| 00:07 | <f1lt3r> | Leave my sense of logic at the door eh? :) Sounds like a fun place to hang out. |
| 00:12 | <MikeSmith> | does anybody know if the WebM subset of Matroiska includes multitrack support? (for caption tracks, additional audio tracks, etc.) |
| 00:19 | <kinetik> | MikeSmith: http://www.webmproject.org/code/specs/container/ |
| 00:19 | <kinetik> | MikeSmith: no captions. you can include multiple tracks, but it's not clear how the non-default ones are supposed to be handled yet. |
| 00:19 | <zcorpan_> | "Initial WebM release does not support subtitles. |
| 00:19 | <zcorpan_> | WHATWG / W3C RFC will release guidance on subtitles and other overlays in HTML5 <video> in the near future. WebM intends to follow that guidance." |
| 00:30 | <nessy> | interesting! |
| 00:31 | <nessy> | kinetik: but it would technically be trivial to include the subtitling functionality of Matroska into WebM, right? |
| 00:32 | <kinetik> | nessy: fairly simple, yes |
| 00:32 | <nessy> | got a lot of reading specs to do :) |
| 00:32 | <kinetik> | nessy: i think the main problem right now is that the matroska spec specifies three or four subtitle formats |
| 00:32 | <nessy> | and they only want one? |
| 00:32 | <kinetik> | ideally we'd want to support one, and that one being whatever is web friendly |
| 00:33 | <nessy> | I like the "we" in that sentence ;) |
| 00:34 | <f1lt3r> | lol |
| 00:34 | <nessy> | any mailing lists or stuff you'd recommend to join? where does the webm community hang out? |
| 00:34 | <kinetik> | nessy: http://www.webmproject.org/about/discuss/ |
| 00:35 | <nessy> | thanks! (not actually that easy to find...) |
| 00:36 | <f1lt3r> | zcorpan_, where can i find out more about w3c subtitle work? |
| 00:40 | <zcorpan_> | https://groups.google.com/a/webmproject.org/group/webm-discuss/browse_thread/thread/21ec1286b09e5616# |
| 00:40 | <zcorpan_> | f1lt3r: i think the work being referred to was removed from the w3c copy of the spec |
| 00:40 | <f1lt3r> | ok |
| 00:40 | <zcorpan_> | f1lt3r: but there's http://whatwg.org/html5#websrt |
| 00:41 | <zcorpan_> | (the previous link was unrelated) |
| 00:54 | <MikeSmith> | kinetik: thanks for the info |
| 00:54 | <MikeSmith> | and zcorpan_ |
| 00:58 | <Lachy> | nessy, kinetik, the plan is for WebM to eventually support whatever format is specced for use in HTML5, which at this stage, is the WebSRT proposal. |
| 00:59 | <kinetik> | Lachy: that's my understanding too, but it's not clear from webmproject.org |
| 00:59 | <Lachy> | kinetik, I know cause I've been involved in the discussions for the past few weeks |
| 01:03 | <kinetik> | Lachy: on seaotter? |
| 01:06 | <Lachy> | yes |
| 01:06 | <MikeSmith> | "Overall, VP8 appears to be significantly weaker than H.264 compression-wise" |
| 01:07 | <MikeSmith> | who's Jason Garrett-Glaser? |
| 01:07 | <kinetik> | one of the lead x264 (an h.264 encoder) developers |
| 01:08 | <MikeSmith> | ah, OK |
| 01:18 | <MikeSmith> | <snort> ASS gets my vote for best-named subtitle/caption format http://www.matroska.org/technical/specs/subtitles/ssa.html |
| 01:24 | <karlcow> | MikeSmith, Lachy: Do you know if there are asian device makers who would be annoyed by vp8. thinking about samsung, nintendo and sony for example. Or do they use chipsets from elsewhere. |
| 01:25 | <Lachy> | karlcow, I don't have any inside info on hardware vendors |
| 01:26 | <karlcow> | the list of people participating to the new webm consortium is missing some companies. Apple, Nokia, Samsung, Sony, it will be interesting to see how it evolves. Let's hope it doesn't turn into another war |
| 01:35 | <MikeSmith> | dunno know what any of those companies might think in regard to this news |
| 01:35 | <MikeSmith> | and wouldn't want to speculate |
| 01:36 | <MikeSmith> | and wouldn't say even if I did know anything :) |
| 01:37 | <karlcow> | hehe |
| 01:38 | <karlcow> | MikeSmith: I wouldn't hear what you would not have to say about it ;) |
| 01:43 | <wycats> | Hixie: from your perspective, what is the best way to ensure that forms post UTF-8 |
