns> | hixie: on TextMetrics what about s/ding/t/ ? ascent and descent are the terms I'm used to, and they're shorter |
| 00:09 | <Hixie> | "ascent font bounding box" isn't gramatically correct |
| 00:09 | <Hixie> | (ascent is a noun, not an adjective) |
| 00:10 | <Hixie> | if it was just .ascent, or .fontBoundingBoxAscent, i guess that could work? |
| 00:11 | <astearns> | I like fontBoundingBoxAscent - also fixes the problem of *Box returning a length |
| 00:11 | <TabAtkins> | Good point. |
| 00:11 | <Hixie> | seems reasonable |
| 00:11 | <Hixie> | can you update the wiki accordingly? there's six attributes to change |
| 00:13 | <astearns> | hmm - don't see where to create an account |
| 00:14 | <Hixie> | e-mail address? |
| 00:14 | <Hixie> | (i have to create it for you) |
| 00:14 | <astearns> | stearns⊙ac |
| 00:15 | <Hixie> | check e-mail |
| 00:15 | <Hixie> | k, i think i have gone through all the pending canvas bugs and the pending canvas e-mail |
| 00:15 | <Hixie> | i'll start going through http://wiki.whatwg.org/wiki/Canvas#Proposals and putting all that in the spec friday |
| 00:20 | <astearns> | done |
| 00:33 | <smaug____> | MikeSmith: https://bugzilla.mozilla.org/show_bug.cgi?id=731516#c4 |
| 01:46 | <MikeSmith> | smaug____: thanks |
| 04:59 | <Hixie> | astearns: thanks |
| 05:22 | <Hixie> | what's the best way in JS to get the number out of a string in the format "123 ..." as in, numeric digits, a space character, and then anything else? |
| 05:22 | <Hixie> | split(' ', 1)? |
| 05:22 | <Hixie> | (followed by parseInt()?) |
| 05:30 | <miketaylr> | parseInt("123 blah") should work, no? |
| 05:47 | <Hixie> | yeah i guess that might technically work too |
| 05:55 | <mhausenblas> | morning |
| 05:55 | <mhausenblas> | Hixie, in which time zone are you in, actually? |
| 05:55 | <Hixie> | technically PST |
| 05:55 | <mhausenblas> | technically? :) |
| 05:56 | <Hixie> | i'm much better about it now, but there was a time where my timezone really had no correlation to my online hours |
| 05:56 | <mhausenblas> | he he ... so, you're based in Mountain View? |
| 05:57 | <Hixie> | yes |
| 05:57 | <mhausenblas> | just wondering cause I don't recall a pattern when I see you here (it's 6am here in Galway) |
| 05:58 | <Hixie> | a pattern of what? |
| 05:58 | <mhausenblas> | anyways. back to fiddling with some JS - have fun |
| 05:58 | <mhausenblas> | when you're around |
| 05:58 | <mhausenblas> | online, here, etc. ;) |
| 05:58 | <Hixie> | i'm often around and not actively talking |
| 05:58 | <mhausenblas> | right |
| 05:58 | <Hixie> | and my irc client is always online (it's in norway) |
| 05:58 | <mhausenblas> | but always watching I guess :) |
| 05:58 | <mhausenblas> | ah |
| 05:59 | <mhausenblas> | one last one: where does the .ch in your domain come from |
| 05:59 | <Hixie> | .ch is switzerland |
| 05:59 | <Hixie> | i'm swiss |
| 05:59 | <Hixie> | (half swiss) |
| 05:59 | <mhausenblas> | oh |
| 05:59 | <jamesr> | Hixie: are you swi or iss ? |
| 06:00 | <mhausenblas> | I mean, I knew that .ch is Switzerland, but didn't know you've actually some physical connection into it |
| 06:00 | <Hixie> | jamesr: i cut it horizontally, top half |
| 06:00 | <mhausenblas> | PMSL |
| 06:00 | <Hixie> | mhausenblas: yup born there. not far from cern, actually. left around the time tim was coming up with the idea for the web though. :-) |
| 06:01 | <mhausenblas> | he he - a sign! |
| 06:01 | <mhausenblas> | so you speak French? |
| 06:01 | <Hixie> | about as well as a ten year old, but yeah |
| 06:01 | <mhausenblas> | cool |
| 06:02 | <mhausenblas> | reason being we have three children (youngest 4) and their German is getting worse and worse (we moved over to Ireland in 2009) |
| 06:02 | <mhausenblas> | so I'm always wondering what other people have experienced in that respect as I don't have any other references |
| 06:04 | <mhausenblas> | (well, besides my sister who lives in Norway, she has two children but their language mix is even worse - German, Dutch, English and now Norwegian ;) |
