| 07:27 | <krijnh> | Philip`: I have no idea (re the search.live.com referrers) |
| 09:35 | <hsivonen> | I got a huge email, so Mail.app has been busy with one message all morning |
| 09:36 | <hsivonen> | so I can't be distracted by the RDFa thread for a while still... |
| 09:36 | <Dashiva> | Well, unless you're a glutton for punishment... :) |
| 09:36 | <zcorpan> | hsivonen: great opportunity to get acctual work done :) |
| 09:37 | <Dashiva> | "We can't use classes because people might try to dereference them. Instead we use namespace URLs, which magically always can be dereferenced." |
| 09:37 | <annevk> | these "Digital Bazaar" people sure write a lot |
| 09:38 | <Hixie> | some of the needs of these rdfa people are things we should probably address |
| 09:38 | <Hixie> | i'm just not sure any of them actually require spec changes |
| 09:38 | <Dashiva> | "Only URIs are ever unique. A classname with all the same characters as a URI will magically cause collisions." |
| 09:38 | <Hixie> | but i'll get back to rdfa once i've dealt with the workers and parsing feedback |
| 09:38 | <hsivonen> | Dashiva: I have a solution for that one |
| 09:39 | <hsivonen> | Dashiva: defining that class names that contain a colon MUST be IRIs |
| 09:39 | <Dashiva> | "We know squat about why namespaces are bad." |
| 09:40 | <Dashiva> | hsivonen: That doesn't stop collisions, though. Anyone's free to type whatever they want |
| 09:40 | <jgraham> | Dashiva: I was going to ask how they plan to keep URIs both unique and dereferencable since it sems to me that atm having one property means that you cannot ensure the other, even assuming good faith |
| 09:40 | <Dashiva> | There is no URI police that will come to your house and stop you |
| 09:40 | <hsivonen> | Dashiva: sure |
| 09:40 | <Dashiva> | jgraham: Oh? |
| 09:40 | <hsivonen> | Dashiva: the same holds true for RDFa property attribute |
| 09:41 | <Dashiva> | hsivonen: Indeed |
| 09:41 | <jgraham> | Dereferenceable => URLs, right? and URLs imply DNS which has finite-time registrations. |
| 09:42 | <jgraham> | (in practice there are shorter term problems) |
| 09:42 | <Dashiva> | They work around that with purl et.al. |
| 09:42 | <annevk> | the long term problems are somewhat annoying too; if I die, will I need to set up a fund to keep annevankesteren.nl alive? |
| 09:42 | <Dashiva> | Yes |
| 09:43 | <jgraham> | Dashiva: So they basically build a second centralised registry |
| 09:43 | <hsivonen> | who is the purl overlord? |
| 09:43 | <Dashiva> | jgraham: Ssh, it's impolite to point out elephants |
| 09:43 | <jgraham> | In which case why bother putting http://purl.org at the front of everything |
| 09:43 | <jgraham> | Why not just have unique human-readable prefixes |
| 09:44 | <hsivonen> | jgraham: well, purl.org NS uris depend on 3 centralized registries that are independent, so it's decentralized in that sense |
| 09:45 | <hsivonen> | billyjack: I'm around now |
| 09:45 | <jgraham> | hsivonen: what are the 3 independent registries? Presumably there is only one body handling out purl.org uris, so purl.org, not dns becomes the bottleneck in that case |
| 09:45 | <jgraham> | ? |
| 09:46 | <hsivonen> | jgraham: purl.org itself (for its subspace), domain registry and the URI scheme registry |
| 09:47 | <billyjack> | hsivonen: can't remember what I wanted to ask you about now.. will ping you if/when I remember |
| 09:47 | <hsivonen> | billyjack: ok |
| 09:47 | <jgraham> | hsivonen: I think that's just three things that could go wrong, not three layers of redundancy |
| 09:47 | <hsivonen> | jgraham: yes. but the three things are distributed :-) |
| 09:48 | <jgraham> | "distributed modes of failiure" doesn't sound that confidence inspiring :) |
| 09:52 | <Hixie> | jgraham: i don't think hsivonen is defending RDFa :-) |
| 09:53 | <annevk> | RDF sounds very nice in theory though, still |
| 09:53 | <Dashiva> | annevk's question was interesting, though. Would they actually endorse a situation where an entire vocabulary becomes unusable because someone takes their webpage down? |
| 09:54 | <hsivonen> | I'd love to have a unified parsing model for microformats but without namespaces, CURIEs or overlong identifiers |
| 09:54 | <hsivonen> | my guess is that two to four letter prefixes disambiguate enough |
| 09:54 | <hsivonen> | IIRC, in the wikimedia commons SVG analysis, the namespace prefix collisions were: |
| 09:55 | <hsivonen> | 1) autogenerated ns1 |
| 09:55 | <hsivonen> | 2) cc changing its URI |
| 09:55 | <hsivonen> | 3) Adobe and Microsoft claiming the same one-letter prefix |
| 09:56 | <annevk> | so unified parsing means you can easily create a graph out of some nodes? |
| 09:56 | <Hixie> | i love that the response to 2) is "but because you can dereference it, it's ok! we can map them to each other!" |
| 09:56 | <annevk> | because after that you will still have to perform format specific operations, which are probably more complicated than the parsing |
| 09:57 | <Hixie> | the syntax introduces a problem that is solved by the syntax, i love it |