| 01:43 | <wycats> | the most reliable combination I can find is accept-charset, a hidden field with unicode character as value, and _charset_ to validate (if available) |
| 01:43 | <wycats> | is there anything else? |
| 01:43 | <wycats> | is this unreliable? |
| 01:44 | <Philip`> | Is the host document UTF-8? |
| 01:45 | <wycats> | Philip`: yes |
| 01:45 | <wycats> | Philip`: but the user can change it :/ |
| 01:45 | <wycats> | it seems that accept-charset + hidden field with unicode value results in all browsers (including IE) ignoring user-overridden charsets |
| 01:46 | <wycats> | I'm asking here because there was a lot of discussion on back-compat in the WF2 days around this topic (from what I can glean) |
| 03:09 | <MikeSmith> | if I were a tech reporter, I think right now I'd be asking for comments on WebM from some companies with serious business stakes in IPTV |
| 03:31 | <MikeSmith> | http://kynesim.blogspot.com/2010/04/interesting-times-for-ott-video.html is interesting |
| 03:31 | <MikeSmith> | "From a royalty regime point of view it's much more important to replace AC3 and MP3 with Vorbis than it is to replace HTML5 with Theora. Particularly since the MP3 licensors, in particular, are apt to charge punitive rates to implementations which take other formats too - this being one reason why IPTV broadcasts generally avoid mp3" |
| 03:42 | <othermaciej> | mp3 patents should expire within a few years... |
| 03:59 | <MikeSmith> | othermaciej: so I'm sure the patent holders will do everything they can to make as much money as possible on them during those few remaining years |
| 03:59 | <MikeSmith> | I would if I were them |
| 04:06 | <nimbupani> | Would the companies not try to extend the life of the patent? http://en.wikipedia.org/wiki/Evergreening |
| 04:09 | othermaciej | is not a patent law expert |
| 04:09 | <othermaciej> | MikeSmith: many of them are also patent holders on AAC and I expect they would not want to scare people off AAC by being dicks about <P3 |
| 04:43 | <MikeSmithX> | nimbupani: that's a depressing article |
| 04:43 | <nimbupani> | MikeSmithX: evergreening? |
| 04:43 | <MikeSmithX> | one more among the list of things I guess I should know about but would almost rather not |
| 04:43 | <MikeSmithX> | nimbupani: yeah |
| 04:44 | <nimbupani> | yeah thats how big pharmas make their money :( |
| 04:44 | <MikeSmith> | well, everybody's got a right to make money |
| 04:45 | <nimbupani> | indian companies making AIDS meds for 1/10th cost were sued out of existence coz of patent infringements |
| 04:45 | <MikeSmith> | well, that's definitely screwed up then |
| 04:45 | <nimbupani> | so big NGOs in Africa get grants to pay big pharmas LOTS of money to get these expensive meds and then give them for free for AIDS victims who are poor. |
| 04:45 | <MikeSmith> | I guess it could be argued that they are still playing by the rules |
| 04:46 | <MikeSmith> | and that it's the rules that need to change |
| 04:46 | <nimbupani> | MikeSmith: yes, exactly. |
| 04:46 | <nimbupani> | this patent law is screwed I think. |
| 04:47 | <MikeSmith> | I'm sure they use a good part of their profits to lobby against changing the rules |
| 04:48 | <nimbupani> | i think so too. |
| 04:49 | <nimbupani> | Plus US has all these Free Trade Agreements that let them sue companies in the countries they have FTAs with (for patent infringements) |
| 04:50 | <nimbupani> | the carrot in the case of Singapore was visa waiver for Singapore citizens and special H1B category for them, one of the things in return was stricter enforcement of anti-piracy practices (among others). |
| 05:00 | <MikeSmith> | sounds more like extortion than a carrot |
| 05:03 | <nimbupani> | :) |
| 09:26 | <hsivonen> | I downloaded the 64-bit Ubuntu WebM build of Opera, but it doesn't support WebM. How do I make sure the app doesn't load libs from the Opera installed from .deb and actually loads from the directory extracted from .tar.gz? |
| 09:27 | <hsivonen> | 32-bit WebM-enabled MozillaDeveloperPreview on Mac, WFM, on the other hand |