| 06:05 | <Hixie> | dunno what to tell you |
| 06:05 | <Hixie> | it's all about practicing |
| 06:05 | <mhausenblas> | sure |
| 06:05 | <Hixie> | my written french is atrocious |
| 06:06 | <Hixie> | and my vocab completely lacks any computer terminology |
| 06:06 | <Hixie> | but i think if i went to live in a french speaking area i'd get back up to speed relatively quickly |
| 06:06 | <mhausenblas> | fair enough, guess English is appropriate there to fill in the blanks |
| 06:06 | <mhausenblas> | yeah |
| 06:06 | <mhausenblas> | if there is a basis, the missing vocabs come over time |
| 06:06 | <Hixie> | not that that's gonna happen unless the bay area falls into the ocean |
| 06:07 | <mhausenblas> | :) |
| 06:07 | <mhausenblas> | well, maybe you wanna move to CND? |
| 06:07 | <izhak> | Wow, I have just realized, that tokenizer calls parser, not vice versa! |
| 06:08 | <Hixie> | mhausenblas: i'm not leaving the bay area unless i'm driven out by some sort of apocalypse |
| 06:08 | <mhausenblas> | he he, fair enough. actually I envy you a bit - not because of the GOOG position, that is well deserved, but the weather ... |
| 06:08 | <mhausenblas> | it's so friggin damp and windy here |
| 06:09 | <Hixie> | the weather and the people/culture are the main factors for me |
| 06:09 | <mhausenblas> | so, when I'm a grown-up, in some 20y, I also want to work in the SF area ;) |
| 06:09 | <mhausenblas> | right |
| 06:09 | <Hixie> | plus this is where it's at as far as the industry goes |
| 06:09 | <mhausenblas> | yeah |
| 06:12 | <mhausenblas> | OMFG, luv it... http://wulffmorgenthaler.com/2012/03/01/ |
| 06:31 | <kennyluck> | sigh… I hate how centralized the industry is. |
| 06:36 | <zcorpan> | it's nice that the chairs seem to not only not read the content of the draft they're commenting on, but also not read the content of the email they are replying to |
| 06:37 | <zcorpan> | and sam was quick to ask for a different editor, that's nice |
| 07:31 | <zcorpan> | Hixie: did you not finish your sentence here? https://www.w3.org/Bugs/Public/show_bug.cgi?id=16142#c6 |
| 07:31 | <Hixie> | i did not. i thought i had deleted what i had written. |
| 09:17 | <izhak> | As I know data comes in chunks from network. Does html5 speculative parser glues all the chunks to form a whole sequence of PRUnichar*, in mozilla's firefox? |
| 09:17 | <izhak> | I just can't find an entry point to mozilla's html 5 parser, to track the whole story. |
| 10:18 | <hsivonen> | izhak: the parser has the capability to parse up to any PRUnichar boundary and suspend |
| 10:19 | <hsivonen> | izhak: first http://mxr.mozilla.org/mozilla-central/source/parser/html/nsHtml5Parser.cpp#222 is called |
| 10:20 | <hsivonen> | izhak: the caller also obtains a pointer to nsHtml5StreamParser from http://mxr.mozilla.org/mozilla-central/source/parser/html/nsHtml5Parser.cpp#178 |
| 10:20 | <hsivonen> | (created when http://mxr.mozilla.org/mozilla-central/source/parser/html/nsHtml5Parser.cpp#614 got called) |
| 10:21 | <hsivonen> | izhak: then, the networking code calls nsHtml5StreamParser's OnStartRequest, OnDataAvailable (potentially many times) and OnStopRequest |
| 10:23 | <izhak> | hsivonen: Thanks a lot! Your answer is exactly what I needed. |
| 10:27 | <hsivonen> | izhak: what are you trying to do? |
| 10:28 | <izhak> | hsivonen: I'm learning :). |
| 10:28 | <hsivonen> | ok |
| 10:32 | <hsivonen> | izhak: have you seen https://developer.mozilla.org/en/Gecko/HTML_parser_threading ? |
| 10:34 | <izhak> | hsivonen: No. Seems very helpful too. |
| 10:50 | <bga> | small idea. make analog of innerHTML for document fragment. Many frameworks has something like div.innerHTML = html; while(v = div.firstChild) df.appendChild(v). Imho its too common task but this loop do nothing. IE already has applyElement(v, 'inside') and some impemenations support it so its will be good to just standardize this method http://msdn.microsoft.com/en-us/library/ms536341(v=vs.85).aspx |
| 11:23 | <zcorpan> | has ie always had applyElement? never heard of it |