| 09:57 | <hsivonen> | annevk: I'd settle for a tree |
| 09:57 | <Hixie> | annevk: if you end up with a triple-store, you can use standard tools to do inference |
| 09:58 | <Hixie> | annevk: which supports things like mapping vocabularies to each other, or saying that anything that is a car is also a four-wheeled object |
| 09:58 | <Hixie> | annevk: so even if your tool originally only knew about wheels, it could work out that each car implied four weels |
| 09:58 | <annevk> | yeah, but you still need to create the "export to Vcard" function |
| 09:58 | <Hixie> | yeah well, i'm not sure i understand what the ui would really be |
| 09:59 | <annevk> | I don't think it solves that part (although some people suggest it does) |
| 09:59 | <hsivonen> | Hixie: one semweb enthusiast I used to have lunches with occasionally at least used to advocate a UI like NakedObjects |
| 09:59 | <hsivonen> | http://www.nakedobjects.org/home/index.shtml |
| 10:00 | <hsivonen> | (I don't think that makes a good UI, though.) |
| 10:01 | hsivonen | fails to find Havoc Pennington's essay that identifies "I can generate the UI" as a passing phase |
| 10:02 | <hsivonen> | here it is: http://www106.pair.com/rhp/free-software-ui.html |
| 10:04 | <jgraham> | hsivonen: Yeah, every time I hear about RDFa I think of that... |
| 10:07 | <Philip`> | A library that does everything your applications need except for generating the UI still sounds like a quite useful thing to have |
| 10:08 | <jgraham> | The other question I didn't ask about RDFa is why it will work when rel="home", "next" etc. have effectively failed |
| 10:09 | <Philip`> | What does "failed" mean? |
| 10:10 | Philip` | has never noticed them causing harm, and has found them to be useful when he uses them on his own pages |
| 10:10 | <jgraham> | Philip`: They aren't used that much and IIRC aren't supported by default in 99% of browsers |
| 10:11 | <jgraham> | In fact you even have to turn on support in Opera I think |
| 10:11 | <Philip`> | They're on by default in Opera |
| 10:11 | <Philip`> | or at least the 'next' one is |
| 10:13 | <jgraham> | I think mpt was right that the problem is that because they are only used for 1 in n pages people don't learn to look in the chrome for functions related to intra-site navigation |
| 10:13 | <Philip`> | (e.g. if I go to http://philip.html5.org/tests/canvas/suite/tests/fallback.multiple.html and use a 'forward' shortcut (ctrl+right, right-click-and-drag-rightwards, etc) then it goes to the next page) |
| 10:14 | <hsivonen> | feed autodiscovery got UI |
| 10:14 | <hsivonen> | rel=next is different from the rest |
| 10:15 | <mpt> | That never occurred to me before -- how did feed autodiscovery succeed while rel= mostly failed? |
| 10:15 | <jgraham> | Philip`: You're right of course. |
| 10:16 | <mpt> | or I should say, <link rel= failed -- <a rel= is useful, because you're not relying on browsers to do anything special |
| 10:16 | <mpt> | it's an optional extra, like microformat detection |
| 10:17 | <annevk> | http://ajaxian.com/archives/ubiquity-quicksilver-of-the-firefox-browser could benefit from RDFa annotated data |
| 10:17 | <wilhelm> | Philip`: You can also just press space on that page. It scrolls down, and when it can't scroll anymore, it goes to the next page. |
| 10:17 | <Philip`> | (I think I've seen Opera's next-page thing working far more often when it automatically determines an ordered list from Apache directory indexes and lets you jump from one file straight to the next in the directory, rather than from explicit rel=nexts) |
| 10:22 | <hsivonen> | wilhelm: that feature is awesome. (which is why I think rel=next is not like the rest) |
| 10:23 | <webben> | I guess feed discovery progressed for on-page links to not relying on-page links. |
| 10:23 | <webben> | *from on-page |
| 10:23 | <webben> | maybe if <a rel="next" and associated UI became commonplace |
| 10:23 | <annevk> | rel=next also causes a page to be prefetched in Firefox |
| 10:23 | <webben> | then <link rel="next" would work too |
| 10:23 | <wilhelm> | <link rel='next'> is a pain, though. <a rel='next'> would be much nicer. |
| 10:23 | <jgraham> | mpt: I would hazard a guess that it has something to do with the fact that you can predict which sites have feeds and the fact that they used the same icon that people had got used to on-page offpage |
| 10:23 | <annevk> | (similar to rel=prefetch) |
| 10:24 | <webben> | wilhelm: are feed icons nicer than feed autodiscovery? |
| 10:30 | <annevk> | Does anyone know how Firefox JIT compares with WebKit Squirrelfish? |
| 10:32 | <annevk> | seems to be faster |
| 10:33 | <Philip`> | Apparently it wins on SunSpider |
| 10:33 | <annevk> | right |
| 10:34 | <Philip`> | Meanwhile, the IE team has a blog post saying that JScript performance isn't that important really |
| 10:36 | <hsivonen> | Philip`: even though IE seems to need the most JS libraries between it and the app programmer :-) |
