| 00:10 | <MikeSmith> | Hixie: if/when you have some time for review/sanity-check: |
| 00:10 | <MikeSmith> | http://dev.w3.org/html5/pubnotes/ |
| 00:11 | <Hixie> | jesus |
| 00:12 | <Hixie> | looks fine to me, in my 2 minute scan |
| 00:12 | <Hixie> | and spot checks |
| 00:12 | <Hixie> | i expect will people reasoning for many of the changes |
| 00:13 | <Hixie> | er |
| 00:13 | <Hixie> | data corruption error on output |
| 00:13 | <Hixie> | retrying |
| 00:13 | <jwalden> | will want? |
| 00:13 | <Hixie> | i expect people will want reasoning for many of the changes |
| 00:13 | <Hixie> | which is to say, i think nobody will ever be happy :-) |
| 00:13 | <Hixie> | but wow, that's a big document |
| 00:13 | <Hixie> | did i really do that much in the last 4 months? |
| 00:14 | <MikeSmith> | Hixie: I take "nobody will ever be happy" as a given :) |
| 00:14 | <Hixie> | indeed |
| 00:14 | <jwalden> | MikeSmith: 6.4, target of the event is the window on which it was called, not that window's document |
| 00:14 | <Hixie> | me too :-) |
| 00:15 | <jwalden> | assuming the version diffed here got that change |
| 00:15 | <MikeSmith> | jwalden: thanks, will edit it now |
| 00:16 | <MikeSmith> | jwalden: version diffed there is the current version, up to date with the most recent checkin |
| 00:16 | <jwalden> | okay |
| 00:16 | <jwalden> | it would have, then |
| 00:16 | annevk | deletes www-archive spam from MikeSmith :) |
| 00:16 | <jwalden> | there was a couple-day window where it could have been okay |
| 00:19 | <annevk> | Lachy, standardssuck.com and standardssuck.net now work |
| 00:20 | <Lachy> | LOL |
| 00:22 | <Lachy> | I expected you to just set up a redirect |
| 00:22 | <Lachy> | <meta name="refresh" content="0;url=http://standardssuck.org"/>; should work ;-) |
| 00:23 | <annevk> | Redirect 301 ... too, but that's boring :) |
| 00:23 | <annevk> | (actually, dreamhost provides the Redirect 301 ... option without having to set up some kind of hosting thingie around it) |
| 00:25 | <MikeSmith> | annevk: sorry, probably I should send those to member-only version of www-archive so as to spam fewer people |
| 00:25 | <annevk> | MikeSmith, neh, this is way better |
| 00:25 | <annevk> | we're a public group, the more we can drag into the open the better |
| 00:31 | <annevk> | btw, Ubuntu 8.04 is shipping with Firefox beta 5... isn't that kind of defeating the purpose of having final versions? |
| 00:32 | <Philip`> | Google taught us that we should store our critical data in beta services |
| 00:33 | Philip` | apparently has precisely 16384 messages on "Google(TM) Mail BETA" |
| 00:35 | <takkaria> | annevk: it'll be updated to the final when that comes out |
| 00:36 | <annevk> | takkaria, sure, but users are getting it before "wir" |
| 00:36 | <Philip`> | How quickly do users switch to the latest version of Ubuntu? (i.e. how many will get the pre-release version of Firefox?) |
| 00:37 | <annevk> | through Ubuntu 8.04 |
| 00:37 | Philip` | would assume lots of people migrate immediately when the new version is released, and then it'd quickly drop off until there's just a steady rate of new users |
| 00:38 | <annevk> | all the people getting a new PC with Ubuntu will have it |
| 00:39 | <annevk> | seems people are already discussing this elsewhere: http://ubuntuforums.org/showthread.php?t=734176 |
| 00:40 | <takkaria> | I can't be the only person that finds it funny that people are calling photo-sharing sites like Flickr an edge-case |
| 00:40 | <takkaria> | (and immensely frustrating) |
| 00:40 | <Philip`> | annevk: https://bugzilla.mozilla.org/show_bug.cgi?id=421482#c32 too |
| 00:41 | <Philip`> | takkaria: Flickr is one site out of tens of millions, which seems a bit edgy |
| 00:42 | <Philip`> | and I don't imagine there are hundreds of thousands of other photo sharing sites |
| 00:43 | <annevk> | Philip`, it seems that photo sharing in general is an edge case scenario according to certain people |
| 00:43 | <othermaciej> | Flickr is one very popular site with millions of pages |
| 00:43 | <jgraham> | Photo sharing must be one of the most common ways that most people use the web |
| 00:43 | <annevk> | Though maybe someone should ask to be sure... |
| 00:44 | <othermaciej> | #39 on Alexa top 500 |
| 00:44 | <annevk> | Won't be me though, as I still consider myself lucky that I didn't get any replies to the latest e-mail I threw into that pit |
| 00:44 | <othermaciej> | ImageShack is #34 |
| 00:44 | <jgraham> | it's not just flickr, pretty much all social networking sites have some sort of photo functionality |
| 00:44 | <othermaciej> | photobucket is #26 |
| 00:46 | <jgraham> | Doesn't facebook (#8) have photo sharing? |
| 00:47 | <jgraham> | Also presumably Myspace |
| 00:47 | <takkaria> | yeah, both of those do |
| 00:47 | <Philip`> | I imagine this is the point where trendy people mention the Long Tail - the top few sites are insignicant compared to the hundreds of billions of other web pages, so we shouldn't optimise for the former at the expense of the latter |
| 00:47 | <annevk> | Matt also seems to skew the statistics a bit by including the trivial iconic images Flickr serves into his comparisons... |
| 00:48 | <annevk> | Philip`, is there evidence that the Long Tail doesn't do photo sharing? |
| 00:48 | <jgraham> | (e.g. via gallery and so on) |
| 00:51 | <Philip`> | annevk: I have no idea; but Flickr and ImageShack and Photobucket and Facebook and MySpace aren't counter-evidence since they're not part of the tail at all |
| 00:52 | <othermaciej> | Philip`: that's not what the Long Tail means |
| 00:52 | <othermaciej> | Philip`: that's getting it backwards |
| 00:52 | <othermaciej> | the power law distribution means the top few sites get as much traffic/attention as the whole long tail combined |
| 00:54 | <jgraham> | othermaciej: I think I remember a survey that backed up that hypothesis (although I think it was one of those "ask people what they do" type setups which are horribly biased) |
| 00:54 | <othermaciej> | jgraham: I'm sure anyone with raw traffic stats could prove it, but even just knowing that the distribution is a power law proves it |
| 00:55 | <jgraham> | othermaciej: Yeah I know, but how do you know that the distribution is a power law? |
| 00:56 | <othermaciej> | all available evidence seems consistent with that hypothesis |
| 00:56 | <othermaciej> | it may be that every available sample that has been disclosed is biased |
| 00:57 | <MikeSmith> | hsivonen: I thought hendry may have built his Iris browser from source at George Staikos code.staikos.net, but browsing there now, I don't see any subdirectory for the Iris stuf |
| 00:57 | <jgraham> | I agree if you can actually measure it you can see if it's a power law or not. |
| 00:57 | <othermaciej> | I just had coffee with George Staikos |
| 00:57 | <othermaciej> | I think their WebKit ports are open source but the browser is not |
| 00:58 | <Hixie> | photos are an edge case |
| 00:58 | <Hixie> | haha |
| 00:58 | <Hixie> | that's so funny |
| 00:58 | <othermaciej> | jgraham: power law distribution can be detected from samples, the question is just whether the samples are biased |
| 00:58 | <MikeSmith> | othermaciej: next time you see George, please tell him I said Hi and that I'm very glad I no longer have to be involved with the CABForum :) |
| 00:59 | <jgraham> | The fraction of time people spend on various sites is not obviously a trivial thing to measure |
| 00:59 | <Philip`> | othermaciej: All the samples would be heavily biased towards the head of the distribution, and you'd probably have too little data about the tail to work out the distribution |
| 00:59 | <Philip`> | (Actually, that's probably totally untrue) |
| 00:59 | <MikeSmith> | othermaciej: http://www.torchmobile.com/blog/?p=5 : he says, "Our source code is now imported into the git server at git://code.staikos.net/ alongside the QtWebKit and Torch Mobile Qt WebKit code," |
| 01:02 | <Lachy> | Google Image Search and other similar services is another example |
| 01:03 | <othermaciej> | maybe I am wrong |
| 01:04 | <othermaciej> | Philip`: broad studies of blogs show that, at least there, the power law goes very deep |
| 01:04 | <othermaciej> | the hypothesis that the low end goes off the power law is pretty much eliminated in that case |
| 01:04 | <othermaciej> | it could be that for the web in general it is different but I doubt it |
| 01:09 | <jgraham> | othermaciej: Interestingly there seem to be a lot of people who claim the powerlaw thing is well supported by evidence but the actual evidence is thin on the ground. The only things I could find related to the number of links pointing to different blogs and the pageviews per month on the Sun website |
| 01:10 | <jgraham> | s/is thin/seems to be thin/ |
| 01:10 | <jgraham> | I would be very surprised if the powerlaw thing doesn't hold for the web at large but I'd like to see the data proving it |
| 01:14 | <annevk> | geez, 2 AM |
| 01:14 | <othermaciej> | does alexa distribute raw data for their to sites list? |
| 01:14 | <othermaciej> | if so you could see if it fits |
| 01:14 | <othermaciej> | I have seen technorati's raw data for all the blogs they track and it is a very good fit |
| 01:15 | <annevk> | power laws are everywhere :) |
| 01:16 | MikeSmith | finishes reading Dave Singer <timerange> proposal |
| 01:21 | <jgraham> | othermaciej: I guess if I wanted to be really difficult I would argue that the navigation model for blogs (largely based around links from a pool of other blogs) isn't the same as the navigation model for ebay, amazon and google (based much more around search) |
| 01:21 | <jgraham> | But I don't really want to be difficult and should really sleep instead :) |
| 01:39 | MikeSmith | chuckles at ggaren http://trac.webkit.org/changeset/34091 "Removed terrible lie from ChangeLog" |
| 01:40 | <Dashiva> | Maybe that line itself is a lie, and it actually removed a truth |