| 09:31 | <Lachy> | hsivonen, foolip should know, if he's around |
| 09:32 | <Lachy> | but I will try the build now, see if it works for me |
| 09:33 | <hsivonen> | I'm still on Karmic, FWIW |
| 09:35 | <foolip> | hi there |
| 09:35 | <foolip> | there's a problem with the libc version linked in the Linux builds |
| 09:35 | <hsivonen> | hi |
| 09:36 | <foolip> | if you put those plugins in GST_PLUGIN_PATH and do gst-inspect opera_vp8, do you see some plugin load failed warnings? |
| 09:36 | <Creap> | with ARIA boolean values, does that mean I should have aria-haspopup=true, aria-haspopup=aria-haspopup or just aria-haspopup in html5? |
| 09:36 | <Lachy> | foolip, the build kind of works for me in Ubuntu. |
| 09:36 | <annevk> | Creap, I think ARIA is different from boolean attributes as defined in HTML |
| 09:37 | <annevk> | Creap, we gave Last Call comments to that effect |
| 09:37 | <foolip> | Karmic just might be too old, I built on Lucid |
| 09:37 | <annevk> | Creap, I lost track as to whether they were addressed or not |
| 09:37 | <Lachy> | foolip, but the audio seems to drop out after a few seconds, and comes back only after I seek. |
| 09:37 | <Creap> | ok |
| 09:37 | <foolip> | hsivonen: if that's the issue, then we know about it (too late) and are going to fix it |
| 09:38 | <zcorpan_> | Creap: it should be =true according to the aria spec, iirc |
| 09:38 | <foolip> | Lachy: low bandwidth? |
| 09:38 | <Lachy> | no, playing from localhost |
| 09:38 | <Lachy> | well, sort of. Ubuntu running in a virtual machine is accessing the web server running under Mac on the same computer |
| 09:38 | <foolip> | Lachy: file a bug :) (not in #whatwg) |
| 09:39 | <hsivonen> | foolip: http://pastebin.mozilla.org/726083 |
| 09:39 | <Lachy> | will do soon |
| 09:39 | <foolip> | hsivonen: yeah, that's the same issue |
| 09:39 | <hsivonen> | foolip: thanks |
| 09:39 | <foolip> | sorry about that, linux libraries were rushed |
| 09:41 | hsivonen | looks forward to having WebM in mozilla-central and therefore in my local builds |
| 09:42 | <doublec> | hsivonen, the patches are in the bugs now. I'm just touching my stuff up for review |
| 09:44 | <Lachy> | hmm, maybe the audio issue is caused by VMWare trying to take a snapshot of the virtual machine, and using a lot of resources. I usually have to wait for that to finish before doing things |
| 09:48 | <hsivonen> | doublec: great! |
| 09:53 | <annevk> | "I didn't patent CSS" heh |
| 09:57 | <hsivonen> | I think it's great howcome made that point on stage. |
| 10:09 | <Lachy> | annevk, Håkon may not have, but Microsoft did :-) |
| 10:10 | <Lachy> | according to the CSS patent disclosures |
| 10:11 | <MikeSmith> | about http://html5.org/tools/web-apps-tracker?from=4926&to=4927 .. is that constraint meant to be machine checkable? |
| 10:11 | <annevk> | i think it is |
| 10:12 | <othermaciej> | how can you detect "include data blocks"? |
| 10:12 | <MikeSmith> | how do I determine if an instance of the element contains data blocks? |
| 10:12 | <othermaciej> | I can imagine it for certain known mime types |
| 10:12 | <othermaciej> | but in general, the validator can't know what the browser would treat as script |
| 10:13 | <MikeSmith> | (what othermaciej said ... I type too clow) |
| 10:13 | <othermaciej> | not for sure anywa |
| 10:13 | <MikeSmith> | *slow |
| 10:13 | <othermaciej> | it can identify JavaScript and known non-script types and that's it |
| 10:15 | <MikeSmith> | what's the current state of discussion around the Content Security Policy spec? |
| 10:15 | <MikeSmith> | http://people.mozilla.org/~bsterne/content-security-policy/ |
| 10:16 | <MikeSmith> | have other browser vendors expressed some support for implementing it yet? |
| 10:16 | <othermaciej> | I think the WebKit project may be interested in implementing it |
| 10:16 | <othermaciej> | but |
| 10:17 | <othermaciej> | I dunno how good the current rev is |
| 10:17 | <zcorpan_> | MikeSmith: i think html5 should specify a dedicated mime type for 'data blocks' |