| 10:36 | Philip` | likes how http://home.kairo.at/blog/2008-08/how_fast_is_tracemonkey_in_real_world measures "stuff a real application might use" by calculating Mandelbrot fractals |
| 10:37 | hsivonen | wishes Apple shipped GNU versions of basic Unix tools |
| 11:10 | <hsivonen> | aargh. no wonder there were too many unique messages. part of the consolidation had failed |
| 11:10 | <hsivonen> | sigh |
| 11:15 | <Hixie> | man, justin's comments on workers were disturbing |
| 11:16 | <Hixie> | he had one example where he launched one worker PER PIXEL of a bitmap he wanted to process |
| 11:16 | <Hixie> | i'm scared of what authors will do now |
| 11:25 | <hsivonen> | hmm. I think Mail.app has just gone crazy on its own |
| 11:25 | <Lachy> | Hixie, in the note at the end of the alt text guidelines, "One way to think of alternative text is to think about what how you would read the page containing the image " - s/what how/how/ |
| 11:26 | <Hixie> | please send mail, i'm about to go to bed |
| 11:26 | <Hixie> | just a one liner to ian⊙hc is fine |
| 11:26 | <Hixie> | btw thanks again for that program |
| 11:26 | <Hixie> | whose name escapes me |
| 11:27 | <Lachy> | Requiem |
| 11:27 | <Hixie> | i have however found one major flaw with it |
| 11:27 | <Hixie> | right |
| 11:27 | <Lachy> | what? |
| 11:27 | <Hixie> | requiem |
| 11:27 | <Hixie> | after i unencrypted everything |
| 11:27 | <Lachy> | what flaw? |
| 11:27 | <Hixie> | i mentioned it to my girlfriend |
| 11:27 | <Hixie> | and her response was |
| 11:27 | <Hixie> | "oh! that means i can download DRMed songs again right?" |
| 11:28 | <Lachy> | heh |
| 11:28 | <Hixie> | i didn't really have a good reply! |
| 11:28 | <Hixie> | i mean, i don't want to encourage iTunes to continue with DRM |
| 11:29 | <Lachy> | just say no, because spending money on music with DRM only funds further development of DRM |
| 11:29 | <Hixie> | but it's hard to argue that the DRM is a problem if it can just be removed |
| 11:29 | <Lachy> | sure it can be removed, but it's illegal thanks to the DMCA |
| 11:29 | <Hixie> | i have a little cron job now that scans my library and decrypts anything that's newly added with drm and moves the old drm'ed file out of the way and renames the new file, all automatically |
| 11:29 | <Hixie> | it's pretty sweet |
| 11:30 | <Hixie> | the "breaking encryption is illegal" aspect of the DMCA is inane |
| 11:31 | <webben> | OT: is there a canonical straightforward explanation of the limitations of DTDs for describing the syntax of XML vocabularies (versus schemas of XSD or RNG varieties, or anything else for that matter)? |
| 11:31 | <Hixie> | webben: DTDs can't describe anything but trivial attribute value formats |
| 11:32 | <hsivonen> | webben: not canonical, but there's http://hsivonen.iki.fi/thesis/html5-conformance-checker#dtd |
| 11:33 | <Hixie> | anyway |
| 11:33 | <Hixie> | nn |
| 11:33 | <hsivonen> | nn |
| 11:37 | hsivonen | restores Mail and AddressBook-related ~/Library files from backup |
| 11:37 | hsivonen | wishes Mail.app were more robust |
| 11:41 | <webben> | hsivonen: ta :) |
| 12:00 | <hsivonen> | zcorpan: multiline scripts now work in the HTML5 live DOM |
| 12:22 | <zcorpan> | hsivonen: cool |
| 12:22 | <zcorpan> | hsivonen: now the question is whether we should change to match webkit and firefox or if they should change to match us... consider that %0D%0A is what is sent to the server |
| 12:23 | <hsivonen> | what does IE do? |
| 12:23 | <zcorpan> | IE uses \r (but sends %0D%0A to the server) |
| 12:24 | <hsivonen> | CRLF sucks, and LF is more canonical than CR, so I'd prefer the Gecko/WebKit way |
| 12:24 | <hsivonen> | without proper compat arguments |
| 12:24 | <zcorpan> | we haven't had any compat bugs on this until livedom |
| 12:24 | <zcorpan> | so either way should work fine |
| 12:25 | <annevk> | path of least resistance seems for us to match Gecko/WebKit |
| 12:25 | <zcorpan> | yeah |
| 12:28 | <hsivonen> | hmm. myspace localizers have localized id=bodytype (or something like that) to id=Vartalon tyyppi without quotes |
| 12:28 | <hsivonen> | wow. MySpace has totally clueless id attribute use |
| 12:29 | <hsivonen> | it's a wonder their markup "works" at all |
| 13:06 | <zcorpan> | Hixie: would it be possible to include an Archived-At or X-Archived-At header in whatwg list emails? |
| 13:06 | <annevk> | he's just running some package from dreamhost |
| 13:11 | <zcorpan> | does that mean "no"? |
| 13:12 | <annevk> | likely |
| 13:12 | <annevk> | (it has been asked before, too) |
| 13:13 | <zcorpan> | ok |
| 13:14 | Philip` | sees some discussion of it in http://wiki.list.org/display/DEV/Stable+URLs |
| 13:16 | annevk | was just about to past that |
| 13:17 | <annevk> | seems a non-trivial problem |
| 13:20 | <annevk> | s/past/paste/ |
| 13:38 | <Philip`> | Hmm, my experiments indicate that Googlebot is surprisingly limited in where it looks for URLs - I only see <a href>, <embed src> and <script src> |
| 13:38 | <Philip`> | Yahoo looks at <a href>, <a src> (?!), <embed src>, <link href> and <object data> |