| 01:42 | <MikeSmith> | Dashiva: "All Cretans are liars" |
| 01:42 | <Dashiva> | That's what she said |
| 01:42 | <MikeSmith> | heh |
| 01:43 | <MikeSmith> | hmm, opera dev build making the fan on my mikebook run continuously |
| 01:44 | <Hixie> | i don't understand what parse mode webkit and gecko use for innerHTML of <td> |
| 01:44 | <Hixie> | http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C!DOCTYPE%20html%3Ea%3Ccaption%3Eb%3Ctable%3E%3Ctd%3EFAIL%3C%2Ftable%3E%0A%3Cscript%3Edocument.getElementsByTagName('td')%5B0%5D.innerHTML%20%3D%20'a%3Ccaption%3Eb'%3C%2Fscript%3E |
| 01:44 | <Dashiva> | MikeSmith: Check CPU load |
| 01:46 | <MikeSmith> | Dashiva: just up-installed to new build & trying that now |
| 01:49 | <Dashiva> | I have recurring cases of random CPU spiking for seconds or minutes myself, so it could be what you had |
| 01:58 | <MikeSmith> | Cpu(s): 1.0%us, 21.2%sy, 0.0%ni, 76.2%id, 1.3%wa, 0.0%hi, 0.3%si, 0.0%st |
| 01:58 | <MikeSmith> | 8693 mike 20 0 307m 121m 18m S 16.7 24.0 2:23.09 opera |
| 01:58 | <MikeSmith> | which don't look bad |
| 01:58 | <MikeSmith> | yet still it seems to be causing my fan to kick in |
| 02:05 | <Philip`> | 16.7% CPU load seems pretty heavy... |
| 02:23 | <Hixie> | annevk: yt? |
| 02:32 | <Hixie> | anne: did you ever reply to this one? http://lists.w3.org/Archives/Public/public-webapi/2008May/0171.html |
| 03:06 | <Lachy> | Hixie, if an author uses <col irrelevant="">, does that implicitly make all the cells in that column irrelevant too? |
| 04:29 | MikeSmith | ponders the idea of a commit hook that for each change to the spec would find the nearest ancestor <h2> for each part of the change, and record that data in the diff |
| 04:32 | <MikeSmith> | gnu diff as that built-in behavior that for C code, it tells you which function the change occurred in |
| 04:32 | <MikeSmith> | "-p --show-c-function : Show which C function each change is in" |
| 04:33 | <MikeSmith> | extending that to handle looking for <h2> might be something reasonably do-able and useful |
| 04:34 | <MikeSmith> | or <hN> |
| 06:05 | <Hixie> | MikeSmith: i'd be happy to replace the diff i use if you can code that up :-) |
| 06:05 | <Hixie> | Lachy: no |
| 06:20 | MikeSmith | downloads the GNU diff sources to take a look |
| 06:22 | <MikeSmith> | Hixie: I'm also exploring the idea of adding to the W3C CVS webview interface a way to generate HTML diffs between any two arbitrary versions of an HTML page maintained in W3C CVS space |
| 06:22 | <MikeSmith> | that also should be pretty straightforward to set up |
| 06:22 | <MikeSmith> | we have an online HTML Diff interface already: |
| 06:22 | <MikeSmith> | http://www.w3.org/2007/10/htmldiff |
| 06:23 | <MikeSmith> | it'd mostly be a matter of setting up the CVS webview to be able to pass to the URLs like the following |
| 06:23 | <MikeSmith> | http://dev.w3.org/cvsweb/~checkout~/html5/spec/Overview.html?rev=1.877&content-type=text/plain |
| 06:24 | <MikeSmith> | http://dev.w3.org/cvsweb/~checkout~/html5/spec/Overview.html?rev=1.878&content-type=text/plain |
| 06:24 | <MikeSmith> | pairs of revisions |
| 06:24 | <Hixie> | yeah i use that to generate index-diff |
| 07:12 | <MikeSmith> | othermaciej: yt? |
| 07:12 | <othermaciej_> | MikeSmith: yeah, what's up? |
| 07:14 | <MikeSmith> | othermaciej: about http://www.hixie.ch/tests/adhoc/perf/dom/artificial/core/001.html in iPhone |
| 07:14 | <MikeSmith> | Safari on iPhone |
| 07:14 | <MikeSmith> | I guess you know this already |
| 07:15 | <MikeSmith> | that test doesn't seem work unless the count is cranked down a bit |
| 07:16 | <MikeSmith> | e.g.: http://www.pureanarchy.com/hixietest.html |
| 07:17 | <othermaciej_> | MikeSmith: iPhone has a JS execution time limit and I would not be surprised if that test blows through it |
| 07:17 | <MikeSmith> | aha |
| 07:17 | <MikeSmith> | so I tried it on Opera Mobile on my own handset here |
| 07:18 | <MikeSmith> | and it also didn't work, so yeah, I'd guess it's the time limit |
| 07:18 | <MikeSmith> | other kinda weird thing about the test results with the adjusted-count page at URL above |
| 07:19 | <MikeSmith> | is the Index test result |
| 07:20 | <MikeSmith> | which is much slower relative to the other test than it is in Webkit running on my macbook |
| 07:21 | <othermaciej_> | how recent is your build? |
| 07:22 | <othermaciej_> | in Safari 3.1 the Index time is plenty fast to me, even more so in a trunk, but indeed it used to be much slower before 3.1 |
| 07:22 | <MikeSmith> | othermaciej_: my desktop webkit is the latest nightly |
| 07:23 | <othermaciej> | I just tried it in a fresh build of trunk |
| 07:23 | <othermaciej> | MikeSmith: http://paste.lisp.org/display/61193 |
| 07:23 | <othermaciej> | those are my results |
| 07:24 | <MikeSmith> | yeah, those are in line with what I get on my machine |
| 07:24 | <MikeSmith> | what I meant was the results in Safari on the iPhone |
| 07:25 | <MikeSmith> | on iPhone, the Index test is much slower relative to the other tests |
| 07:25 | <MikeSmith> | on desktop, it is one of the fastest tests |
| 07:25 | <othermaciej> | it's probably got something closer to the Safari 3 code |
| 07:25 | <MikeSmith> | on iPhone it is the slowest] |
| 07:26 | <MikeSmith> | OK |
| 07:26 | <othermaciej> | where the index test was slow |
| 07:26 | <othermaciej> | due to a bug in caching for nodeLists |
| 07:26 | <MikeSmith> | ah |
| 09:34 | <annevk> | lol, somebody did indeed bring up all the other sites out there |
| 09:34 | <annevk> | how very predictable |
| 09:35 | <annevk> | Hixie, replied |
| 09:35 | <Hixie> | hm, you rejected it :-( |
| 09:36 | <Hixie> | Sec-Origin would be really cool because if browsers included it with everything, it could be used to reject CSRFs more reliably than Referer |
| 09:37 | <Hixie> | oh well |
| 09:39 | <Hixie> | annevk: what's the status on the header stuff? did you end up making any decisions on that? |
| 09:42 | <annevk> | Hixie, it's easy to change, just comment |
| 09:42 | <Hixie> | i hope adam will |
| 09:42 | <annevk> | Hixie, though shouldn't we just call it Origin then? |
| 09:42 | <Hixie> | (he's got a better handle on the implications) |
| 09:42 | <Hixie> | (or collin, maybe) |
| 09:43 | <Hixie> | i guess Sec-* isn't safe in old XHRs either huh |
| 09:43 | <annevk> | it will be going forward, but isn't now |
| 09:43 | <annevk> | maybe in WebKit it is already :) |
| 09:43 | <Hixie> | that's the main reason to use Sec- |
| 09:44 | <annevk> | yeah |
| 09:44 | <annevk> | Hixie, I think I just want to leave AC as is |
| 09:45 | <annevk> | Hixie, apart from naming changes such as dropping "Access-Control-" from "Access-Control-Origin" |
| 09:45 | <Hixie> | cool |
| 09:45 | <Hixie> | last call time then? |
| 09:45 | <annevk> | for AC, yes, although maybe I should more explitly call out the problems with Access-Control-Policy-Path |
| 09:46 | <annevk> | in the security section for authors |
| 09:46 | <Hixie> | i'd be so happy if you could make the f2f pointless |
| 09:46 | <Hixie> | i don't want to travel :-) |
| 09:48 | <annevk> | i'll reply to my own e-mail |
| 09:48 | <annevk> | for the origin thingie |
| 09:48 | <Hixie> | heh k |
| 09:49 | <othermaciej> | Hixie: what is Sec-Origin and why would it be more reliable than Referer? |
| 09:49 | <Hixie> | origin wouldn't have the path information, so it wouldn't be stripped by https->http posts, for instance |
| 09:50 | <Hixie> | though i guess it could still be stripped by over-zealous privacy advocates |
| 09:50 | <Hixie> | and so you'd still have to assume its absence meant same-origin |
| 09:50 | <Hixie> | <Hixie> and so you'd still have to assume its absence meant same-origin |
| 09:50 | <Hixie> | adam can explain his idea better |
| 09:51 | <othermaciej> | If path info is the only reason to strip Referer then it would be a simple matter to include Referer but remove path info from it |
| 09:51 | <othermaciej> | but I am not sure it is the only reason |
| 09:55 | <annevk> | if that's the case maybe it should take magical values "same-origin" and "non same-origin" so you can at least know what is happening in case the URI is not exposed |
| 09:56 | <annevk> | though maybe middleware will just strip it out |
| 09:56 | <Hixie> | does hsivonen have a page anywhere that exposes his parser? |
| 09:56 | <Hixie> | oh nm |
| 09:56 | <Hixie> | http://parsetree.validator.nu/ |
| 09:57 | <Hixie> | hm |
| 09:57 | <Hixie> | doesn't work |
| 09:57 | <annevk> | he could add <textarea> now I think |
| 09:57 | <Hixie> | http://parsetree.validator.nu/?doc=http%3A%2F%2Fjunkyard.damowmow.com%2F323&submit=Print+Tree |
| 09:59 | <annevk> | hsivonen, ^^ |
| 09:59 | <Hixie> | he has mail about this already |
| 09:59 | <Hixie> | so do you in fact |
| 09:59 | <Hixie> | :-) |
| 10:00 | <annevk> | i'm sort of doubtful we can avoid the F2F btw |
| 10:00 | <annevk> | we'll see how it goes I guess |
| 10:00 | <Hixie> | that's not the right attitude :-) |
| 10:00 | <Hixie> | i thought you thought you were ready for LC months ago but that mozilla was the only thing blocking you? |
| 10:00 | <Hixie> | we've gotten rid of that blocker now |
| 10:01 | <annevk> | we did? |
| 10:02 | <annevk> | hmm, parsetree does work for other stuff, e.g. http://parsetree.validator.nu/?doc=http%3A%2F%2Fannevankesteren.com%2F |
| 10:06 | <Hixie> | mozilla has sent all the feedback they have and will not be blocking XHR/AC on the grounds of cookies being insecure, yes |
| 10:06 | <Hixie> | at least, that's the assurance i have received |
| 10:08 | <annevk> | i think i'll take your IIS and Bjoern's Apache example and put them in the spec |
| 10:09 | <annevk> | maybe example.org yours first :) |
| 10:10 | <Hixie> | what did i use? :-) |
| 10:10 | <annevk> | microsoft.com |
| 10:10 | <Hixie> | hah |
| 10:11 | <Hixie> | i wish example.com and example.org didn't have the same name |
| 10:12 | <annevk> | there's something.invalid |
| 10:12 | <annevk> | wait, something.example |