| 10:18 | <MikeSmith> | zcorpan_: aye |
| 10:18 | <MikeSmith> | othermaciej: OK, thanks |
| 10:18 | <othermaciej> | no clear statement from vendors |
| 10:18 | <othermaciej> | I prefer security features that work more automatically w/ less work by the content author |
| 10:18 | <othermaciej> | but clearly the idea of dropping priveleges has a certain appeal right now |
| 10:18 | <othermaciej> | good night |
| 10:20 | <MikeSmith> | 'night |
| 10:23 | <hsivonen> | are the right people at Opera aware of the content type trouble with http://devfiles.myopera.com/articles/1891/sunflower-webm.html ? |
| 10:24 | <roc> | Lachy: yesterday you said something about the performance of WebM decoding with the CPU on 1080p |
| 10:24 | <hsivonen> | (the video file is application/octet-stream; should be video/webm) |
| 10:25 | <hasather> | hsivonen: I can mail the guy |
| 10:25 | <hsivonen> | hasather: thanks |
| 10:26 | <annevk> | according to foolip application/octet-stream should play per the specification |
| 10:27 | <roc> | Lachy: Using GL to perform YUV conversion and scaling, we can play fullscreen 1080p using slightly less than one core of my 3-year-old Macbook pro , so it's definitely doable |
| 10:27 | <annevk> | specification is somewhat unclear on media types; e.g. if a UA supports Ogg Theora and WebM, someone puts WebM as video/ogg on the web, it should still play per spec |
| 10:27 | <roc> | which specification? |
| 10:27 | <annevk> | HTML5 |
| 10:28 | <annevk> | "The MIME type "application/octet-stream" with no parameters is never a type that the user agent knows it cannot render." |
| 10:29 | <roc> | ah |
| 10:29 | <roc> | interesting, I wasn't aware of that |
| 10:29 | <annevk> | and fetching resources says |
| 10:29 | <annevk> | "If the media resource is found to have Content-Type metadata that, when parsed as a MIME type (including any codecs described by the codecs parameter), represents a type that the user agent knows it cannot render (even if the actual media data is in a supported format)" |
| 10:29 | <annevk> | it doesn't actually say anything about mismatching |
| 10:30 | <hsivonen> | annevk: looks like the spec is settuing us up for image/* all over again |
| 10:30 | <annevk> | not sure whether to consider this broken or a feature |
| 10:30 | <hsivonen> | right. magic numbers within image/* work |
| 10:30 | <hsivonen> | except for SVG maybe |
| 10:30 | <annevk> | yeah, SVG is noted as exception |
| 10:30 | <zcorpan_> | i noticed this about a year ago or so |
| 10:30 | <annevk> | for images |
| 10:31 | <zcorpan_> | don't remember if i emailed the list, but at least Hixie didn't think it was an obvious bug as to change the spec |
| 10:31 | <annevk> | zcorpan_, was there discussion about it too? I though there was |
| 10:31 | <zcorpan_> | at least in #whatwg, yeah |
| 10:32 | <roc> | hum |
| 10:32 | <roc> | when did this get added? |
| 10:32 | <roc> | adding a requirement for binary sniffing of all media types is a fairly big deal |
| 10:34 | <foolip> | that is what we're doing |
| 10:34 | <annevk> | I think we should remove that |
| 10:34 | <annevk> | I guess Apple advocated it |
| 10:34 | <MikeSmith> | Lachy: about http://www.w3.org/Bugs/Public/show_bug.cgi?id=9707 .. the bug-status guidelines that othermaciej wrote up pretty much reserve the "invalid" state for bug reports that are simply spam. I think we are meant to use "worksforme" instead for cases like this one |
| 10:35 | <annevk> | but all this is based on vague recollections so don't take my word for it :) |
| 10:35 | <foolip> | I wouldn't mind changing the spec, but would make actually changeing the implemention veeery low priority |
| 10:35 | <hsivonen> | Why does http://videojs.com/ have a *fallback* for Opera? |
| 10:35 | <foolip> | zcorpan_: can you check it out? |
| 10:36 | <zcorpan_> | roc: at first there were no checking at all in the spec |
| 10:36 | <annevk> | it was added between april and august 2009 |
| 10:36 | <zcorpan_> | roc: then some checks were added, mostly due to feedback from me iirc |