| 13:38 | <Philip`> | msnbot only looks at <a href>; msnbot-media also does <embed src> and <img src> |
| 13:38 | <Philip`> | Heritrix looks at <* cite>, <* data>, <* href>, <* lowsrc>, <* profile>, <* src> for any * |
| 13:40 | <Philip`> | (Of course they might be doing sufficiently clever things to break my experiments) |
| 13:56 | <annevk> | hmm, #html-wg is starting to look like public-html :) |
| 14:04 | <hsivonen> | Hixie: I'm seeing a number of pages that have not parse errors. So far, I'm not seeing pages that don't have validation errors |
| 14:08 | <Philip`> | Will you be able to easily determine how many don't have validation errors except for some given set of errors? |
| 14:08 | <hsivonen> | Philip`: "easily" will depend of the ratio of RAM and data after the validation run |
| 14:09 | <hsivonen> | Philip`: but grep -v should work |
| 14:09 | <hsivonen> | I did a couple of things that will drastically reduce the data |
| 14:10 | <hsivonen> | I made the validation worker consolidate "bad value 'foo'" messages properly. (it had a bug last night) |
| 14:10 | <hsivonen> | and I picked at most one URL per hostname |
| 14:21 | <hsivonen> | is there an HTML 4 "community" now? |
| 14:21 | <hsivonen> | "community" sounds even more hollow these days than "economy" |
| 14:32 | <zcorpan> | aaronlev: hey |
| 14:33 | <zcorpan> | bah he always disconnects when i ping him :( |
| 15:12 | <zcorpan> | aaronlev: ping |
| 15:13 | <Philip`> | A lot of people seem to be complaining about how IE8's InPrivate Blocking mode (which blocks third-party items that are included on more than ten different sites you visit) is going to break analytics and advertising and will cause the web industry to collapse |
| 15:14 | <Philip`> | I guess they're quite significantly distinct from the people who complain about <a ping> |
| 15:15 | <aaronlev> | hi zcorpan |
| 15:15 | <zcorpan> | aaronlev: i tried to email aria-ua-impl last night |
| 15:15 | <zcorpan> | aaronlev: but it bounced |
| 15:16 | <aaronlev> | did you join? |
| 15:17 | <zcorpan> | yeah |
| 15:18 | <zcorpan> | aaronlev: are the logs public? |
| 15:21 | <aaronlev> | zcorpan: i don't know |
| 15:21 | <aaronlev> | i can make you an administrator |
| 15:21 | <aaronlev> | give me a minute |
| 15:42 | <takkaria> | hsivonen: oh, what's your more interesting URL set? |
| 15:54 | <gsnedders> | benoitc: My navigating skillz are teh 1337. |
| 15:59 | <gsnedders> | benoitc: sorry, wrong person. |
| 16:00 | <benoitc> | np |
| 16:19 | <annevk> | we really need better <input type=file> + XMLHttpRequest support (or async <input type=file max=100>) |
| 16:20 | <annevk> | having to use the commandline to upload Flickr photos easily under Ubuntu/Mac is just silly |
| 16:46 | <takkaria> | does no-one else thing that role="photo" is a mad suggestion? |
| 16:47 | <takkaria> | role="logo" or role="drawing" |
| 16:47 | <takkaria> | since they're not roles, blatantly, they're descriptions |
| 16:48 | <takkaria> | a photo could be role="decorative" but role="photo" seems a really jarring misuse of the word "role" |
| 16:48 | <Dashiva> | It seems more like a subtle attempt at sabotaging role |
| 16:48 | <takkaria> | mind you, all this role business for img elements is astronauting anyway |
| 16:59 | <gsnedders> | gsnedders's advice for today: Don't fall on to solid wood floors. |
| 17:13 | <Xenos> | Concrete floors arn't good either, I hear |
| 17:14 | <Philip`> | Best not to fall on any floor at all, in my experience |
| 17:16 | <gsnedders> | normalize-space() is expensive. |
| 17:31 | <annevk> | whoa, flood of RDF e-mail |
| 17:35 | <annevk> | though a lot of it is due to the dubious cross-list posting (www-archive and whatwg) |
| 17:40 | <annevk> | wow, my own e-mail to an internal mailing list got marked as spam |
| 19:07 | <annevk> | For those with W3C Member access: http://lists.w3.org/Archives/Member/tag/2008Aug/thread.html#msg67 |
| 19:12 | <malde_> | Hi, |
| 19:13 | <malde_> | is my reading of the spec correct that within a sigle window only one transaction can execute simultaneously? |
| 19:13 | <Dashiva> | "The concept of namespaces is something that most regular folks can grasp if you explain it to them in the correct way." |
| 19:13 | <Dashiva> | How many understand namespaces, and how many just understand magical prefixes? |
| 19:15 | <annevk> | malde_, given that two windows can interact with the same database I don't think that's true |
| 19:15 | <malde_> | annevk: I was asking about a single window |
| 19:16 | <annevk> | I realize that and I don't think it's something that goes per window given that multiple windows can interact with the same database |
| 19:16 | <annevk> | (but I haven't checked that part recently) |
| 19:16 | <malde_> | like in db.transaction(...);db.transaction(); the second transaction wont be executed until all callbacks of the first transaction fired |
| 19:16 | <malde_> | It seems that webkit implements it that way |
| 19:17 | <annevk> | that sounds correct, yes |