| 10:12 | <Hixie> | yeah but that looks invalid :-) |
| 10:12 | <Hixie> | and .example looks to long |
| 10:12 | <annevk> | true, they don't give us much choice for describing same-origin and all that |
| 10:13 | <annevk> | (i did use hello-world.example and such fwiw, because I can't be bothered to try setting a new standard there) |
| 10:13 | <Hixie> | heh |
| 10:14 | <annevk> | the reason IE puts <noframes> in <head> btw is because IE ignores </head> |
| 10:17 | <Hixie> | yeah I just put it in head because it was the least change to the spec |
| 10:18 | <Hixie> | the alternative was putting it in between <head> and <body>, and that seemed far, far more complex |
| 10:26 | <annevk> | Hixie, btw, hsivonen also raised the EOF issue and he has a longer list of elements he'd like to see added (besides h4) |
| 10:26 | <Hixie> | if he sent mail, i'm sure i'll get to it |
| 10:27 | <annevk> | the batch processing is over? |
| 10:27 | <Hixie> | i use batch processing when there are enough mail to batch process |
| 10:27 | <Hixie> | and when they are all sent to the same list |
| 10:55 | MikeSmith | would also be really happy to avoid traveling |
| 10:56 | <MikeSmith> | let's have all f2f meetings in Tokyo |
| 10:56 | <MikeSmith> | you can stay in my apartment |
| 10:56 | <MikeSmith> | I have room for 1.5 people |
| 10:56 | <Hixie> | i recommend having the meetings on irc and e-mail |
| 10:56 | <roc> | amen |
| 10:57 | <MikeSmith> | word |
| 11:03 | <gsnedders> | MikeSmith: who you planning on cutting in two? |
| 11:04 | <MikeSmith> | gsnedders: you don't want two of me |
| 11:04 | <MikeSmith> | trust me on that |
| 11:04 | <annevk> | ok, I think AC is ready for LC |
| 11:05 | <annevk> | I think it's better for XHR2 to wait until at least all feedback on XHR1 LC is in |
| 11:05 | <MikeSmith> | annevk: so let's start LC process for AC |
| 11:05 | <MikeSmith> | we will need ArtB |
| 11:06 | <annevk> | I said in "Re: Moving forward with XHR2 and AC" that it is ready for LC |
| 11:06 | <annevk> | should we get ArtB to formally ask the WG? |
| 11:07 | <MikeSmith> | I think we need him to do that |
| 11:08 | <MikeSmith> | likely you are actually more familiar with W3C WG process stuff around this than me |
| 11:08 | <MikeSmith> | but I don't think it can be done without the chair |
| 11:08 | gsnedders | passes MikeSmith a n00b badge |
| 11:09 | <annevk> | MikeSmith, from my recollection we need WG agreement (either implicit or explicit the latter being preferred) |
| 11:10 | MikeSmith | tosses n00b badge into his pile with shitbird and jaggass badges |
| 11:10 | <MikeSmith> | annevk: I guess |
| 11:11 | <MikeSmith> | but getting that is a lot more of potential fubar situation than just getting artb on board |
| 11:12 | <gsnedders> | When does LC on XHR1 end? |
| 11:12 | <gsnedders> | (i.e., by when do I need to get off my lazy ass?) |
| 11:13 | <annevk> | Monday June 2 |
| 11:13 | <gsnedders> | The day of my final exam. Heh. |
| 11:16 | <MikeSmith> | annevk: at this point it's clear that we have to extend the WAF WG charter til end of June at least anyway |
| 11:18 | <MikeSmith> | so getting a WD out for *public* review in that window of time would be great |
| 11:19 | <MikeSmith> | I was going to say that I would honestly not want to make getting that done contingent on consensus of the WG |
| 11:19 | <MikeSmith> | but on second thought |
| 11:19 | <MikeSmith> | .. |
| 11:20 | <MikeSmith> | most of the lets-slow-this-down comments |
| 11:20 | <MikeSmith> | have come from "the public" |
| 11:20 | <MikeSmith> | not from active members of the group |
| 11:21 | <MikeSmith> | e.g., the "I'm commenting on this as an individual, not representing my employer" |
| 11:21 | <MikeSmith> | which is a lie/BS to begin with |
| 11:23 | <MikeSmith> | and then the assorted other people-who-by-very-deliberate-choice-are-not-members-of-the-WG with whatever agenda it is they have |
| 11:23 | <MikeSmith> | anyway, I say too much |
| 11:23 | <MikeSmith> | as usua |
| 11:23 | <MikeSmith> | sual |
| 11:23 | <annevk> | so I just remembered we need to settle on naming of Access-Control-Origin before publication |
| 11:23 | <MikeSmith> | usual |
| 11:23 | <MikeSmith> | annevk: yeah. |
| 11:23 | <annevk> | but hopefully that can be resolved within a few days |
| 11:23 | <MikeSmith> | decide. |
| 11:24 | <MikeSmith> | annevk: resolve it. |
| 11:24 | <annevk> | well, I'd like to hear the opinion of Adam / Collin first |
| 11:24 | <MikeSmith> | annevk: yes |
| 11:24 | <MikeSmith> | they are among the few |
| 11:24 | <MikeSmith> | the very few |
| 11:24 | <annevk> | they're quite awesome from what I've seen :) |
| 11:24 | <MikeSmith> | yep |
| 11:24 | <MikeSmith> | they have earned it |
| 11:24 | <MikeSmith> | through merit |
| 11:25 | <MikeSmith> | they speak rarely |
| 11:25 | <MikeSmith> | but when they do.. |
| 11:25 | <MikeSmith> | they have earned the right to be listened to carefully |
| 11:26 | <Hixie> | annevk: when does XHR1 LC end? |
| 11:27 | <Dashiva> | That's the sage's quandry. The less you speak, the more awed are people when you actually do, but the less often you're able to dispense wisdom :) |
| 11:27 | <Hixie> | adam and collin are awesome |
| 11:27 | <MikeSmith> | Dashiva: actually, speaking ain't worth nearly as much as other things |
| 11:28 | <MikeSmith> | Adam and Collin walk the walk |
| 11:29 | <MikeSmith> | as in (just for one example) coding/patching Webkit code around some of this tuff |
| 11:29 | <MikeSmith> | stuff |
| 11:29 | <annevk> | Hixie, Mon June 2 |
| 11:30 | <Hixie> | is there any feedback you have received on xhr1 and not yet dealt with in xhr2? |
| 11:31 | <annevk> | yes, what to do with send(document) if document can't be serialized for instance |
| 11:31 | <annevk> | well, some people disagree with how it's handled, and apparently it matches neither Firefox or IE, but I think everyone is doing something else... |
| 11:32 | <annevk> | another issue is URI/IRI confusion |
| 11:32 | <annevk> | maybe I should put them in the issue tracker |
| 11:32 | <Hixie> | i recommend dealing with that now then, and then once you've handled all that, pushing to get XHR2 into LC, even if XHR1 hasn't finished LC yet -- then if there are major issues, rerelease XHR2 as a second last call quickly before its first LC is over. |
| 11:32 | <annevk> | that would at least give an overview |
| 11:33 | <Hixie> | that would, i think, be the most efficient way of getting all this to CR |
| 11:33 | <Hixie> | you generally want to avoid putting yourself in a position of having feedback you can't deal with (e.g. by waiting on someone else's input) |
| 11:33 | <MikeSmith> | annevk: fwiw, I think a 2nd LC maybe basically an inevitability anyway |
| 11:34 | <annevk> | Hixie, I agree, but URIs/IRIs are very confusing and it's not at all clear what the right solution is to me :) |
| 11:34 | <annevk> | Hixie, I think I might need the HTML5 URL stuff |
| 11:35 | <Hixie> | read the specs :-) |
| 11:35 | <Hixie> | and test implementations :-) |
| 11:35 | <Hixie> | that's the way to make things clearer :-) |
| 11:36 | MikeSmith | records milestone of checkin number 1700 against HTML5 spec source, notes that at current rate, checkin 2000 will be reached within 2 monhts |
| 11:37 | <Dashiva> | MikeSmith: Have you seen the issues graph? |
| 11:37 | <Hixie> | the number of checkins per day depends more on the latency of the various servers i use and on my mood than on how much work i do |
| 11:37 | <annevk> | MikeSmith, btw, aside from sandboxing and details like document.charset I'd be interested in hearing what html5-diff doesn't cover currently |
| 11:37 | <MikeSmith> | Dashiva: yep |
| 11:38 | <Hixie> | because if the computers aren't working with me, or if i'm being too productive, i batch changes into bigger changesets |
| 11:38 | <Hixie> | (too productive for the spec regen script to keep up, i mean) |
| 11:38 | <Hixie> | annevk: i've no idea :-) |
| 11:38 | <MikeSmith> | annevk: read through my doc. I lack the time to enumerate what may be missing |
| 11:39 | <annevk> | k, because you thought it missed something you'd like to see there... |
| 11:39 | <annevk> | MikeSmith, the chair needs to be involved in LC stuff most definitely btw, e-mailing chairs⊙wo and such |
| 11:40 | <Hixie> | annevk: oh? |
| 11:41 | <annevk> | Hixie, not you, MikeSmith :) |
| 11:41 | <Hixie> | oh |
| 11:41 | <Hixie> | oh, oops, the line got highlighted cos it included the word "html5" and i assumed you were talking to me! |
| 11:41 | <MikeSmith> | annevk: yeah. but ArtB is generally supportive of moving work forward rather than allowing it to be bogged down or tar-babied |
| 11:41 | <annevk> | hah |
| 11:41 | <Hixie> | nm :-) |
| 11:42 | <Dashiva> | MikeSmith: It's a bit worrying how the issues value in the graph doesn't seem to change even though the revision number goes up and up :) |
| 11:42 | <MikeSmith> | and Art is not particularly happy about the fact that the grand WG merger has now taken 7+ months to not happen |
| 11:43 | <MikeSmith> | so in the mean time, we hav things to get done |
| 11:44 | <MikeSmith> | Dashiva: /me points to Hixie about that.. |
| 11:45 | <Hixie> | Dashiva: the XXX count will go up until the e-mails hit near zero |
| 11:45 | <MikeSmith> | annevk: Access Control actually getting implemented in some UAs will be a huge help |
| 11:45 | <Hixie> | Dashiva: i'm not spending any time tracking down XXXs |
| 11:45 | <Hixie> | Dashiva: i just add them when i see something wrong as i'm editing other things |
| 11:45 | <MikeSmith> | ++ to Webkit and and Sam Weinig and othermaciej for helping get AC actually implemented |
| 11:46 | <MikeSmith> | annevk: maybe you can do something about getting some Opera support behind it |
| 11:46 | <Hixie> | gotta love people who have clearly not yet worked out the scale of the web |
| 11:46 | <Hixie> | "1.7% is nothing!" |
| 11:46 | <MikeSmith> | Hixie: heh |
| 11:46 | <Hixie> | uh no. welcome to the web. |