| 11:24 | <zcorpan> | although it sounds nice as a convenience feature |
| 11:24 | <zcorpan> | but it's not like innerHTML at all |
| 14:47 | <hsivonen> | how come Opera is shipping WebGL without either requestAnimationFrame or oRequestAnimationFrame? Am I testing wrong? |
| 16:02 | <zcorpan> | http://dev.w3.org/html5/html4-differences/Overview.src.html#changes-2011-05-25 - review appreciated (non-HTML5 changes are in a comment currently) |
| 16:03 | <zcorpan> | (i haven't updated everything in the rest of the document to take teh changes into account yet) |
| 16:07 | <miketaylr> | window.onerror now supports a forth argument for column position. s/forth/fourth/ |
| 16:11 | <miketaylr> | "Things that use EventTarget now inherits from it instead of using "implements"." s/inherits/inherit/ |
| 16:12 | <miketaylr> | "The crossorigin attribute has been added to img, video and audio to use CORS. CORS" (extra CORS) |
| 16:16 | <zcorpan> | the extra CORS is a reference |
| 16:16 | <zcorpan> | thanks! |
| 16:17 | <miketaylr> | np |
| 17:25 | <MikeSmith> | Ms2ger: should Anne be added to the DOM Parsing and Serialization spec as a co-editor? |
| 17:25 | <MikeSmith> | per https://bitbucket.org/ms2ger/dom-parsing-and-serialization/changesets |
| 18:04 | <bga> | http://www.google.com/patents/US6732330?dq=6,405,366 |
| 18:08 | <AryehGregor> | What's the right way in CSS-ese to say "if <length> is zero or negative, treat it as a parse error"? |
| 18:08 | <AryehGregor> | Oh, I see. |
| 18:08 | <AryehGregor> | "Properties may restrict the number value to some range. If the value is outside the allowed range, the declaration is invalid and must be ignored. For unrestricted values, UAs must support at least up to ±230; unsupported values must be clamped to the closest supported value." |
| 18:11 | <Ms2ger> | 2^30, I hope? |
| 18:12 | <AryehGregor> | No, it's 230. Values and Units is being very conservative about hardware limitations. Everything has to fit in nine bits with breathing room. |
| 18:12 | <AryehGregor> | (also, isn't it annoying how superscript/subscript doesn't copy-paste properly from HTML to plaintext?) |
| 18:13 | <dglazkov> | good morning, Whatwg! |
| 18:14 | <Ms2ger> | Good, dglazkov |
| 18:14 | <dglazkov> | :) |
| 18:14 | <bga> | i wonder if somebody already patented css vars :/ |
| 18:14 | <TabAtkins_> | Good question. The spec counts as prior art, at least. |
| 18:15 | <TabAtkins_> | I should probably submit a patent request to our legal team, though, just so we can be in the clear (and obviously offer it under the standard free w3c license). |
| 18:17 | <dglazkov> | how do you call an untrue statement in an invention disclosure? |
| 18:17 | <dglazkov> | ... |
| 18:17 | <dglazkov> | a patent lie! |
| 18:18 | <TabAtkins_> | ... |
| 18:18 | <Ms2ger> | ... |
| 18:19 | <dglazkov> | Also: battle-scarred HTML badger: <p>.</p> |
| 18:42 | <AryehGregor> | What's the policy for adding new content attributes for new event types? |
| 18:43 | <AryehGregor> | There are attributes like onvolumechange in HTML. |
| 18:43 | <AryehGregor> | So if adding new content attributes fell out of favor, it must have been relatively recent. |
| 18:43 | <Ms2ger> | It fell out of favour? |
| 18:43 | <TabAtkins> | I think we just add all of them. |
| 18:43 | <AryehGregor> | Did it? |
| 18:44 | <AryehGregor> | There are no content attributes currently specced for transition and animation events. |
| 18:44 | <AryehGregor> | So the question is, should there be? |
| 18:44 | <TabAtkins> | I think so, yes. |
| 18:44 | AryehGregor | also thinks so, but wants more opinions before he posts to www-style |
| 18:44 | <TabAtkins> | Any event which can have an element as a target should have a corresponding content attribute. |
| 18:45 | <AryehGregor> | There's an onaddtrack IDL attribute on TextTrackList, so it must still be in vogue to add them. |
| 18:46 | <TabAtkins> | I presume it was just a consequence of this being a CSS spec and so the editors forgot. |
| 18:46 | <TabAtkins> | As opposed to all the things that get dumped into HTML or DOM. |