| 10:37 | annevk | goes to find the exact diff |
| 10:37 | <zcorpan_> | oh maybe you're discussing something else |
| 10:37 | <annevk> | foolip, wouldn't it be a trivial change? |
| 10:37 | <foolip> | annevk: not at all |
| 10:38 | <foolip> | once we hand off the media to the platform, it doesn't even have the MIME type, it uses only sniffing to set up the decoding pipeline |
| 10:38 | <annevk> | http://www.w3.org/Bugs/Public/show_bug.cgi?id=7977 |
| 10:39 | <roc> | thanks anne |
| 10:39 | <annevk> | foolip, ah yeah, but the application/octet-stream one should be more doable |
| 10:40 | <foolip> | annevk: to just not support it? |
| 10:40 | foolip | hsan't looked at the bug |
| 10:40 | <foolip> | oh, simon's old bug |
| 10:41 | <foolip> | we already handle application/octet-stream as per spec |
| 10:41 | <roc> | that bug report is not very helpful |
| 10:42 | <annevk> | foolip, right, to not support it |
| 10:42 | <annevk> | roc, not really, no :/ |
| 10:42 | <zcorpan_> | annevk: why would we want to not support application/octet-stream? |
| 10:43 | <foolip> | I'm beginning to think that MIME types are just silly |
| 10:43 | <foolip> | well, it's nice that things aren't served as text/plain I guess |
| 10:43 | <jgraham> | That seems to be a surprisingly common conclusion |
| 10:44 | <annevk> | zcorpan_, I could be convinced that we want to support lack of a Content-Type header everywhere, but then we should be consistent about it |
| 10:44 | <foolip> | bah, I'm not writing a manifesto for/against MIME today, life goes on |
| 10:44 | <annevk> | zcorpan_, e.g. <html manifest> does not support lack of a content type, neither does EventSource() iirc |
| 10:45 | <zcorpan_> | <video> is more like <img> though |
| 10:46 | zcorpan_ | doesn't really care much either way |
| 10:47 | <annevk> | we might gain some more friends by getting rid of Content-Type :p |
| 10:50 | <roc> | hmmm |
| 10:50 | <hasather> | hsivonen: the video should work now |
| 10:50 | <zcorpan_> | annevk: i think there are more effective ways to gain friends :P |
| 10:50 | <roc> | I wonder how we handle unknown content-types for video right now |
| 10:51 | <zcorpan_> | what was the issue with http://videojs.com/ ? |
| 10:51 | <roc> | ah, I think we just fail |
| 10:52 | <hsivonen> | hasather: thanks |
| 10:52 | <zcorpan_> | i get a <video> there in opera |
| 10:53 | <hsivonen> | zcorpan_: opera isn't on the support list but on the fallback list |
| 10:53 | <hsivonen> | zcorpan_: problem with advertising, not code, then |
| 11:38 | <annevk> | ooh I found the wrong bug? |
| 11:38 | <annevk> | sorry |
| 11:39 | annevk | looks once more |
| 11:42 | <annevk> | the actual revision was http://html5.org/tools/web-apps-tracker?from=3497&to=3498 |
| 11:42 | <annevk> | it is not tied back to any bug report or email unfortunately |
| 11:42 | <annevk> | though you could look in the archives of mailing lists to see if hixie confirmed such a change around 2009-07-30 |
| 11:47 | annevk | cannot find something in the whatwg archives so fast |
| 11:58 | <Philip`> | annevk: Do you mean http://article.gmane.org/gmane.ietf.http-wg/4340 ? |
| 11:59 | <Philip`> | (or the public-html equivalent) |
| 11:59 | <Philip`> | (which Google is less happy to find for me) |
| 12:00 | <annevk> | guess so |
| 13:12 | <annevk> | TabAtkins, awake by any chance? |
| 13:12 | <annevk> | TabAtkins, I was wondering if the split between display values is really intuitive enough for authors |
| 13:12 | <annevk> | TabAtkins, I think we should at least provide some simple shorthands for flex usage |
| 14:22 | <zcorpan_> | ah, so it was mozilla guys asking for application/octet-stream |
| 14:25 | <annevk> | and now they wonder why lol |
| 14:25 | <annevk> | typical that only we ended up implementing it :p |
| 14:26 | <zcorpan_> | i think the spec is fine as is, fwiw |
| 14:30 | <annevk> | yeah, paving the path for getting rid of content negotiation and what not :p |
| 14:31 | <annevk> | on the other hand, it makes me wonder why we'd support media types on the HTTP level at all... |