| 11:47 | <Hixie> | 1.7% is probably more pages than you have ever looked at it in your entire life. |
| 11:47 | <Hixie> | and more people than you've met or dealt with ever. |
| 11:47 | <MikeSmith> | we need better need ways of modding down all the noise |
| 11:47 | <Hixie> | i mean, elements used only on 0.01% of pages can still have giant interop issues |
| 11:48 | <MikeSmith> | Aaron Leventhal has some good ideas about how to deal with representing "truth structures" in public discussions |
| 11:49 | <MikeSmith> | ... hope we can make time to actually do some of that |
| 11:50 | <Dashiva> | Hixie: Yeah, you told me that before. I'm just enjoying the perspective of someone who doesn't know :) |
| 11:50 | <MikeSmith> | e.g., I never want to see any message about something that has proven to be false |
| 11:50 | <MikeSmith> | ... or that at least there is general agreement is false |
| 11:51 | <MikeSmith> | what I want is: bubble up to my attention the arguments that have some degree of agreement as being viable |
| 11:51 | <MikeSmith> | and please god give me a way to drown out the noise |
| 11:51 | <MikeSmith> | e.g., RB |
| 11:51 | <Hixie> | Dashiva: :-P |
| 11:52 | <Hixie> | Dashiva: actually i do occasionally deal with XXXs. Mostly when my mail server goes down. :-) |
| 11:52 | <roc> | a friend of mine is working on a social Web site for representing truth structures |
| 11:52 | <roc> | it's doomed |
| 11:52 | <Dashiva> | Hixie: Do you still do that, or did it stop when you got GTA4? |
| 11:52 | <roc> | BTW doesn't Jonas deserve a ++ for getting AC implemented too? |
| 11:53 | <Hixie> | Dashiva: you can see where i got GTA4 on that chart |
| 11:53 | <Hixie> | Dashiva: it's just before the last hump before the last big fall |
| 11:53 | <Hixie> | Dashiva: (april 29th) |
| 11:53 | <annevk> | roc, yeah |
| 11:53 | <annevk> | sicking++ |
| 11:53 | <Hixie> | Dashiva: took me out for a week :-) |
| 11:53 | <Dashiva> | Hehe, yeah |
| 11:54 | <annevk> | roc, been quite useful in discussions so far when people complain it's too hard |
| 11:54 | <MikeSmith> | roc: yeah, Jonas deserves ++ for many many things |
| 11:54 | <annevk> | MikeSmith, e-mailed Art + you |
| 11:54 | <MikeSmith> | annevk: thanks |
| 11:55 | <MikeSmith> | general comment: the way things are now, we have N different ones of us taking time to set up our own private filtering |
| 11:55 | <MikeSmith> | Hixie gots his AAA-important folder and filter |
| 11:56 | <MikeSmith> | I got my my less-tactfully-worded bozo filter |
| 11:56 | <MikeSmith> | we could really benefit from some way to share our bozo rules |
| 11:56 | <Hixie> | it's called "aaa-productivity" |
| 11:56 | <Hixie> | for extra irony |
| 11:57 | MikeSmith | is taking a moment for a belly laugh |
| 11:57 | <MikeSmith> | anyway, a pipe dream |
| 11:57 | <Dashiva> | If you can get away with it without being accused of censoring and favoritism, excellent |
| 11:58 | <MikeSmith> | we will continue to be plagued by the likes of RB I guess |
| 11:58 | <MikeSmith> | Dashiva: not censoring, not favortism |
| 11:58 | <MikeSmith> | not at all |
| 11:58 | <MikeSmith> | that't the pont |
| 11:58 | <MikeSmith> | point |
| 11:59 | <MikeSmith> | the whole "wisdom of the crowds" ideal |
| 11:59 | <MikeSmith> | modding it collectively |
| 11:59 | <Dashiva> | MikeSmith: I know, but do you think the same people who create all this noise would stand silent, regardless of rational arguments? |
| 11:59 | <MikeSmith> | which may or may not actually work in practice |
| 12:00 | <MikeSmith> | Dashiva: we don't need a way for them to stand silent |
| 12:00 | <MikeSmith> | we just need a way to mod them down |
| 12:00 | <Hixie> | nn |
| 12:00 | <MikeSmith> | Hixie: otsukare-sama and o-yasumi |
| 12:00 | <Dashiva> | Does Hixie dream of electric spec writers? |
| 12:02 | <Dashiva> | MikeSmith: I guess I'm a bit pessimistic with regard to factionism |
| 12:02 | <MikeSmith> | not factionism |
| 12:12 | <MikeSmith> | annevk: see my reply |
| 12:17 | MikeSmith | turns to reading http://lists.w3.org/Archives/Public/public-html/2008May/0598.html |
| 12:18 | <MikeSmith> | hsivonen: I can't say I'm particularly happy about the completely arbitrary requirement that charset encoding markup declaration must occur within first 512 bytes |
| 12:19 | <MikeSmith> | real-worl documents may have comments much longer than that |
| 12:20 | <annevk> | what's wrong with <!DOCTYPE html><meta charset=x><!-- really really long comment --> ? |
| 12:20 | <MikeSmith> | Hixie reply about "multimegabyte comment" seems to me like a red herring |
| 12:20 | <MikeSmith> | or whatever other rhetorical term is appropriate here |
| 12:21 | <MikeSmith> | there are in practice no "multimegabyte comments" |
| 12:21 | <MikeSmith> | so not even sure sure wtf he is talking about there |
| 12:21 | <annevk> | and even if they are, there's been no reason given they can't occur after a <meta> |
| 12:21 | <takkaria> | gsnedders: re: http5, for some sites (like royalmail.com) require you to send an Accept: header or they don't serve up any content |
| 12:22 | <annevk> | <meta> can have a reparsing effect, longs comments won't |
| 12:22 | <MikeSmith> | but 512 seems a totally arbitrary and effectively counter-productive criterion to me |
| 12:22 | <annevk> | well, if browsers do a prescan, it's likely over the first packet or something, which is 512 bytes |
| 12:23 | <gsnedders> | MikeSmith: HTML 5 has some really large comments in it :P |
| 12:24 | <MikeSmith> | and then there's the criterion that meta@charset must be the 1st child element of head.. I really don't care why that was added, but it it is an absolutely broken requirement |
| 12:24 | <gsnedders> | takkaria: FWIW, I'm currently only working on defining parsing. Things like that are currently out of scope — HTTP5 is a purely theoretical spec that could be based upon the parsing spec. |
| 12:24 | <gsnedders> | takkaria: But feel free to send me an email about it |
| 12:24 | <gsnedders> | takkaria: That's the sort of thing I have informatively mentioned currently |
| 12:24 | <takkaria> | gsnedders: k, will do |
| 12:26 | <annevk> | MikeSmith, that's to prevent re execution of scripts and such |
| 12:27 | <MikeSmith> | annevk: I see. but I guess you know that getting authors to comply with it in practice is, well... optimistic |
| 12:29 | <MikeSmith> | ... along with a good number of instances of autoring-conformance criteria in the current draft that basically are completely arbitrary from the POV of authors |
| 12:30 | <annevk> | I thought we tried to reduce those |
| 12:30 | <MikeSmith> | annevk: trying is not doing |
| 12:31 | <annevk> | Sorry, did reduce those... |
| 12:31 | <annevk> | Anyway, maybe you should comment on the draft? |
| 12:31 | <gsnedders> | takkaria: Sorry, I'm a bit busy rediscovering the awesomeness of NFS:HP2 |
| 12:34 | <billyjack> | annevk: speaking of arbitrariness |
| 12:34 | <billyjack> | I'm told that the guy who maintains http://logopoeia.com/wisdom/ |
| 12:35 | <billyjack> | .. that there are some bits of wisdom there (easter eggs) |
| 12:35 | <billyjack> | ... that if you care to take the time |
| 12:35 | <annevk> | isn't "NORMALISATION" with a "Z" in the US? |
| 12:35 | <billyjack> | ... if you reload that a few tims |
| 12:35 | <billyjack> | times |
| 12:36 | <billyjack> | e.g. 10 times or so |
| 12:36 | <billyjack> | you might discover some real bits of wisdom |
| 12:36 | <annevk> | heh |
| 12:36 | <billyjack> | wisdom wisdom which you would really not want to paste into this channel |
| 12:36 | <billyjack> | or some other channels |
| 12:37 | <billyjack> | but wisdom for your own personal enlightenment |
| 12:37 | <billyjack> | like Zen |
| 12:37 | <billyjack> | annevk: yeah, Z |
| 12:37 | <annevk> | "I live in Tokyo and work as a software non-executive." |
| 12:38 | <billyjack> | Hixie has this weird thing for wanting to spell everyhing with S's |
| 12:38 | <gsnedders> | annevk: Technically, it should be z in en-gb too. It is in en-gb-oed. |
| 12:38 | <billyjack> | it's nouts |
| 12:38 | <billyjack> | nuts |
| 12:38 | <annevk> | I think he fixed some of that recently |
| 12:38 | <billyjack> | annevk: yeah, he did in many places |
| 12:39 | <billyjack> | though I think "Tokenisation" remains |
| 12:39 | <billyjack> | in truth, I couldn't care less myself about the UK vs. US spelling |
| 12:39 | <gsnedders> | en-gb-x-hixie is defined to use -ize and not -ise, FWIW |
| 12:40 | <billyjack> | I think he changed it based on feedback from others |
| 12:40 | <gsnedders> | (as -ise is a made up thing from the 20th cent. — -ize is technically correct) |
| 12:41 | <billyjack> | maybe in regard to search engines that are too dumb to know that Tokenisation and Tokenization mean exactly and precisely the same thing.. i dunno |
| 12:41 | <billyjack> | this is one of the beautiful things about Japanese |
| 12:41 | <billyjack> | (the need to not deal with Tokenisation vs Tokenization |
| 12:42 | <gsnedders> | billyjack: Google does some normalization |
| 12:42 | <billyjack> | I know G does |
| 12:42 | <billyjack> | lesser engines seem to have more trouble.. |
| 12:48 | <jgraham> | gsnedders: You're missing the whole language evolution thing |
| 12:48 | <MikeSmith> | gsnedders: btw, so have seen mention here about your university plans and such |
| 12:48 | <MikeSmith> | ... curious as to what your actual plans are |
| 12:48 | <gsnedders> | MikeSmith: Congrats on being able to read! |
| 12:48 | <gsnedders> | :P |
| 12:48 | <gsnedders> | jgraham: Hey! I want to read anglo-saxon (sp?)! |
| 12:48 | <jgraham> | (like pluralising with s is a made-up since roman times thing iirc |
| 12:48 | <jgraham> | ) |
| 12:48 | <MikeSmith> | I hate reading |
| 12:49 | <MikeSmith> | plurals are a mostly unnecessary language construct |
| 12:50 | <MikeSmith> | you can almost always get away without plurals and still avoid ambiguity |
| 12:50 | <MikeSmith> | (works fine in Japanese at least, for the most part) |