| 20:03 | <Hixie> | AryehGregor: what's the use case for animation event handler content attributes? |
| 20:04 | <Ms2ger> | What's the use case for onvolumechange? :) |
| 20:04 | <AryehGregor> | Hixie, the only argument I see would be consistency. |
| 20:04 | <Hixie> | Ms2ger: updating the volume ui |
| 20:04 | <Hixie> | AryehGregor: k |
| 20:04 | <AryehGregor> | I don't see a specific use-case. |
| 20:04 | <Hixie> | there's plenty of events that don't have them |
| 20:05 | <AryehGregor> | I mean, I don't know offhand what a use-case is for animation events at all. |
| 20:05 | <Hixie> | and the implementation cost in some implementations is non-zero (depends on whether they implement them all individually or have some sort of hash for the ones that are set) |
| 20:06 | <Ms2ger> | It's non-zero for Gecko, but smaller than epsilon, I'd say |
| 20:12 | <TabAtkins> | Ms2ger: Are you computing with the hyperreals or the surreals? |
| 20:12 | <Ms2ger> | Gecko, so must be surreal |
| 20:13 | <TabAtkins> | Excellent, the surreals are much more interesting anyway. |
| 20:13 | <gsnedders> | Reals are so moring. |
| 20:13 | <gsnedders> | morning? boring. |
| 20:13 | <gsnedders> | Been listening to dglazkov too much. |
| 20:14 | <Ms2ger> | Not even algebraically closed |
| 20:14 | <TabAtkins> | The surreals aren't algebraicly closed either. |
| 20:14 | <Ms2ger> | Bah |
| 20:32 | <AryehGregor> | The surcomplexes are, though, right? |
| 20:33 | <TabAtkins> | Are the surcomplexes defined? |
| 20:33 | <Ms2ger> | ow |
| 20:34 | <Ms2ger> | They are now, that is |
| 20:34 | <TabAtkins> | Ah, indeed. |
| 20:35 | <AryehGregor> | Of course, the fact that the reals aren't algebraically closed is most of what makes them *interesting*. |
| 20:35 | <AryehGregor> | The complex numbers are far simpler. |
| 20:35 | <AryehGregor> | . . . it's kind of a misnomer. |
| 20:35 | <TabAtkins> | Hm? The fact that there is no solution to sqrt(-1) in the reals is what makes them interesting? |
| 20:35 | <Ms2ger> | You say interesting, I say annoying |
| 20:35 | <AryehGregor> | Real analysis is crazy more complicated than complex analysis, yeah. |
| 20:35 | <Ms2ger> | tomato, tomato |
| 20:35 | <TabAtkins> | They're called complex because they are *a* complex (of two numbers). |
| 20:36 | <AryehGregor> | Granted, I don't think the complication makes them more interesting. |
| 20:36 | <AryehGregor> | I also think reals are much more annoying. |
| 20:36 | <AryehGregor> | Things are so much easier in complex-land. |
| 20:37 | <AryehGregor> | All nonconstant polynomials have a zero. All differentiable functions are analytic. All analytic functions have a ludicrously large set of very powerful properties. |
| 20:37 | <AryehGregor> | Witness: Picard's great theorem. W. T. F. |
| 20:38 | <AryehGregor> | Or even his little theorem. |
| 20:38 | <AryehGregor> | "All entire functions are onto except possibly for one point." |
| 20:38 | <Ms2ger> | Also: well-order of the reals, wut |
| 20:38 | <AryehGregor> | You can well-order the complex numbers if and only if you can well-order the reals, no? |
| 20:39 | <TabAtkins> | I thought the reals were trivially well-orderable? |
| 20:39 | <TabAtkins> | Just on the standard < relation? |
| 20:39 | <AryehGregor> | Only if you assume full choice. |
| 20:39 | <AryehGregor> | That's not a well-ordering. |
| 20:39 | <Ms2ger> | Not trivially |
| 20:39 | <AryehGregor> | Well-ordering means every set has a least element. |
| 20:40 | <AryehGregor> | Every nonempty subset, to be pedantic. |
| 20:40 | <TabAtkins> | Gah, duh. |
| 20:40 | <TabAtkins> | Sorry. |
| 20:40 | <AryehGregor> | Like the natural numbers. |
| 20:40 | <Ms2ger> | You apparently can't even give a constructive definition of the well-order of the reals |
| 20:40 | <AryehGregor> | Countable choice isn't enough to well-order the reals. |
| 20:40 | <AryehGregor> | Well, since it requires full choice, yeah, constructive definitions are probably not going to happen. |
| 20:41 | <AryehGregor> | Nor an additive nonlinear continuous function, or the Banach-Tarski paradox, or any of that good stuff. |