| 19:17 | <Hixie> | malde_: it's per database, not per window |
| 19:18 | <annevk> | morning |
| 19:18 | <malde_> | OK, so it should be save to store the transaction handle inside a global variable? |
| 19:19 | <Hixie> | malde_: i'm not sure i'd recommend that, but it depends on what you're doing really |
| 19:19 | <Hixie> | zcorpan: in case you read this, i'd love to be able have X-Archived-At (i suggestd it for w3c originally) but i'm just using old stock mailman that dreamhost runs |
| 19:20 | <malde_> | I wrote an OR mapper that work with both html5 db and gears and I wont to get the tx out of the ORM's API |
| 19:20 | <annevk> | Hixie, it's standardized as Archived-At now, fwiw |
| 19:20 | <annevk> | (and W3C switched to using that) |
| 19:22 | <jgraham> | Dashiva: I thought "We realized that it would be much easier for people to understand" |
| 19:22 | <jgraham> | "namespaces if we used URIs as the method of namespace expression" |
| 19:22 | <Dashiva> | So "Namespaces are much easier to understand if we ignore that pesky indirection thing" |
| 19:23 | <jgraham> | Dashiva: I think the point was "Namespaces are much easier to understand if we use URIs to denote the namespace rather than human-readable strings" |
| 19:24 | <Dashiva> | And then bring CURIEs into it |
| 19:25 | <Philip`> | Dashiva: Regular folks can use the magical prefixes that they copy-and-paste from the instructions given to them by the namespace experts who understand how to use the power and flexibility to provide all the desired features and avoid collisions and everything |
| 19:26 | <Philip`> | But I don't think that reasoning works, because there are far too few experts, and so regular folks make it up as they go along and get it all wrong |
| 19:28 | <jgraham> | Hmm. the RDFa emails seem to se using URI and URL interchangably in a situation where the difference is important |
| 19:32 | <Dashiva> | Philip`: And we all know the non-experts would never forget to copy the enclosing element defining the CURIE prefixes :) |
| 19:33 | <Philip`> | Dashiva: The same non-experts who write XHTML pages with xmlns="http://www.w3.org/1999/xphp";? Surely not! |
| 19:45 | <hsivonen> | takkaria: the more interesting set is a set of URL from Google that shouldn't have the kind of front page bias dmoz likely has |
| 19:45 | <hsivonen> | s/URL/URLs/ |
| 19:45 | <takkaria> | ah, I see :) |
| 19:46 | hsivonen | notes that the RDFa examples omit ns declarations |
| 19:48 | <annevk> | and even when they omit those they still look far more complex than the microformats they try to emulate |
| 19:51 | <Philip`> | That's a problem with designing scalable solutions: you want examples to be as small as possible, and much smaller than real-world uses are likely to be, so the scalability is not apparent to people who only look at the examples |
| 20:33 | <gsnedders> | IE8b2 is available, FWIW |
| 20:37 | <Philip`> | Ooh, a bug-hunting opportunity |
| 20:43 | <Philip`> | Seems unusual for them to talk about improving the performance of Gmail, rather than of any Microsoft site |
| 20:56 | <gsnedders> | I have only one item on my to-do list left for spec-gen 1.0RC1 |
| 20:56 | <gsnedders> | checking all the docs are right. |
| 20:57 | <hsivonen> | hmm. IE8b2 EULA prohibits publication of benchmark results |
| 20:57 | <hsivonen> | that sucks |
| 20:58 | <webben> | seems a bit harsh |
| 20:58 | <webben> | (also seems highly unlikely to be an effective deterrent) |
| 20:58 | gsnedders | thinks hsivonen complained about that when IE8b1 was released (with the slight change of the IE8b1 EULA being the EULA in question) |
| 20:58 | <hsivonen> | webben: harsh to prohibit it or harsh to say it sucks? |
| 20:58 | <webben> | I mean the prohibition is harsh. |
| 20:59 | <webben> | Seems fair to say it sucks. :) |
| 21:00 | <Philip`> | VMware from version 4 (I think) prohibits distribution of benchmark results too, but then the VMware people complained that it was unfair when other virtualisation people distributed benchmarks comparing against VMware 3 |
| 21:01 | <Philip`> | which is kind of their own fault |
| 21:02 | hsivonen | wonders if one presses cancel to the IE8b2 install when a cancel button is offered after the installer has said it removed previous IE |
| 21:02 | hsivonen | wonders what happens if.. |
| 21:05 | Philip` | had to manually uninstall IE8b1 and reboot and then see Windows fail to start up and offer to restart in safe mode but then it worked the next time |
| 21:06 | <hsivonen> | to me, it seems like a bad idea to offer a cancel button in the second phase of a rebooting installer |
| 21:07 | <hsivonen> | whee. a second reboot |
| 21:08 | <Philip`> | It's rather odd that in these modern times it can take twenty minutes to install a web browser |
| 21:08 | <Dashiva> | Maybe it's more than just a browser :) |
| 21:09 | <annevk> | if only hardware improved installing a browser sure would be fast |
| 21:10 | <hsivonen> | What Gordon giveth, Bill taketh. |
| 21:11 | <hsivonen> | eww. Validator.nu is ugly in IE8b2 |