| 12:51 | <jgraham> | MikeSmith: It's just style. Japanese has counters or something instead, right? |
| 12:51 | <jgraham> | s/style/language style/ |
| 12:51 | <MikeSmith> | counters yeah |
| 12:52 | jgraham | can't actually remember how those work |
| 12:52 | <MikeSmith> | every language has counters |
| 12:52 | <MikeSmith> | e.g, "four *sheets* of paper" |
| 12:52 | <MikeSmith> | "four *pieces* of paper" |
| 12:53 | <MikeSmith> | nothing unique about those in Japanese, really |
| 12:53 | <MikeSmith> | four glasses of bear |
| 12:53 | <MikeSmith> | beer |
| 12:53 | <jgraham> | MikeSmith: Right. I seem to recall that they are a bigger deal in Japanese than in English |
| 12:53 | <MikeSmith> | yeah |
| 12:53 | <gsnedders> | MikeSmith: In short, as I'm running off, Cambridge's Comp.Sci. BA, Edinburgh's Comp. Phys. MPhys, or York's Theoretical Phys. MPhys |
| 12:53 | <MikeSmith> | they are a bit more |
| 12:53 | <jgraham> | In the same way that morphology is a bigger deal in Romance languages |
| 12:53 | <MikeSmith> | yup |
| 12:54 | <MikeSmith> | but in Japanese the general rule is that you can get by without much of that |
| 12:54 | <MikeSmith> | and still be "correct"] |
| 12:55 | <MikeSmith> | e.g., a grammatically correct sentence requires only a verb |
| 12:55 | <MikeSmith> | and no explii |
| 12:55 | <MikeSmith> | explicitly-stated subject |
| 12:56 | <jgraham> | I don't know any Japanese but it seems like it would be a pretty nice language except for a) Kanji and b) the social structures baked into the language |
| 12:56 | <MikeSmith> | kanji is a major major PITA, that's for sure |
| 12:56 | <MikeSmith> | but it is not absolutely intellectually difficult |
| 12:57 | <MikeSmith> | learning kanji is just brute-force memorization |
| 12:57 | <MikeSmith> | basically |
| 12:57 | <jgraham> | Right, but learning 10,000 hard-to-draw symbols need not be intellectually difficult to be a major problem :) |
| 12:57 | <MikeSmith> | social constructs are a totally different matter |
| 12:58 | <MikeSmith> | but for the most part non-native speakers get a "by" on much of those |
| 12:58 | jgraham | has difficulty drawing 26 alphabetic characters with sufficient uniqueness |
| 12:59 | <jgraham> | I guess it ends up being just another way that your foreignness is made obvious though |
| 12:59 | <MikeSmith> | jgraham: the amount of information that are represented by the complete set of kanji in normal use pales in comparison to the set of information you have in your mind about many other subject areas |
| 13:00 | <MikeSmith> | I promise you you that |
| 13:00 | <MikeSmith> | and for the most part you don't need to write them |
| 13:00 | <MikeSmith> | don't need to hand-write them |
| 13:00 | <Lachy> | memorising 10,000 different symbols seems like it would be impossible |
| 13:01 | <MikeSmith> | you just need to be able to real them |
| 13:01 | <MikeSmith> | Lachy: not impossible |
| 13:01 | <MikeSmith> | not by a long shot |
| 13:01 | <Lachy> | on average, how many of them do most japenese people know? |
| 13:01 | <MikeSmith> | you have in your mind now many sets of info that are orders of magnitude larger than the set of kanji |
| 13:02 | <MikeSmith> | Lachy: minimal adult literacy is on the order of 2000 kanji |
| 13:02 | <jgraham> | MikeSmith: I think for one brief part of my life I could recognise the Kanji for Railway Station |
| 13:02 | <MikeSmith> | 駅 |
| 13:02 | <MikeSmith> | eki |
| 13:03 | <MikeSmith> | part of that character means "horse" |
| 13:03 | <Lachy> | how do you type those characters without having 10,000 keys on the keyboard? |
| 13:03 | <jgraham> | Yeah, I remember the word. Getting on buses and going "Eki?" was a frequent part of my trip to Japan |
| 13:03 | <Lachy> | or even just a few thousand, given modifier keys? |
| 13:03 | <MikeSmith> | that's the other thing with many kanji -- they can be decomposed into familiar parts |
| 13:04 | <jgraham> | Lachy: There's a system where you type ascii and combinations are turned into kanji |
| 13:04 | <jgraham> | I think |
| 13:04 | <MikeSmith> | Lachy: invest some time in learning about CJK input modes |
| 13:04 | <MikeSmith> | :) |
| 13:04 | <MikeSmith> | talk to Wilhelm |
| 13:04 | <Lachy> | so not only do japenese people need to learn their own symbols, but they need to learn to type them phonetically in ASCII? |
| 13:04 | <MikeSmith> | Lachy: yep |
| 13:05 | <MikeSmith> | my daughter who is just 10 years old is now learning for the first time how to represent Japanese in the "roman" alphabet |
| 13:06 | <Lachy> | wouldn't it be easier if they just learned a language that used the roman alphabet, like Engrish? |
| 13:06 | <MikeSmith> | and the majority of people in japan type Japanese in roman chars on their computer keyboards |
| 13:07 | <jgraham> | Lachy: The effort of learning a whole new language seems huge compared to learning a different representation of a familiar grammar |
| 13:07 | <MikeSmith> | so if I want to type 駅, I type "e"+"ki" to get えき, then use the IME to change that to 駅 |
| 13:08 | <MikeSmith> | Lachy: roman alphabet sucks in many ways |
| 13:08 | <Lachy> | in what way? |
| 13:08 | <MikeSmith> | alphabets in general are deficient |
| 13:08 | <MikeSmith> | Lachy: for one thing - |
| 13:09 | <MikeSmith> | take the the symbol, 人 |
| 13:09 | <MikeSmith> | it means the same thing in Chinese and Japenese |
| 13:09 | <MikeSmith> | Japanese |
| 13:10 | <MikeSmith> | though its sound/reading is different |
| 13:10 | <Lachy> | I asked about the roman alphabet though. |
| 13:10 | <MikeSmith> | but you don't need to know the the sound/reading/pronunciation |
| 13:10 | <jgraham> | MikeSmith: Are you claiming that is a feature? |
| 13:10 | <MikeSmith> | jgraham: yes, absolutely |
| 13:11 | <MikeSmith> | it is a superior way to convery information |
| 13:11 | <Lachy> | MikeSmith, why is it a feature for the same symbol to be pronounced differently? I call that a bug |
| 13:11 | <MikeSmith> | a pure symbol |
| 13:11 | <jgraham> | (lots of words in english and, say German have similar alphabetic representations but different pronunciations) |
| 13:11 | <MikeSmith> | Lachy: same meaning |
| 13:11 | <jgraham> | MikeSmith: I think I tend to disagree |
| 13:11 | <MikeSmith> | that's why |
| 13:11 | <Lachy> | same meaning should have same sound |
| 13:11 | <MikeSmith> | Lachy: no |
| 13:12 | annevk | disagrees too, though not on the sound/reading basis but because 9000 symbols is way harder to grasp than 26 letters |
| 13:12 | <Lachy> | Why? It makes communicating aurally harder |
| 13:12 | <MikeSmith> | jgraham: you would disagree less if you were a Japanese speaker not fluent in Chinese and found yourself trying to read signs and menus and such in China |
| 13:12 | <jgraham> | The problem with using one symbol per concept is that the number of concepts is much larger than the reasonable number of symbols. Therefore you have to start combining symbols and soon enough you have an alphabet |
| 13:13 | <MikeSmith> | jgraham: no, not really |
| 13:13 | <MikeSmith> | the majority of written communication in Japanese is done in pure symbols |
| 13:13 | <Lachy> | MikeSmith, I agree it's good to use the same symbol for the same meaning. I just think it's a bug that it's pronounced differently |
| 13:14 | <Lachy> | isn't all *written* communication done with symbols? |
| 13:14 | <MikeSmith> | Lachy: I encourage you to get Chinese people and Japanese to pronounce those symbols in the same way |
| 13:15 | <jgraham> | In what way? I would have thought that you could either have a simple alphabet which makes learning easy but allows more combinations of symbol or a more complex alphabet which makes learning hard but leads to fewer different combinations to represent the same concept |
| 13:15 | <MikeSmith> | Lachy: yeaha, all written communication is done in symbols, but some symbols are better than others |
| 13:15 | <Lachy> | I would. But I'm having enough trouble trying to get Norwegian's to stop mispronouncing their own letters. :-) |
| 13:15 | <jgraham> | I don't think optimising for Japanese speakers reading Chinese menus is the right optimisation |
| 13:15 | <MikeSmith> | as far as alphabets, Korean Hangul is absolutely a better alphabet than anything else |
| 13:16 | annevk | finds http://www.logoi.com/notes/pictograph.html |
| 13:16 | <MikeSmith> | and roman alphabet is relatively very ppor |
| 13:16 | myakura | wonders he can read and understand chinese menus if it's simplified |
| 13:16 | <Lachy> | I find it annoying the, for example, the letter 'a' is prounced differently in norwegian |
| 13:17 | <annevk> | the letter 'a' is pronounced differently all over the place |
| 13:17 | <Lachy> | I know. It's annoying |
| 13:17 | <jgraham> | MikeSmith: I certainly don't claim that the Roman alphabet is perfect (I was also using the word alphabet to include things like Kanji that perhaps are strictly not alphabets) |
| 13:17 | <annevk> | Lachy, it's annoying not everyone speaks the same language, yes... |
| 13:17 | <MikeSmith> | Lachy: Norwegian pronunciation is a mess.. you have some people who are fairly understandable, then you have those like Arve who do that sing-song thing |
| 13:17 | <Lachy> | but 'i' and 'y' are even more annoying here. They're pronounced exactly the same in norwegian, even though the norewegians try to claim otherwise |
| 13:18 | <jgraham> | Lachy: In what context? |
| 13:18 | <MikeSmith> | btw, myakura knows infinitely more about this than me |
| 13:18 | <jgraham> | I mean English isn't phonetic so it's not like we have a way of pronouncing "a" |
| 13:18 | <jgraham> | I think that phonetic writing systems make more sense than pictoral ones |
| 13:19 | <jgraham> | (although some Euopean languages manage the phonetic thing much better) |
| 13:20 | <MikeSmith> | jgraham: my 10 year old daughter seems to have no trouble learning two different Japanese phonetic alphabets while at the same time having learned to write and read hundreds of kanji and now learning the roman alphabet |