| 20:41 | <AryehGregor> | s/good/evil but necessary if you want any algebra to work/ |
| 20:41 | <AryehGregor> | Analysis could get by just fine with countable choice, but algebra would kind of be dead. |
| 20:42 | <AryehGregor> | I mean, countable choice means that uncountable vector spaces need not have a basis. Bad. |
| 20:42 | <Philip`> | I'm sure most people in the world could get on just fine if algebra was dead |
| 20:42 | <AryehGregor> | (more precisely, vector spaces would only need have a basis if they have a countable spanning set, I guess) |
| 20:42 | <AryehGregor> | . . . well, yes, although I've heard some lame people use it for physics and cryptography and stuff instead of recreation. |
| 20:43 | AryehGregor | would like to mentally rehearse the proof of the equivalence of the axiom of choice, Zorn's lemma, the well-ordering principle, and the fact that every vector space has a basis, but decides that returning to billable work is a better idea |
| 20:44 | TabAtkins | agrees that rehearsing proofs is a fun way to spend one's time. |
| 20:44 | <TabAtkins> | I used to put myself to sleep by rehearsing the reasoning behind efficient prime searches. |
| 20:46 | <AryehGregor> | Which efficient prime searches, in particular? |
| 20:46 | <TabAtkins> | Just the basic techniques to speed up a linear search. |
| 20:46 | <TabAtkins> | This was back in middle school. |
| 20:46 | <AryehGregor> | Like the sieve of Eratosthenes? |
| 20:46 | <AryehGregor> | That's the best I know, but I have some vague idea it's not considered efficient by fancy modern standards. |
| 20:47 | <TabAtkins> | Yeah, like that. |
| 20:47 | <gsnedders> | Heh — that came up in my maths lecture today. |
| 20:47 | <TabAtkins> | I didn't know fancy modern techniques back in the day. |
| 20:47 | <gsnedders> | (Both the Sieve of Eratosthenes, and other techniques) |
| 20:47 | <AryehGregor> | Does anyone have any clever ideas of how to deal with this? https://www.w3.org/Bugs/Public/show_bug.cgi?id=15709 |
| 20:47 | <AryehGregor> | No CSS specs specify anything about rounding right now, but it's kind of hard to test programmatic APIs in that case. |
| 20:48 | <AryehGregor> | Also, reftests implicitly test rounding in some cases anyway. |
| 20:48 | <TabAtkins> | But I knew why checking only things that were 1 or 5 mod 6 was efficient and hwo to improve it, or why searching for factors only up to the sqrt of your target was okay. |
| 20:49 | gsnedders | was given an unconvincing proof of that earlier today… |
| 20:50 | <TabAtkins> | Unconvincing proof of which? |
| 20:50 | <gsnedders> | I should probably look at it again, and probably realize I'm just being silly |
| 20:50 | <gsnedders> | 1 or 5 mod 6 |
| 20:50 | <TabAtkins> | 0, 2, 3, and 4 all imply the number is divisible by 2 or 3. |
| 20:51 | <gsnedders> | Yes, but the existance of divisors greater than 6? |
| 20:51 | <TabAtkins> | Huh? |
| 20:51 | <TabAtkins> | It *only* gives a guarantee that the number is/isn't divisible by 2 or 3. |
| 20:51 | <TabAtkins> | It says nothing about any other divisors. |
| 20:52 | <TabAtkins> | 25, for example, is 1mod6. |
| 20:52 | <gsnedders> | "Every prime p >= 5 is of the form 6k±1" |
| 20:52 | <Philip`> | Sounds like an advancement on simply counting sheep - you collect six sheep at once, fan them out in six different directions, drop two thirds of them into cunningly-placed bottomless pits, and then merge the remaining streams of sheep together before applying a more complex filter and only counting the ones that remain alive at the end, and then you fall asleep |
| 20:52 | <gsnedders> | Somehow in my head I turned that around. |
| 20:52 | <gsnedders> | (i.e., Every pos. int. k has some prime 6k±1) |
| 20:52 | <TabAtkins> | Oh yeah, that's clearly false. |
| 20:53 | <gsnedders> | Exactly. |
| 20:53 | <gsnedders> | I had just read a statement later in the proof that almost implies that, but it's just poor phrasing |
| 20:55 | <gsnedders> | Also: I'm vaguely wondering why I'm not doing joint CS and maths. |