| 21:11 | <gsnedders> | hsivonen: can you see if http://gsnedders.com is still badly screwed up? |
| 21:12 | <hsivonen> | gsnedders: it isn't |
| 21:12 | <hsivonen> | except Windows font rendering is hideous |
| 21:12 | <hsivonen> | (and some people wanted Safari on Windows to use it) |
| 21:13 | <Hixie> | so does ie8 work in the live dom viewer yet |
| 21:13 | <hsivonen> | Hixie: it at least doesn't render the link to it on damowmow portal |
| 21:14 | <hsivonen> | Hixie: it works! |
| 21:16 | <Hixie> | woot |
| 21:16 | <Philip`> | "Do you want to discover websites you might like based on websites you've visited?" - hmm |
| 21:17 | <Philip`> | Things like "<b style>" still don't show the attribute in the DOM view, though |
| 21:18 | <Hixie> | is that because the attribute isn't there? |
| 21:18 | <Hixie> | i just want it to reflect the dom |
| 21:18 | <Philip`> | Hmm, it works if you do <b style=color:black> |
| 21:19 | <Philip`> | where "works" means it shows B style="COLOR: black" in the DOM view |
| 21:19 | <Hixie> | ok i need to eat before i die of starvation |
| 21:19 | <Hixie> | and then i guess i'll respond to this svg proposal for parsing |
| 21:19 | <Philip`> | but <b style=color:blac> makes the attribute disappear |
| 21:19 | <Hixie> | Philip`: that's probably an accurate representation of the DOM |
| 21:21 | <annevk> | IE has some weird stuff going on |
| 21:22 | <annevk> | (though it probably made sense at the time (I hope)) |
| 21:26 | <Philip`> | Hmm, I can't find any situations in which authoritative=true does anything at all |
| 21:31 | <Philip`> | (nor does mshtml.dll contain the string "authoritative", so maybe they just haven't implemented it?) |
| 21:32 | <Philip`> | (unless it's implemented in some other component of Windows) |
| 21:32 | <Philip`> | (but it still doesn't seem to do anything at all in practice) |
| 21:36 | gsnedders | fails to find any further issues blocking release of the spec-gen |
| 21:38 | <Philip`> | gsnedders: It needs a clever name before it's released |
| 21:39 | <Philip`> | "spec-gen" is a bit generic :-p |
| 21:39 | <gsnedders> | :P |
| 21:39 | <annevk> | also, drop the hyphen |
| 21:40 | <annevk> | specgen is easier |
| 21:41 | <annevk> | (you could call it "chameleon" referring to how you mimicked the original) |
| 21:45 | <gsnedders> | annevk: chameleon is taken :P |
| 21:45 | <annevk> | not for spec generators |
| 21:46 | <annevk> | I should ask, what do you mean with "taken"? |
| 21:46 | <gsnedders> | annevk: Used by other software |
| 21:47 | <annevk> | "Chameleon Specgen" isn't |
| 21:47 | <annevk> | http://www.google.com/search?q=%22chameleon%20specgen%22 |
| 21:47 | <annevk> | though that's not really a good name :) |
| 21:47 | <gsnedders> | Anole? |
| 21:48 | <gsnedders> | Used too, actually |
| 21:48 | <annevk> | call it Anolis then or something like that |
| 21:49 | <gsnedders> | Inkfish? |
| 21:51 | <BenMillard> | how about Geoffrey's Specification Generator (GGen/ggen)? |
| 21:51 | <annevk> | I don't really see the relation |
| 21:52 | <annevk> | "ggen is a 2-D geometry generator expressed as a c++ class library." |
| 21:52 | <Philip`> | Use the Hebrew alphabet for the name |
| 21:53 | <gsnedders> | BenMillard: Re: you're ability to navigate — I can fine. We should make it there and back fine :) |
| 21:53 | <Philip`> | (Er, not the whole alphabet, just some letters from it) |
| 21:53 | <BenMillard> | gsnedders, coolies :) |
| 21:53 | <BenMillard> | gsnedders, I take it we'll stay at the hotel annevk did 2 years ago? |
| 21:54 | <gsnedders> | BenMillard: What hotel did annevk stay at two years ago? :) |
| 21:54 | <BenMillard> | gsnedders, oh, forwarded to you |
| 21:55 | <Lachy> | Philip`, do you mean something like this? א |
| 21:55 | <Lachy> | no idea what that means, but it's a hebrew character aparently |
| 21:55 | <annevk> | oops, forgot to cc gsnedders |
| 21:56 | gsnedders | feels too lazy to bother to click the "English" link |
| 21:56 | <gsnedders> | BenMillard: We have our TV for the PS2! :) |
| 21:57 | <BenMillard> | gsnedders, good, the trip will be worthwhile after all! :D |
| 21:57 | <gsnedders> | umm¬ |
| 21:57 | <gsnedders> | *umm… |
| 21:57 | <gsnedders> | "3eme personne dans la chambre, (sans lit) " |
| 21:57 | <gsnedders> | So, what? You sleep on the floor? |
| 21:57 | <gsnedders> | (that means three people in the bedroom, without bed) |
| 21:58 | gsnedders | wonders if we could really book online when we only want the extra bed for three nights |
| 21:59 | <gsnedders> | heh. They're English is in parts on the website a literal translation from the French and odd :) |
| 21:59 | <gsnedders> | *Their |
| 21:59 | <BenMillard> | looking here, I see what you mean: http://www.villaparisiana.com/price.htm |
| 22:00 | <BenMillard> | "3rd person in room, (without bed) : 6 euros" |
| 22:00 | <gsnedders> | But also, "Lit supplémentaire" |
| 22:00 | <gsnedders> | (extra bed) |
| 22:00 | <gsnedders> | (however they translate that) |