| 13:21 | <Lachy> | MikeSmith, that's because children still have the ability to learn new languages, which adults generally lose |
| 13:21 | <MikeSmith> | ture |
| 13:22 | <MikeSmith> | true |
| 13:22 | <myakura> | yeah |
| 13:22 | <jgraham> | MikeSmith: Picking a 10 year old isn't really fair because below the age of ~10 humans are built for learning language. Above that age we seem to be much worse at it (compare foreign language ability in the UK where people start learning at 11 to countries where they start younger) |
| 13:22 | <jgraham> | Hmm Lachy said that but shorter and faster |
| 13:23 | <Lachy> | plus, children's hearing is generally better so they can pick up the slight nuances in the sounds easier than adults |
| 13:24 | <MikeSmith> | jgraham: yeah, I will concede that. problem is that, as you point out, educational system don't start to teach languages to children until they are past the point of optimal language learning |
| 13:24 | <Lachy> | so while a kid may not have difficulty differentiating between 'i' and 'y' in norweign, I certainly can't. |
| 13:25 | <MikeSmith> | Lachy: in Japanese, the troublesome sounds are "r" and "l" |
| 13:25 | <myakura> | and 'b' and 'v' |
| 13:25 | <MikeSmith> | ah yeah |
| 13:25 | <MikeSmith> | video vs. bideo |
| 13:26 | <myakura> | i don't think i can easily differentiate them |
| 13:26 | <MikeSmith> | myakura: I think you can :) |
| 13:28 | <MikeSmith> | myakura: in terms of pronunciation and other things, it's far easier for me to communicate with you face to face than is it for me to communicate with many "native" English speakers |
| 13:28 | <MikeSmith> | e.g., I was in Dublin a couple weeks ago... |
| 13:28 | <Lachy> | MikeSmith, what's your native language? |
| 13:28 | <myakura> | ah |
| 13:29 | <MikeSmith> | And in several instances in Dublin, I found myself, in the "nod my head and pretend I understand" situation |
| 13:30 | <MikeSmith> | Lachy: I was born in Oklahoma and lived for the first 3 years of my life in Japan |
| 13:31 | <takkaria> | hehe, Dublin |
| 13:31 | <MikeSmith> | ... but my native language is essentially English, I guess |
| 13:32 | <MikeSmith> | with large amounts of mexican/central-american spanish thrown in |
| 13:32 | <myakura> | MikeSmith: i don't think i can easily communicate with people with brogue either |
| 13:32 | <MikeSmith> | and french-Canadian |
| 13:32 | <MikeSmith> | myakura: you and me both :) |
| 13:33 | <MikeSmith> | e.g., a guy in Dublin was talking about "cooking a stick" |
| 13:33 | <MikeSmith> | ... and I nod my head |
| 13:33 | <MikeSmith> | ... yeah, cooking a stick (what the hell is this guy talking about) |
| 13:34 | <MikeSmith> | ... but apparently the word "steak" is pronounced much closer to "stick" in Dublin |
| 13:35 | <Lachy> | MikeSmith, did the guy have an irish accent? |
| 13:35 | <MikeSmith> | Lachy: heh, yeah |
| 13:35 | <MikeSmith> | ... that would be an understatement |
| 13:35 | <MikeSmith> | He was Irish |
| 13:35 | <Lachy> | where I come from, steak is pronounced the same as stake |
| 13:36 | <MikeSmith> | I personally reckon the Irish to be among the great peoples of the earth |
| 13:36 | <Philip`> | Lachy: That causes terrible problems during vampire attacks |
| 13:37 | <MikeSmith> | ... in terms of their collective sense of humor and love for life and poetry and music |
| 13:37 | <MikeSmith> | ... and just about anything else you could imagine |
| 13:37 | <takkaria> | and catholicism |
| 13:37 | <MikeSmith> | takkaria: yeah |
| 13:37 | <MikeSmith> | ... and except speaking English |
| 13:38 | <MikeSmith> | ... which is is no way their natural native language |
| 13:39 | <MikeSmith> | Lachy: steak is not stake in your language |
| 13:39 | <MikeSmith> | ... you guys love to stretch out vowels to ridiculous extremes |
| 13:40 | <Lachy> | MikeSmith, what? |
| 13:40 | <MikeSmith> | more like "stai-aike" |
| 13:40 | <Lachy> | MikeSmith, no. Aussies don't do that |
| 13:41 | <MikeSmith> | oK |
| 13:41 | <Lachy> | maybe some Americans who have bastardised our language do that |
| 13:42 | <MikeSmith> | well, there are a treasure trove of other examples of OZ embrace-and-extend of pronunciation that I can't think of right now |
| 13:42 | <MikeSmith> | ... not that I'm saying it's bad |
| 13:42 | <MikeSmith> | I like the sound of it, |
| 13:42 | <MikeSmith> | the cadence |
| 13:43 | <MikeSmith> | e.g. the way that chaals talks |
| 13:43 | <Lachy> | MikeSmith, we're proud of our Aussie slang! |
| 13:44 | <MikeSmith> | the way that Arve talks both in English and Norski, I like the sound |
| 13:44 | <Lachy> | chaals doesn't speak like a typcial aussie |
| 13:44 | <MikeSmith> | Lachy: oK, well, here's another example: |
| 13:44 | <Lachy> | he speaks like a victorian |
| 13:44 | <MikeSmith> | David Storey |
| 13:45 | <Lachy> | (Victoria is a state in Aus, where I think Chaals is from) |
| 13:45 | <MikeSmith> | I challenge you to translate wtf David is talking about when he slips into his Geordi-speak |
| 13:46 | <Lachy> | what is Geordi-speak? |
| 13:46 | <MikeSmith> | (which he does pretty often) |
| 13:46 | <MikeSmith> | Lachy: Newcastle, Northumberland |
| 13:46 | <MikeSmith> | the north-north |
| 13:47 | <MikeSmith> | of that tiny little counry |
| 13:47 | <MikeSmith> | country |
| 13:47 | <MikeSmith> | which is a very different world than the south |
| 13:48 | <takkaria> | my sister lived by newcastle for a few years, it's not too bad |
| 13:48 | <MikeSmith> | people in the North in england are quite a bit different than those you find in London and environs |
| 13:49 | <MikeSmith> | I like that Newcastle accent... just that I can't figure out what the hell they are saying most of the time |
| 13:49 | <MikeSmith> | ... seems to involve dropping most vowels |
| 13:49 | <MikeSmith> | and saying things under your breath |
| 13:50 | <jgraham> | glottal stops! |
| 13:50 | jgraham | goes to eat lunch |
| 13:50 | <MikeSmith> | jgraham: yep |
| 13:50 | <takkaria> | in the north we like our glottal stops. :) |
| 13:50 | takkaria | is from manchester |
| 13:50 | <MikeSmith> | takkaria: aha |
| 13:51 | MikeSmith | estimation of takkaria goes up a few notches |
| 13:51 | <takkaria> | heh |
| 13:52 | <takkaria> | though I'm from south manchester, not north, so my accent is reasonably understandable |
| 13:52 | <MikeSmith> | manchester, for one thing, has produced many innovative/exceptional musicians/songwriters/drunkards over the last 15 years or more |
| 13:55 | <MikeSmith> | and as far as accents, "the sound of Ian Brown", for good or bad |
| 13:55 | <MikeSmith> | and many others |
| 13:56 | <takkaria> | manchester has a lot to recommend it |
| 13:56 | <takkaria> | though also far too many unsigned musicians |
| 13:57 | <MikeSmith> | takkaria: you could say the same about many places... Berlin, for one |
| 13:57 | <MikeSmith> | ... or Birmingham even |
| 13:57 | <MikeSmith> | (me is a huge fan of Broadcast, from Birmingham) |
| 13:58 | <takkaria> | I say it about manchester because I'm one of them and it makes it very hard to get noticed by anyone at all. :) |
| 13:58 | <MikeSmith> | takkaria: you got some high expectations to compete with there .. |
| 14:00 | <takkaria> | it's true |
| 14:00 | <MikeSmith> | e.g., Johnny Marr.. what can be said there/ |
| 14:00 | <MikeSmith> | ? |
| 14:01 | <MikeSmith> | even Morrissey |
| 14:01 | <MikeSmith> | nutjob |
| 14:01 | <MikeSmith> | but brilliant |
| 14:01 | <MikeSmith> | and never quits |
| 14:02 | <takkaria> | on a related note, Andy Rourke recommended us to a promoter once and we got a good gig as a result |
| 14:02 | <MikeSmith> | I'm intrigued |
| 14:02 | <MikeSmith> | who is "us" ? |
| 14:03 | <takkaria> | my band, System Fault |
| 14:03 | MikeSmith | notes that he worked in record stores during most of the time he was in university |
| 14:03 | <MikeSmith> | takkaria: first impression: extremely dumb name |
| 14:04 | <MikeSmith> | change that |
| 14:04 | <takkaria> | yeah, it's bad, isn't it? |
| 14:04 | <MikeSmith> | yeah, really |
| 14:04 | <MikeSmith> | unequivocally |
| 14:04 | <MikeSmith> | hmm, |
| 14:05 | <takkaria> | it was chosen in, er, earlier days |
| 14:05 | <MikeSmith> | so which of thsese four ruffians is you? |
| 14:05 | <takkaria> | I've been lobbying for a name change for quite a while but other members aren't having it |
| 14:05 | <MikeSmith> | aha |
| 14:05 | <MikeSmith> | #4 |
| 14:06 | <MikeSmith> | Guitar, Synth & Backing Vox |
| 14:06 | <takkaria> | yeah, that's the one |
| 14:06 | <MikeSmith> | takkaria: do more than lobby |
| 14:06 | <MikeSmith> | that name is millstone |
| 14:08 | <MikeSmith> | and "synth"... well, |
| 14:08 | <gsnedders> | MikeSmith: Want any more specific answers about uni now I return? |
| 14:09 | <MikeSmith> | takkaria: "keyboard wizardy" .. "fingers" |
| 14:09 | <MikeSmith> | please anything but "synth" |
| 14:10 | <takkaria> | MikeSmith: synth is where it's at here in the UK |
| 14:11 | <MikeSmith> | gsnedders: yeah, if you got the time,. the whole point of this is to try to document for "casual readers" what changes have been made, at a high lelve |
| 14:11 | <MikeSmith> | level |
| 14:11 | <MikeSmith> | takkaria: that will pass |
| 14:12 | <MikeSmith> | try breaking some new ground in terms of the way you describe your skills |
| 14:12 | <MikeSmith> | ... that almost always pays off |
| 14:12 | <takkaria> | really? |
| 14:13 | <MikeSmith> | takkaria: this seems obvious to me.. but I guess I may come from a different plane |
| 14:13 | <MikeSmith> | takkaria: just as a data point |
| 14:13 | <gsnedders> | MikeSmith: Huh? |
| 14:13 | <takkaria> | MikeSmith: well, your advice is valued, as someone who really isn't UK-centric at all |
| 14:13 | <MikeSmith> | ... do you now what the "Amen break" means? |
| 14:14 | <takkaria> | of course |
| 14:14 | <MikeSmith> | takkaria: OK, good |