| 20:56 | <gsnedders> | Though the real analysis my gf had for test today just scared me. |
| 20:57 | <TabAtkins> | I'm definitely not a math person. I could do math well enough to minor in it (you got a math minor for free if you did a BS in CompSci at my school), but no more. |
| 20:58 | <AryehGregor> | No one has input on <https://www.w3.org/Bugs/Public/show_bug.cgi?id=15709>;? |
| 20:58 | <AryehGregor> | gsnedders, I'm curious. What was it? |
| 20:59 | <gsnedders> | I could likely manage, but I didn't do the majority of maths courses this year, so that chance has passed. Linguistics and CS it is. |
| 20:59 | <gsnedders> | AryehGregor: Nothing that complex in real analysis terms, but I've never really done any real analysis. |
| 20:59 | <Ms2ger> | Really? |
| 20:59 | <Ms2ger> | (Excuse the pun) |
| 20:59 | <gsnedders> | Well, okay, I've done the basic sort of things. |
| 21:00 | <gsnedders> | But some of the finding limits in test my gf had to do were well beyond me. |
| 21:01 | gsnedders | logs on to uni website to file abscene report and will look up the learning outcomes of that course… |
| 21:01 | AryehGregor | wants an example! |
| 21:01 | <Ms2ger> | ^ |
| 21:02 | <gsnedders> | In particular, students should be able to: deal with implications and equivalences; interpret the negation of a statement involving quantifiers; recognise various methods of proof (direct, contrapositive, counterexample, contradiction, induction); show that a function is bounded/unbounded; show, directly from the definition, that a given number is the limit of a given sequence; evaluate sequence limits using arithmetic and order properties; show th |
| 21:03 | <gsnedders> | (where did that get cut off?) |
| 21:03 | <Ms2ger> | order properties; show t |
| 21:03 | <TabAtkins> | "order properties; show th" |
| 21:03 | <gsnedders> | show that a given sequence is monotonic; investigate sequences defined recursively; use subsequences to establish non-convergence; test series for convergence/divergence; test series for absolute/conditional convergence; determine, directly from the definition, whether a function is continuous; use the sequential characterisation to establish discontinuity;solve problems using the intermediate value and extreme value theorems. |
| 21:03 | <Ms2ger> | Hey, you got an h too? |
| 21:03 | <TabAtkins> | My client is better! |
| 21:03 | <Hixie> | i got an h |
| 21:03 | <Ms2ger> | xchat-- |
| 21:03 | <TabAtkins> | irssi++ |
| 21:04 | <Ms2ger> | That doesn't sound too bad |
| 21:04 | <AryehGregor> | That sounds like a standard advanced calculus course. |
| 21:05 | <gsnedders> | It probably wouldn't seem to bad if I'd done it and been to lectures and stuff :) |
| 21:05 | <Ms2ger> | My first university calculus course was basically building up real calculus from scratch, while proving everything along the way |
| 21:05 | Ms2ger | didn't like it either |
| 21:06 | <gsnedders> | (My gf is a chem. student, FWIW) |
| 21:07 | <gsnedders> | (Originally a CS student, but struggled having done far less programming than most people in the course) |
| 21:11 | <gsnedders> | I'd claim she's better than me at maths, and she'd claim the opposite. Nevertheless, both perfectly capable of it. |
| 21:11 | <Ms2ger> | Heh |
| 21:12 | <gsnedders> | Ms2ger: First year maths courses here are quite easy — and huge numbers take them, because they're required for practically all subjects in the faculty of sciene. |
| 21:12 | <Ms2ger> | Mm |
| 21:12 | <gsnedders> | (Actually, for most subjects they're not compulsary, just "strongly recommended", and not doing them massively limits course choice in later years) |
| 21:13 | <Ms2ger> | Everyone here gets their own math courses |
| 21:13 | <gsnedders> | Though the perception of how easy they are is likely down to the fact that traditionally in Scotland you go to uni after five years at secondary school — one fewer than in England, but with one more year of the degree. |