| 22:00 | <BenMillard> | "Extra bed : 15 euros" |
| 22:00 | <BenMillard> | win! |
| 22:00 | <gsnedders> | totally. |
| 22:00 | <annevk> | just means you pay EUR 21 for the third person |
| 22:00 | <gsnedders> | BenMillard: Can you justify the cost of coming for a whole week now? :P |
| 22:01 | <annevk> | I suggest taking the breakfast as well |
| 22:01 | gsnedders | will undoubtedly get up too late for it :) |
| 22:02 | <gsnedders> | Last hotel I was at, I made breakfast zero out of three times. |
| 22:02 | <BenMillard> | I absolutely *must* eat a breakfast; ideally hot chocolate and cereal with a banana |
| 22:02 | <gsnedders> | BenMillard: Now, do we try and get three people in a small room or not? |
| 22:03 | <annevk> | I usually miss a few minutes of the relevant meeting :) |
| 22:03 | gsnedders | had no breakfast today :P |
| 22:03 | <BenMillard> | gsnedders, I don't think we need to slum it too badly |
| 22:03 | gsnedders | got into school late anyway |
| 22:03 | <gsnedders> | Nor do I :) |
| 22:04 | <gsnedders> | I mean, even the big rooms are cheaper than half what you'd pay sharing a room at the TPAC hotel |
| 22:04 | <BenMillard> | gsnedders, rooms don't look amazingly generous: http://www.villaparisiana.com/rooms.htm |
| 22:04 | <BenMillard> | so their idea of "small" might be "cupboard" |
| 22:04 | <gsnedders> | BenMillard: That was my thought :) |
| 22:04 | <BenMillard> | medium or large sounds good to me |
| 22:04 | <gsnedders> | We're middle/low season |
| 22:04 | <gsnedders> | (as we're there over the change) |
| 22:05 | <BenMillard> | gsnedders, yeah |
| 22:05 | gsnedders | still hasn't bothered clicking that English link |
| 22:05 | <BenMillard> | as well as 3 people (I'm about 6ft tall) we also must consider 3 people's bags of stuff |
| 22:06 | <BenMillard> | you two are there for like a week, so that volume may be non-trivial |
| 22:06 | <gsnedders> | And I'm away from home for two weeks |
| 22:06 | <BenMillard> | I'm thinking Large Room to be safe (and their idea of "Large" may not be that big anyway) |
| 22:06 | <BenMillard> | plus we're adding an extra bed |
| 22:06 | <gsnedders> | (Though that won't be much issue — I'll be staying with my uncle, so washing and the like can be done) |
| 22:06 | <gsnedders> | (so I only really need a week's stuff) |
| 22:07 | <BenMillard> | gsnedders, Large Room might get a better TV and more power outlets, too |
| 22:09 | <gsnedders> | 528EUR in total, by my calculation |
| 22:09 | <gsnedders> | (big room, inc. breakfast) |
| 22:09 | <gsnedders> | (exc. the 0.40 EUR tax per day per person) |
| 22:09 | <gsnedders> | EUR 176 if we split it equally, which is still less than one night in the TPAC hotel. |
| 22:09 | <gsnedders> | Heh. |
| 22:10 | <gsnedders> | BenMillard: Power outlets may be useful with three geeks :) |
| 22:10 | <BenMillard> | gsnedders, I have a mobile phone which eats battery life but no laptop |
| 22:10 | <gsnedders> | BenMillard: I have a mobile phone which doesn't eat battery life and a laptop |
| 22:10 | <gsnedders> | (which does) |
| 22:11 | <BenMillard> | gsnedders, splitting the price evenly seems fair enough to me, as I'll be making the room more cramped for the 2 nights I'm there, you'll be doing the meatspace navigation for me and I'm not footing the bill personally |
| 22:11 | <gsnedders> | BenMillard: :) |
| 22:11 | <BenMillard> | plus I'm bound to whip you at GT and make you cry |
| 22:12 | <BenMillard> | :P |
| 22:13 | <gsnedders> | BenMillard: If I get back in practice at playing such games on the PS2… :) |
| 22:13 | gsnedders | has been playing Forza 2 on 360 a lot recently |
| 22:13 | <gsnedders> | it takes me a week or two to get fully up to speed on the PS2 controller |
| 22:13 | <gsnedders> | So, I guess I could start practising… :) |
| 22:14 | <gsnedders> | BenMillard: On Forza 2, in the time trial rankings, I'm #8xx, so rather high up :) |
| 22:16 | <BenMillard> | gsnedders, we can use different tyres and classes of car to balance things out, handicap starts, etc |
| 22:18 | <Xenos> | After the fifth day of PS2, smoking pot and eating junk food, someone goes "Wait, wasn't there a meeting we were supposed to go to?" |
| 22:19 | <BenMillard> | gsnedders, looking at the booking form, it seems 1 person books for everyone and we balance the money out afterwards? http://www.villaparisiana.com/contact_e.htm |
| 22:19 | <BenMillard> | ("Number of persons: 3" and "Number of rooms: 1") |
| 23:15 | <gsnedders> | BenMillard: different classes of cars has its limits. In GT3, I once beat a friend, driving an F1 car, with a Mini. |
| 23:15 | <gsnedders> | BenMillard: Yeah, so it seems |
| 23:16 | <BenMillard> | gsnedders, we can arrange how to square up the costs in e-mail |
| 23:17 | <gsnedders> | BenMillard: Yeah, seems sane |
| 23:17 | gsnedders | emails smedero with link to the logs |
| 23:17 | <BenMillard> | gsnedders, as I won't be there the whole time I guess one of you should do the booking |
| 23:17 | <Hixie> | i can drive trucks much more easily than sports cars in gta4 |