| 14:14 | <MikeSmith> | gsnedders: sorry, misunderstood |
| 14:15 | <MikeSmith> | gsnedders: I think you should weigh Hixie advice against other considerations |
| 14:16 | <MikeSmith> | e.g., focused coursework in CS |
| 14:16 | <MikeSmith> | as opposed to physics |
| 14:16 | <MikeSmith> | (I say this as someone who minored in physics myself at university) |
| 14:17 | <MikeSmith> | ... and physical sciences in genera; |
| 14:17 | <MikeSmith> | general |
| 14:18 | <MikeSmith> | in particular, there is a good argument for spending a lot of time at university learning the math fundamentals underlying CS |
| 14:18 | jgraham | doesn't understand what MikeSmith means about focussed coursework |
| 14:18 | <gsnedders> | MikeSmith: Most of the advice I've got agrees with Hixie though :P |
| 14:19 | <MikeSmith> | learning while you have the luxury of time |
| 14:19 | <Lachy> | gsnedders, advice about what? |
| 14:19 | <gsnedders> | MikeSmith: In the case of the comp.phys. at Edi, in first year you do the two introductory CS courses |
| 14:19 | <gsnedders> | Lachy: uni |
| 14:21 | <Lachy> | gsnedders, so how did you go in your physics exam? Was that the one you had yesterday? |
| 14:21 | <MikeSmith> | jgraham: algorithms |
| 14:21 | <MikeSmith> | (in so many words) |
| 14:21 | <gsnedders> | Lachy: Yeah. I thought the paper was hard, but so did everyone. I think I did quite well in it though. |
| 14:21 | <MikeSmith> | basics |
| 14:21 | <MikeSmith> | structures |
| 14:21 | <gsnedders> | (There again, "quite well" in my world means a low A) |
| 14:21 | <MikeSmith> | math |
| 14:22 | <Lachy> | physics is easy. It's just remembering a bunch of formulas and knowing a few constants |
| 14:22 | <jgraham> | I understand that learning CS basics is more likely to happen on a CS course but I'm still not clear on what "focused coursework" means |
| 14:22 | <Philip`> | gsnedders: Do you know how the maths content varies between all the courses? |
| 14:22 | <Lachy> | at least, it was for me in high school |
| 14:22 | <jgraham> | Lachy: I like to think that it's actually harder than that :) |
| 14:23 | <MikeSmith> | Lachy: physics is not easy |
| 14:23 | <MikeSmith> | at a certain level it is |
| 14:23 | <MikeSmith> | like many things |
| 14:23 | <Lachy> | MikeSmith, I know. |
| 14:23 | <gsnedders> | Philip`: In the case of Edi. and pure maths, the maths in comp.phys. is identical to that of phys. (but I don't know what that is) |
| 14:24 | <MikeSmith> | beyond basics, physics and math grow into the realm of the non-intuitive |
| 14:24 | <MikeSmith> | that's the problem for most people |
| 14:25 | <jgraham> | gsnedders: In general physics courses will have you doing "continuous" maths (e.g. vector calculus, differential equations) |
| 14:25 | <jgraham> | whereas CS courses have you doing discrete maths |
| 14:25 | <jgraham> | which I am less sure about |
| 14:25 | <jgraham> | so I'll let someone else describe it in case I am wrong |
| 14:26 | <MikeSmith> | jgraham: I guess I mean investing time over 3-4 years in learning the math and fundamentals behind modern CS while at university |
| 14:26 | <MikeSmith> | ... instead of having to take/make time to learn it later |
| 14:26 | <jgraham> | MikeSmith: I guess that assumes that your end goal is to know CS |
| 14:26 | <Philip`> | There's quite a bit of partly-continuous maths like probability and Fourier transform stuff in CS here |
| 14:27 | <jgraham> | In which case taking CS is a good idea :) |
| 14:27 | <MikeSmith> | jgraham: that is a worthy end goal, as far as I'm concerned |
| 14:27 | <jgraham> | MikeSmith: As someone who knows less CS than he would like, I agree, but I would also say the same about Physics |
| 14:27 | <MikeSmith> | this gets back in part to the idea of having a common framework for discussion |
| 14:28 | <jgraham> | Or indeed Mathematics itself |
| 14:28 | <Philip`> | (AI in particular had lots of integral equations and stuff) |
| 14:28 | <MikeSmith> | jgraham: well, we here are not here because of common interest in Physics |
| 14:29 | <jgraham> | MikeSmith: Sure. But gsnedders might be here because he is interested in web stuff but still be more interested in Physics |
| 14:29 | <jgraham> | Therefore CS would be the wrong choice for him |
| 14:29 | gsnedders | would have more enthusiasm to be doing something like comp.phys. than CS |
| 14:30 | jgraham | thinks that the combination of computers and differential equations is the root of all evil |
| 14:30 | <gsnedders> | And from a lot of what I've seen CS is badly taught at a heckuva lot of places, which doesn't add to the enthusiasm |
| 14:30 | Philip` | guesses jgraham wouldn't want to do electronic engineering :-) |
| 14:30 | <Lachy> | what does CS stand for? |
| 14:30 | <gsnedders> | Lachy: Computer Science |
| 14:30 | <takkaria> | depending on your uni, switching courses can be more or less easy, but if you find you're on the wrong one, then changing shouldn't be too bad |
| 14:31 | <Lachy> | ok. We call that IT where I come from |
| 14:31 | <jgraham> | FWIW I still maintain that, were he to get in the Cambridge CS course is ideal for gsnedders because he would do both Physics and CS in the first year |
| 14:31 | <MikeSmith> | part of the context for what I say is that here in Japan, I sometimes (often) meet "engineers" who have gone through university course in CS and seem to me to be lacking in understanding of some fundamentals that any graduate with a CS degree from a major university in europe or n. america would be assumed to have |
| 14:31 | <gsnedders> | And I could always change to doing Physics in second year, too |
| 14:31 | <jgraham> | gsnedders: Indeed. Hence ideal |
| 14:32 | gsnedders | has a Cam prospectus next to him |
| 14:32 | <jgraham> | Also, I would worry that a computational physics course wouldn't teach much in the way of actual CS |
| 14:32 | <jgraham> | Though I could be wrong |
| 14:33 | <gsnedders> | It's the only one of the three that has arrived. There again, I bet the other two unis haven't even done their '09 prospectuses yet :P |
| 14:33 | <takkaria> | gsnedders: which unis are you looking into? |
| 14:33 | <gsnedders> | http://www.ed.ac.uk/studying/undergraduate/finder/degree.php?id=0,4,F355 — that has a shortish description of it |
| 14:33 | <gsnedders> | takkaria: Cam, Edi, York |
| 14:33 | <MikeSmith> | jgraham: based just on anecdotal datea, I guess i should also note that some of the cleverest/most-effective developers I personally know are people who did not major in CS at university |
| 14:34 | <Philip`> | Lachy: The impression I have is that IT here is about things like "how do I make an Access database to store my business's client list", whereas CS is about things like "how do I model a database using relational algebra, and how do I implement it efficiently taking account of disk geometry" :-) |
| 14:34 | <takkaria> | gsnedders: I would imagine they all have their 09 prospectuses sorted, I'm pretty sure Manchester has |
| 14:35 | <gsnedders> | takkaria: Then Edi and York are just really slow at sending it :) |
| 14:35 | <Lachy> | Philip`, I covered all of that stuff in my Bachelor of IT course that I did at uni. |
| 14:35 | <gsnedders> | It's hardly as if it has a long way to come from either, especially Edi. |
| 14:36 | <jgraham> | Lachy: Which uni? |
| 14:36 | <gsnedders> | http://www.ph.ed.ac.uk/undergraduate/degrees/structure/ — that has far more information about the course content for the comp.phys. |
| 14:37 | <takkaria> | gsnedders: they'll be receiving hundreds and hundreds of requests for prospectuses though, so some delay is inevitable |
| 14:37 | <gsnedders> | takkaria: And I guess Cam's propaganda department is good :) |
| 14:38 | <Lachy> | jgraham, Charles Sturt University in Bathurst, NSW, Australia |
| 14:38 | <Lachy> | csu.edu.au |
| 14:39 | <Philip`> | takkaria: I don't see why it'd take long to print only hundreds of copies of something, and the demand should be entirely predictable based on previous years |
| 14:41 | <jgraham> | gsnedders: That course seems to be mostly physics with a bit of computing. I would try and find the detailed syllabus to check if the "computing" bits are solving differential equations using fortran 77 |
| 14:42 | <takkaria> | Philip`: I would guess sending out prospectii to colleges and the like would take precedence, and perhaps "hundreds" was a little on the low side |
| 14:42 | <takkaria> | Philip`: but you have a fair point |
| 14:43 | <jgraham> | gsnedders: have you seen http://www.inf.ed.ac.uk/teaching/degrees/cs_ph.html? |
| 14:43 | <gsnedders> | takkaria: I know that "we" (my school) only get the Scottish ones, FWIW |
| 14:43 | <gsnedders> | jgraham: yes |
| 14:44 | <Philip`> | jgraham: It points to http://www.ph.ed.ac.uk/undergraduate/degrees/tasters/compsim2.html which menions Java which is probably a bit more sensible than Fortran :-) |
| 14:44 | <Philip`> | s//t/ |
| 14:44 | <gsnedders> | jgraham: Yeah, it's certainly Java that's used |
| 14:45 | <Philip`> | Clearly it would be better to use ML |
| 14:47 | <gsnedders> | Philip`: Haskell _may_ come in to it, I can't remember which of the CS courses you do |
| 14:48 | <takkaria> | common lisp > ML |
| 14:48 | <gsnedders> | I'm sure I found something better about the course before |
| 14:49 | <gsnedders> | Yeah, Haskell comes into it |
| 14:49 | <gsnedders> | <http://www.inf.ed.ac.uk/teaching/courses/inf1/1Aguide.html>; |
| 14:51 | <gsnedders> | <http://www.drps.ed.ac.uk/06-07/dpts/SCE_FINAL/147.html>; and <http://www.ph.ed.ac.uk/teaching/progspecs/ProgSpecCP.pdf>; |
| 14:52 | <Philip`> | "Applicable Mathematics" - is that implying the rest of mathematics is inapplicable? |
| 14:52 | <gsnedders> | Philip`: My thoughts exactly :) |
| 14:53 | <gsnedders> | Oh, Fortran does come into it |
| 14:53 | <Philip`> | "Parallel Fortran in Physics" |
| 14:53 | <gsnedders> | But that was inevitable. |