| 21:14 | <gsnedders> | Most people nowadays do six years of secondary school in Scotland, to the standard needed for English universities, and have done a lot of the first year courses for science subjects. |
| 21:14 | <gsnedders> | (Arts subjects tend to be a far higher level than even sixth year of school courses, so it makes little difference there.) |
| 21:15 | <gsnedders> | I did an unusually high number of subjects of the highest level at school — trying to get into Cambridge — which probably is a lot of the reason why I found first year quite so easy. |
| 21:15 | <gsnedders> | My gf only did five years of secondary school (which admittedly I too almost did), so actually learnt quite a bit in first year. |
| 21:19 | <gsnedders> | (I turned up to a single maths lecture a week — to hand in compulsary homework, and learnt almost nothing through it. She went to all five lectures a week and learnt a lot.) |
| 21:20 | <gsnedders> | (I would've gone straight into second year had it not for doing joint linguistics and CS — the linguistics side dictated I had to do first year) |
| 21:20 | <Ms2ger> | Yay, linguistics |
| 21:26 | <kennyluck> | Speaking of math, I really hope some classic textbooks that are full of definitions such as Hartshone's algebraic geometry can be turned into something as readable as the HTML LS. |
| 21:27 | <gsnedders> | HTML LS? |
| 21:27 | <gsnedders> | Living Standard? |
| 21:27 | <kennyluck> | yeah |
| 21:28 | <kennyluck> | xrefs are really helpful. |
| 21:28 | <Philip`> | It's a shame nobody has figured out a decent way to do maths in HTML yet |
| 21:29 | <Philip`> | which is presumably limiting for maths textbooks |
| 21:29 | <jgraham> | Is MathML decent? |
| 21:29 | <jgraham> | Not hand authoring it obviously |
| 21:29 | <jgraham> | That's like slow suicide |
| 21:30 | <Philip`> | Hand authoring seems like an important consideration to me |
| 21:30 | <Philip`> | but maybe people in real life write maths textbooks in Word instead of LaTeX? |
| 21:31 | <Ms2ger> | Heresy! |
| 21:31 | <jgraham> | It might depend what you consider "real life" |
| 21:31 | <Ms2ger> | Outside of academia? |
| 21:31 | <Ms2ger> | Probably so |
| 21:31 | <jgraham> | Probably people who are university professors writing undergrad level textbooks use LaTex |
| 21:32 | <jgraham> | But doing some-TeX-like-dialect-to-MathML is a problem that has been solved many times |
| 21:32 | <jgraham> | So probably anyone writing that kind of book would start by solving it again, differently |
| 21:33 | <Ms2ger> | It's somewhat annoying, I now cringe when I have to study from non-LaTeXed texts |
| 21:35 | <Philip`> | jgraham: "real life" is what people other than me experience |
| 21:38 | <jgraham> | In real life one learns the way to the supermarket in < 4 years, perhaps. I'm not sure how that affects their choice of typesetting software, but does make it easier to get milk. |
| 21:38 | <jgraham> | s/their/one's/ |
| 21:40 | <Ms2ger> | Relevant: http://xkcd.com/1015/ |
| 22:45 | <hober> | abarth (and whoever else cares about registerProtocolHandler): any thoughts on http://www.w3.org/html/wg/wiki/User:Eoconnor/ISSUE-189 before I send it to public-html? |
| 22:46 | <abarth> | reading |
| 22:47 | <abarth> | looks reasonable |
| 22:47 | <hober> | ok, thanks |
| 22:50 | <Velmont> | Ah nice. -- Although I hoped it was a reply to the "deprecate register*handler in favour of webintents" |
| 22:51 | <Velmont> | I'd really like that to read, "deprecate webintents in favour of extending registerprotocolĥandler" instead. Which IMHO makes more sense. |
| 22:52 | <hober> | heh |
| 22:57 | <othermaciej> | how do registerProtocolHandler and Web Intents differ in their purpose? |
| 23:00 | <othermaciej> | hober: you could add an example of a specific hypothetical future use case for a web+ URI, that would be a stronger rationale than just citing the general desire to be upward compatible |
| 23:06 | <Velmont> | othermaciej: WebIntents lets you do a "share" intent, -- so that you don't hardcode Facebook. Nice for me since I don't use Facebook but rather identi.ca. So it's doing verbs and stuff. Lets the browser have some UI to delegate your wanted action to your service. |