| 14:31 | <annevk> | for video |
| 14:32 | <zcorpan_> | to make videos playable in top-level browsing context without sniffing everything for video |
| 14:34 | <annevk> | so in a top-level browsing context we'd not play video labeled as application/octet-stream? |
| 14:35 | <zcorpan_> | not sure |
| 14:35 | <zcorpan_> | but we wouldn't for text/plain or text/html |
| 15:47 | <Dashiva> | "Awesome for legacy browsers like IE9 and Safari." :) |
| 16:04 | <paul_irish> | (sorry about the flood flood. eek) |
| 16:22 | <zcorpan_> | Lachy: Hixie said websrt would be used for chapters also |
| 16:23 | <Lachy> | really? |
| 16:23 | <Lachy> | but it's not a chapter format |
| 16:24 | <zcorpan_> | http://krijnhoetmer.nl/irc-logs/whatwg/20100504#l-77 |
| 16:39 | <TabAtkins> | annevk: I'm up now. |
| 16:39 | <TabAtkins> | What were you thinking? |
| 18:51 | <zcorpan_> | what are the requirements for media chapters? |
| 18:52 | <zcorpan_> | and what are the solutions being used today for media chapters? |
| 18:54 | <gour> | i'm reading diveintohtml5 and wonder if it is safe to use 'html' doctype and some similar html5 enhancements for a web site (which will be generated by static site generator)? |
| 18:55 | <zcorpan_> | depends on what you mean by 'safe' |
| 18:55 | <TabAtkins> | The html doctype is completely safe; it's been usable for years. |
| 18:55 | <gour> | that's it will be useful for those users using old IE crap |
| 18:55 | <TabAtkins> | Many of the simpler new elements can be used, but you have to use the html5 shim in IE, and apply proper styling to them yourself, as they're not officially recognized yet. |
| 18:56 | <paul_irish> | gour: yup. it'll kick it into standards mode, which means your css and js will act the exact same as if you had the xhtml1 or html4 strict doctype up there |
| 18:56 | <gour> | i plan to use YAML framework which takes care of (almost all) IE 'idioms' |
| 18:57 | <TabAtkins> | Alternately, you may look at this page for valid html doctypes that will trigger standards mode in all browsers: http://www.xanthir.com/etc/doctype.php |
| 18:57 | <gour> | let me see... |
| 18:57 | <zcorpan_> | TabAtkins: they're not valid |
| 18:58 | <TabAtkins> | zcorpan_: Really? I discussed it here in the room, and thought it was valid a while back. |
| 18:58 | <TabAtkins> | Did something change on me recently? |
| 18:58 | <zcorpan_> | they trigger standards mode |
| 18:58 | <zcorpan_> | they're not valid |
| 18:58 | <TabAtkins> | Ah, right. Yes. |
| 18:58 | <TabAtkins> | That's fine. ^_^ |
| 18:59 | <TabAtkins> | But I will amend my statment in the future. I didn't mean that sense of "valid". |
| 19:00 | <gour> | TabAtkins: http://code.google.com/p/html5shiv/ is the place of html5 shiv? |
| 19:01 | <TabAtkins> | Yes. |
| 19:01 | <gour> | ta |
| 19:13 | <paul_irish> | <!dOcTYpE HTMl > is valid though. :) |
| 19:13 | <paul_irish> | html5 doctype, myspace style. |
| 19:16 | <gour> | :-) |
| 19:23 | <TabAtkins> | msg paul_irish You work for g now, right? |
| 19:23 | <TabAtkins> | Shit. |
| 19:23 | <TabAtkins> | Eh. |
| 19:26 | <paul_irish> | :) I don't start for another 2 weeks or so. |
| 20:54 | jgraham | finishes /The Thousand Autumns of Jacob de Zoet/, wonders what he was planning to do |
| 20:54 | <jgraham> | It's a fantastic book by the way |
| 20:56 | <jgraham> | Although possibly I feel too sad to do anything useful |
| 20:56 | <jgraham> | Finishing good books sucks |
| 21:01 | <jgraham> | (it is also possible that listening to the Eels isn't the most chhering way to recover) |
| 21:01 | <jgraham> | *cheering |
| 21:47 | <jgraham> | Leaving the xkcd book lying around is dangerous. Now my life is soundtracked by intermittent bursts of laughter mingled with the occasional "that is /so/ like you" |
| 21:50 | <jgraham> | Although I guess the fact that I am writing a monolouge about it on irc does provide some truth to the claims |
| 21:58 | <TabAtkins> | ^_^ |
| 23:58 | <TabAtkins> | othermaciej: Both 89 and 92 have proposals on them now. |