| 14:54 | <gsnedders> | MikeSmith: How much of CS is just maths without doing much application of it? Or does that vary massively? |
| 14:57 | <gsnedders> | Philip`: ^^ |
| 14:59 | <gsnedders> | http://www.drps.ed.ac.uk/08-09/dpts/SCE_FINAL/147.html — that's '08–'09, which is more up to date, and is the latest available |
| 15:02 | <gsnedders> | It seems for the comp.phys. you can do the norm CS courses too as your outside courses |
| 15:10 | <Philip`> | gsnedders: I can't think of much maths I did that wasn't leading to some kind of application - I suppose probability was kind of left on its own, but other maths lead to databases and cryptography and digital signal processing and automated theorem proving and proving program correctness and whatever, and so pretty much all the maths was done with that kind of final goal in mind and was rarely taught as a pure maths course |
| 15:12 | <Philip`> | though I could be misremembering, particularly since I gave up on some of the maths ones since they were too hard and I was too lazy :-) |
| 15:15 | <Philip`> | The Denotation Semantics course taught me that "⊤" was called "top" and "⊥" was "bottom" but I'm not sure I managed to follow much else of it :-/ |
| 15:15 | <Philip`> | *Denotational |
| 15:17 | <takkaria> | grr, I want my IRC client to do UTF-8 |
| 15:17 | <Lachy> | takkaria, which IRC client are you using? |
| 15:18 | <takkaria> | irssi via SSH |
| 15:18 | <Philip`> | irssi via (screen via) SSH does UTF-8 fine for me, though its font fails to draw ⊥/⊤ except as boxes :-( |
| 15:19 | <takkaria> | hmm, I'm just getting ? instead of boxes |
| 15:21 | <Philip`> | Maybe your font doesn't have the box glyph either |
| 15:22 | <takkaria> | I seem to be using a unicode font OK |
| 15:22 | <takkaria> | oh, I'm not |
| 15:22 | <takkaria> | that explains a lot |
| 15:53 | <Lachy> | hey, does anyone have any ideas for some images I could use to represent each of the design principles in the slides? |
| 15:54 | <Lachy> | it can be a bit abstract if there's nothing that directly conveys the concept |
| 16:06 | <Dashiva> | Priority of constituencies could be a pyramid |
| 16:09 | <Lachy> | Dashiva, interesting idea |
| 16:11 | <Lachy> | FYI, whatwg blog article was just republished here http://css.dzone.com/news/html-5-reverse-ordered-lists - I hope it gets some intersting comments |
| 16:13 | <Dashiva> | I'm tempted to suggest an crossed out ivory tower, or in one of those "do not" traffic signs, but I can't decide which one it should be :) |
| 16:14 | <Lachy> | for which principle? |
| 16:15 | <Lachy> | I have this, or something similar, for pave the cowpaths http://www.flickr.com/photos/adactio/298913307/ |
| 16:19 | <takkaria> | hm, looks like html5 doesn't describe how to handle whitespace in <a href=""> |
| 16:22 | <Lachy> | takkaria, does it need to define any special handling? |
| 16:23 | <Lachy> | do you mean within the element's text content, or its attributes, or something else? |
| 16:23 | <takkaria> | within the href="" attribute in particular |
| 16:24 | <Lachy> | like handling spaces as %20? |
| 16:24 | <takkaria> | yeah |
| 16:24 | <takkaria> | and trimming leading spaces |
| 16:24 | <takkaria> | Fx3 seems to ignore all but one trailing space too |
| 16:25 | <takkaria> | (in the DOM) |
| 16:25 | <takkaria> | but then following that link strips all trailing and leading spaces |
| 16:28 | <takkaria> | I'd imagine it's required for interopability too |
| 16:29 | <gsnedders> | Yeah, it certainly is. |
| 16:30 | <gsnedders> | WebKit's URL class always strips leading/trailing spaces, no matter where it comes from |
| 16:32 | <Dashiva> | Lachy: Support Existing Content, Solve Real Problems, or Handle Errors |
| 16:33 | <Dashiva> | Maybe all three? :) |
| 16:34 | <Lachy> | for support existing content, I think I'm going with a bridge. |
| 16:35 | <Lachy> | Either this one http://www.flickr.com/photos/hb2/288721287/ or this one http://www.flickr.com/photos/andyinnyc/421695304/ |
| 16:37 | <takkaria> | could someone with Safari around check to see what the href="" attribute looks like in the DOM at http://tinyurl.com/5wx9wa? |
| 16:40 | <Lachy> | takkaria, href=" http://google.com/search?q=hello world " |
| 16:41 | <takkaria> | thanks |
| 17:23 | <takkaria> | it's interesting how the number of private mails Hixie has is roughly the same as it was last October |
| 17:23 | <takkaria> | (for html5 issues, anyway) |
| 18:41 | <Lachy> | damn. Searching flickr for photos to use in my presentation that use creative commons licences, but don't use share-alike (cause copyleft licences suck), is difficult. |
| 18:47 | <jgraham> | Lachy: Copyleft would just mean that we had to make the slides CC licensed too, right? that seems acceptable |
| 18:47 | <jgraham> | (Well CC SA specifically) |
| 18:48 | <Lachy> | jgraham, I'd prefer to make it public domain and use CC by or by-nc licenced photos |
| 18:51 | <jgraham> | I don't mind either way but if not making it SA makes it worse it seems worth making it SA. I'm struggling to think of any situation in which not-SA would be a significant advantage |
| 18:52 | <jgraham> | Like when would someone try to distribute a derived-work under a more restrictive license? |
| 18:58 | <Lachy> | maybe I could get away with using share-alike licenced photos, since it's possible to mix public domain with anything, and using an SA licence doesn't on some parts doesn't change that. |
| 18:58 | <Lachy> | it would just have to be clear that photos use their own licences anyway, none of which are more restrictive than SA |
| 18:59 | <Lachy> | but I'd prefer to avoid that whole issue since IANAL and copyright licences are confusing |
| 19:00 | <Philip`> | It just means you can license the composite work only as CC, which is alright if it's composed of non-CC things that still allow to license them as CC |
| 19:00 | <Philip`> | (The only problem is when you're using something that requires the composite to be licensed as CC, and something else that requires it to be licensed as something else different) |
| 19:02 | <Philip`> | (*different and incompatible) |
| 19:02 | <Lachy> | I want everything but the photos to be free from copyright, so if someone took out the photos and replaced them with their own, there would be no restrictions |
| 19:03 | <Philip`> | You could just not care about copyright at all, and use whatever pictures look good, because it's not like anyone is going to get arrested for presentation slide piracy :-) |
| 19:04 | <Lachy> | since the slides will be published on the web afterwards, I'd rather avoid any possible complications |
| 19:05 | <Lachy> | any suggestions for what to use for Degrade Gracefully? |
| 19:05 | <jgraham> | Lachy: IANAL but IIRC it's pretty hard to public domain your work. Like harder than saying "this is public domain". So you're better off picking a suitable license that grants people the permissions that you want to give them. |
| 19:06 | <Lachy> | jgraham, it's not hard to say something is in the public domain |
| 19:07 | <jgraham> | Lachy: The problem is, I think, that in certain jurisdictions just saying that doesn't make it so |
| 19:09 | <jgraham> | Lachy: see http://en.wikipedia.org/wiki/Wikipedia:Granting_work_into_the_public_domain for example |
| 19:09 | <Lachy> | so you can just say "This is in the Public Domain. You are free to use this for any purpose, without any restrictions" |
| 19:10 | <jgraham> | The second status there has some weight yes. Although you should probably be more clear |
| 19:10 | <Lachy> | http://creativecommons.org/licenses/publicdomain/ |
| 19:11 | <Lachy> | but I don't care that much. If someone doesn't understand what "Public Domain" means, or lives in a stupid juristiction, that's not my problem. They need to fix their country's laws. |
| 19:12 | <Dashiva> | What matters is that Lachy isn't going to prosecute anyone :) |
| 19:13 | <jgraham> | Dashiva: It also matters if anyone can prosecute Lachy |
| 19:14 | <jgraham> | Lachy: Note that it says on the page before "Please note that the Public Domain Dedication may not be valid outside of the United States" |
| 19:14 | <jgraham> | (in the UK Public Domain doesn't mean anything, apparently) |
| 19:15 | <jgraham> | So it seems like the safest thing to do is to take the part of the PD license that reads "the Work may be freely reproduced, distributed, transmitted, used, modified, built upon, or otherwise exploited by anyone for any purpose, commercial or non-commercial, and in any way, including by methods that have not yet been invented or conceived." |
| 19:15 | <jgraham> | and just use that |
| 19:19 | <Lachy> | fine. I'll adjust my site-wide copyright licence to say something like that http://lachy.id.au/about/copyright |
| 20:35 | <annevk> | takkaria, did you test Opera 9.5?? |
| 21:42 | gsnedders | moves SP inline with HTML 5's stuff on character encoding names |
| 21:51 | <Lachy> | I have images for most design principles now. I just need Media Indpendence and Well Defined Behaviour |
| 21:51 | <Lachy> | any suggestions? |
| 21:52 | <gsnedders> | Lachy: I want to say some sort of crash for Well Defined Behaviour |
| 21:52 | <hdh> | is the cowpath suitable for well-defined? |
| 21:52 | <hdh> | I don't have the link here to check the licence |
| 21:52 | <Lachy> | hdh, I'm using a cowpath for Pave the Cowpaths. |
| 21:52 | <Lachy> | gsnedders, I've got a train crash for Handle Errors |
| 21:53 | <Philip`> | That ought to be a dragon |
| 21:53 | <gsnedders> | Lachy: Someone being mad? |
| 21:53 | <Philip`> | (Guess it's hard finding photos of one, though) |
| 21:53 | <gsnedders> | (mad in a silly way) |
| 21:53 | <Lachy> | Philip`, a dragon for what and why? |
| 21:54 | <Lachy> | I thought of finding a photo of a sign with lots of rules on it, but couldn't find an appropriate one |
| 21:54 | <Philip`> | Lachy: For (non-)draconian error handling |
| 21:54 | <annevk> | Looking at the SVG minutes there hasn't been much progress on SVG in HTML |