| 23:07 | <hober> | othermaciej: ok, will do |
| 23:07 | <Velmont> | othermaciej: So registerprotocolhandler could do the same with e.g. a special intent+share:// protocol or something like that. |
| 23:07 | <Velmont> | Instead of NIH-ing a totally new system. |
| 23:07 | <othermaciej> | Velmont: it seems like one could imagine registering a web+share: protocol instead |
| 23:07 | <Velmont> | othermaciej: Ah, but the problem is ofc that noone will really use that. |
| 23:07 | <othermaciej> | I guess web intents can be (as currently proposed) registered declaratively |
| 23:08 | <Velmont> | othermaciej: And browsers will not implement it. |
| 23:08 | <Velmont> | If I like blog about it and put it on my home page. |
| 23:08 | <Velmont> | It kinda needs to have some weight. |
| 23:13 | <Hixie> | we should make web intents and the handler methods just be one feature |
| 23:13 | <Hixie> | that does declarative and imperative and so on all together |
| 23:13 | <Hixie> | no need for two features |
| 23:15 | <Velmont> | Yep. |
| 23:16 | <zewt> | heh, even after a restart, firefox is still in this weird "everything gets a text cursor" mode |
| 23:16 | <zewt> | as if all text on every page is an uneditable text box |
| 23:18 | <Hixie> | zewt: press f7 |
| 23:18 | <zewt> | ... f7? |
| 23:18 | <Velmont> | caret mode browsing? |
| 23:18 | <TabAtkins> | You're in the caret-navigation mode. |
| 23:19 | <TabAtkins> | It's very confusing when it happens. |
| 23:19 | <zewt> | what a braindamaged feature |
| 23:19 | <TabAtkins> | Took my wife a week to figure out how to back out of it last time it happened. |
| 23:19 | <Hixie> | it's quite an important feature, for certain users |
| 23:19 | <Hixie> | it does give you a prompt the first time you invoke it |
| 23:19 | <zewt> | sometimes I seriously wonder if anyone's at the wheel with FF UX |
| 23:19 | <Velmont> | I was also baffeled once it happened. Took me a long time as well. |
| 23:19 | <TabAtkins> | Hixie: That doesn't help when it was triggered by your cat while you weren't looking at the computer. |
| 23:20 | <zewt> | i definitely havn't seen that prompt before, but I see the default is "yes", so it's probably not improbable to fat-finger enter and never realize there was a dialog (or see it flash and go "well, I wonder what that was") |
| 23:21 | <zewt> | my favorite is when programs steal focus with a dialog that accepts space to select "yes" to something, which invariably happens while I'm typing |
| 23:21 | <Philip`> | Maybe it should pop up a "Did you know you're in caret navigation mode?" message every 2^n minutes, to give you chances to get out of it |
| 23:22 | <Velmont> | Philip`: Along with a "never show this again" checkbox so that the people who use it don't go crazy? |
| 23:23 | <zewt> | well, for starters the default action on the initial confirmation should be no |
| 23:23 | <TabAtkins> | Hm. I wonder where I can find a font that doesn't have the necessary metrics to extract an x-height from? |
| 23:23 | <Philip`> | Velmont: No, because your cat might select that option |
| 23:23 | <Philip`> | Velmont: The exponential decrease in frequency means that long-term users won't get too annoyed |
| 23:24 | <zewt> | heh, well, no, the word would be "infuriated" |
| 23:24 | <Velmont> | Philip`: I get very annoyed about some Opera defaults every time I install it, - like tab focus mode. -- It's very easy to change, -- but it's very very frustrating to always have to change it to be sane. |
| 23:25 | <Velmont> | Sane as in "what a power user would expect". |
| 23:25 | <jamesr__> | Velmont, does Opera have a settings sync feature? |
| 23:26 | <Velmont> | jamesr__: hmmm. working on it at least, I think. Although I could actually figure out how the settings are stored and just put it in my unison configuration file, then it'd be synced to all my computers. |
| 23:27 | <Velmont> | Anyway, -- if I'm irritated by stuff like that, I always think that there's someone else who is as well. |