| 23:18 | <BenMillard> | Hixie, we're talking about Gran Turismo, but yes I have the same experience of past GTAs |
| 23:18 | <Hixie> | i usually prefer to do the races (against sports cars) in a bus, because the bus can take more damange. :-) |
| 23:18 | <gsnedders> | Hixie: Because of them skidding around more, due to their lack of weight? |
| 23:18 | <Hixie> | BenMillard: yeah, same kind of thing with GT |
| 23:18 | gsnedders | much prefers the sports cars, but he can cope with a madly light car |
| 23:19 | <BenMillard> | the original CD case for Grand Theft Auto describes the handling as "Hollywood" and "outrageous" which, to a greater or lesser extent, has been true in all editions of it |
| 23:19 | <Hixie> | gsnedders: partly, yes, but also with the bus i don't have to be as precise, and i can just plow through the competition and watch them splash into walls and buildings, while i don't even slow down |
| 23:19 | <gsnedders> | There again, I don't manage to be within the top 400 around the Nürburgring in Forza 2 by being a bad driver :) |
| 23:20 | <gsnedders> | Hixie: Peh. Being precise ain't hard. Though driving buses is fun. |
| 23:21 | <BenMillard> | Hixie, do you have and of the 3D GTA's PC? if so, I did a "handling overhaul" for each of them as a mod: http://projectcerbera.com/gta/3/handling-overhaul http://projectcerbera.com/gta/vc/handling-overhaul http://projectcerbera.com/gta/sa/handling-overhaul |
| 23:21 | <BenMillard> | s/and of/any of/ |
| 23:21 | <Hixie> | nice |
| 23:21 | <Hixie> | i just have gta:vc and sa for ps2 and gta4 for ps3 |
| 23:22 | <Hixie> | haven't played my ps3 in months, it needs to be set up in the new room |
| 23:22 | <BenMillard> | Hixie, you're welcome to join the PS2 Gran Turismo part with gsnedders and I :) |
| 23:22 | <Hixie> | :-) |
| 23:22 | <BenMillard> | I'll be bringing my console as it's in near-perfect running order |
| 23:22 | <BenMillard> | feel free to bring memory cards |
| 23:22 | <gsnedders> | BenMillard: You have two controllers, I assume? |
| 23:22 | <Hixie> | be aware of PAL vs NTSC issues |
| 23:22 | <BenMillard> | gsnedders, yes |
| 23:22 | <gsnedders> | Hixie: We're both in PAL Europe :) |
| 23:22 | <Hixie> | oh right, meeting is in europe |
| 23:23 | <Hixie> | cool! |
| 23:24 | gsnedders | ought to bring NFS:HP2 (IMO, the best NFS game) |
| 23:24 | <gsnedders> | Crazy Taxi is scratched :( |
| 23:25 | <Hixie> | well if y'all are going to be bringing consoles, that's certainly going to make it more interesting |
| 23:27 | <gsnedders> | Bringing a 360 by air is a bit too complicated, so I won't be. |
| 23:27 | <gsnedders> | But I'll probably bring one or two PS2 games :) |
| 23:27 | <BenMillard> | gsnedders, on closer inspection I have a single PS2 controller and two PS controlls |
| 23:27 | <gsnedders> | BenMillard: k, if you remind me, I can bring a PS2 controller |
| 23:27 | <Dashiva> | I have a ps2 controller that sends random signals without me pressing buttons |
| 23:27 | <gsnedders> | That could be fun. |
| 23:27 | <Dashiva> | I'm considering enrolling it in a fight game tournament |
| 23:28 | <Hixie> | i have a slim pal ps2 i can bring if necessary |
| 23:28 | <Hixie> | and a ps2 controller and some games (like worms 3d, which is great for parties) |
| 23:29 | <Hixie> | but the ps3 is bigger than my backpack, which is all i'm planning on bringing |
| 23:29 | <BenMillard> | Hixie, I have the slim one |
| 23:29 | <gsnedders> | Hixie: I think the simple solution is we let BenMillard bring the console :) |
| 23:30 | <Hixie> | i'm just sayin' :-) |
| 23:30 | <BenMillard> | unless you know how to connect 2 together, which could be awesome |
| 23:30 | <BenMillard> | although one 1 TV...hmm |
| 23:30 | <BenMillard> | *only 1 TV |
| 23:30 | <gsnedders> | For GT3 at least you can use the Firewire port… |
| 23:30 | <gsnedders> | But I don't think that's on the slim console |
| 23:32 | <gsnedders> | GT4 supports LAN with ethernet |
| 23:45 | <BenMillard> | LAN requires 1 TV per player: http://www.us.playstation.com/Content/OGS/SCUS-97328/Site/gt4manual.pdf |
| 23:46 | <BenMillard> | (page "twenty") |
| 23:53 | <gsnedders> | peh. |
| 23:55 | <BenMillard> | Hixie, don't bother bringing your PS2. we'll only have 1 TV |
| 23:56 | <Philip`> | Bring a TV |
| 23:56 | <BenMillard> | lol |
| 23:56 | <Philip`> | as long as it's flatscreen |
| 23:56 | <gsnedders> | Actually, I have a video capture thing for my computer! |
| 23:57 | <Philip`> | Do any laptops have video-in that's compatible with the PS2's outputs? |
| 23:57 | <gsnedders> | Like, one that has next to no lag and can be used with a PS2 |
| 23:57 | <Philip`> | Ah |
| 23:57 | <gsnedders> | And it's what I normally use with my PS2 |
| 23:57 | <gsnedders> | Hixie: Do bother bringing your PS2. We have two TVs. |
| 23:58 | <gsnedders> | We also need two copies of GT4 to do that though |
| 23:58 | <gsnedders> | And mine is scratched, but mostly works |
| 23:58 | <gsnedders> | Just need to reset the console after every race :P |