| 00:00 | <Hixie> | can we just have a data-foo thingy that we put on a <dfn> that says "this term is actually defined over there"? |
| 00:00 | <Hixie> | i guess that's a bit more maintenance |
| 00:00 | <Hixie> | i dunno |
| 00:00 | <Hixie> | anyway |
| 00:00 | <AryehGregor> | Something like @media print { [data-anolis-spec]::after { content: attr(data-anolis-spec); text-transform: uppercase } } or whatever. |
| 00:00 | <Hixie> | i'm happy to use whatever if it makes things better |
| 00:00 | <AryehGregor> | It's not tenable to do that given the number of terms we have to reference. |
| 00:00 | <AryehGregor> | DOM Range uses zillions of terms from DOM Core and HTML. |
| 00:00 | <Hixie> | we really should reduce the number of specs, man |
| 00:01 | <Hixie> | DOM Range should just be in DOM Core |
| 00:01 | <AryehGregor> | HTML isn't so affected because it mostly just defines stuff itself. |
| 00:01 | <annevk> | not if DOM Range depends on HTML |
| 00:01 | <AryehGregor> | I don't think we need to merge specs if we have good xspec xrefs. |
| 00:01 | <Hixie> | merging specs is for sanity of not having a bazillion specs, not for the xrefs |
| 00:01 | <Hixie> | we should have like a dozen or so core specs for the web platform |
| 00:02 | <AryehGregor> | I think only the Selection part of DOM Range depends on HTML. |
| 00:02 | <annevk> | yeah, but we first need to figure out the Web platform |
| 00:02 | <annevk> | then we can organize :) |
| 00:03 | <annevk> | I mean, we don't even agree on mutation listeners |
| 00:03 | <annevk> | :p |
| 00:04 | <annevk> | oh, IE10 is removing conditional comments |
| 00:04 | <annevk> | about time |
| 00:04 | <AryehGregor> | Yay. |
| 00:04 | <TabAtkins> | Hixie: Where'd you get the VALUE=TEXT:ERROR thing from? |
| 00:04 | TabAtkins | just read the vcard spec and, as expected, this situation is not addressed. |
| 00:04 | <annevk> | ms2ger, ^^ all of the above :) |
| 00:05 | <Hixie> | core (dom core, dom events, dom range), i/o (xhr, cors, fetch, from-origin, sniffing, file api), user interaction (mouse events, geo, touch, wheel, device orientation), semantics/api (html, web workers, clipboard, aria), css, svg, mathml, js, unicode, http, woff, webgl, indexeddb... |
| 00:05 | <Hixie> | TabAtkins: just made it up |
| 00:06 | <Hixie> | TabAtkins: yeah vCard is a spec from the old times |
| 00:06 | <TabAtkins> | Hixie: Ah, kk. |
| 00:06 | <jamesr> | Hixie: audio |
| 00:06 | <Hixie> | jamesr: part of html |
| 00:06 | <TabAtkins> | Hixie: I could tell from the fact that it's a Word doc. |
| 00:06 | <jamesr> | webaudio isn't (and imo shouldn't be, it's a leaf node) |
| 00:06 | <Hixie> | jamesr: along with webrtc, video, etc |
| 00:06 | <jamesr> | the <audio> tag is html and insufficient |
| 00:06 | <Hixie> | jamesr: i'm saying webaudio should be in the same spec as web workers |
| 00:07 | <Hixie> | jamesr: "semantics/api" |
| 00:07 | <roc> | webaudio should not be in a silo of its own |
| 00:08 | <Hixie> | (note that what i'm saying here doesn't map to who should be working on what) |
| 00:08 | <roc> | I guess jamesr hasn't seen my rant about this, because the Audio WG is off in a silo of its own! |
| 00:08 | <Hixie> | heh |
| 00:08 | <jamesr> | yeah, why is that? |
| 00:08 | <Hixie> | why is what? |
| 00:08 | <roc> | because that's the way the W3C works |
| 00:08 | <jamesr> | hm |
| 00:09 | <jamesr> | do we have another process set up other than 'email everything to the whatwg list'? |
| 00:09 | <TabAtkins> | Argh, I forget how to get the coordinates of a click relative to the clicked element. ;_; |
| 00:09 | <jamesr> | the web perf WG is another problem, but i don't intend to do anything important enough there for it to matter |
| 00:10 | <Hixie> | jamesr: i'm talking about how we publish specs, not how we write them |
| 00:10 | <annevk> | TabAtkins, offsetX/Y |
| 00:11 | <roc> | unfortunately how we write specs strongly influences how they get published, and how they are designed too |
| 00:12 | <Hixie> | yeah |
| 00:12 | <annevk> | not so much guidelines for writing specs :( |
| 00:12 | <Hixie> | we're in much need of overhaul of much more important stuff before we starting worrying about how we split the specs |
| 00:12 | <Hixie> | i'm just blue-skying here |
| 00:13 | <annevk> | also need some kind of manual to tell people how to write specs |
| 00:13 | <roc> | for example, I think one reason the Audio WG doesn't like my proposal for general media stream processing, is that they're the Audio WG and they're not chartered for or interested in general media stream processing |
| 00:13 | <jamesr> | annevk: does anything anywhere define what the DOMTimestamp on events represent, precisely? |
| 00:13 | <TabAtkins> | Yay for Conway's Law! |
| 00:13 | <Hixie> | annevk: we have a wiki page on whatwg.org, i've helped a bit with it. i'd love to help more but don't want to do it all myself. :-) |
| 00:14 | <annevk> | jamesr, http://dvcs.w3.org/hg/domcore/raw-file/tip/Overview.html#dom-event-timestamp |
| 00:14 | <annevk> | Hixie, good point, maybe once I'm back from Berlin I'll remember that and fix it up a bit |
| 00:14 | <jamesr> | ok, so same as ECMAScript Date.now() |
| 00:14 | <annevk> | jamesr, the definition of DOMTimeStamp is in Web IDL btw |
| 00:14 | <Hixie> | roc: implementations win |
| 00:14 | <jamesr> | yeah, i'm interested in the value not the representation |
| 00:14 | <jamesr> | thanks |
| 00:15 | <Hixie> | roc: are there implementations of either of the proposals? |
| 00:15 | <roc> | jamesr: the relevant thread is here, if you care: http://lists.w3.org/Archives/Public/public-audio/2011AprJun/0102.html |
| 00:16 | <roc> | Hixie: yes, Chris Rogers has his in Chrome and I'm working on one |
| 00:16 | <jamesr> | oh doug |
| 00:16 | <roc> | doug was mostly an innocent bystander in this case |
| 00:16 | <jamesr> | with a bad suggestion |
| 00:17 | <roc> | Hixie: yes, implementations win, but that's not always a god thing |
| 00:17 | <roc> | er, a good thing |
| 00:17 | <Hixie> | i wasn't making a value judgement |
| 00:18 | <Hixie> | merely observing reality :-) |
| 00:18 | <roc> | I thought so |
| 00:18 | <Hixie> | looks like i said so in http://lists.w3.org/Archives/Public/public-audio/2011AprJun/0118.html too :-) |
| 00:19 | <Hixie> | personally i know nothing about audio so i'm really not in a position to make a proposal |
| 00:19 | <roc> | for other people, saying the exact same thing would amount to a value judgment |
| 00:19 | <jamesr> | as per your last point audio APIs are one of the main drivers of flash in a lot of cases |
| 00:19 | <Hixie> | (that was the case with video conferencing too; after nobody did anything for like a year i finally broke down and taught myself the relevant specs and wrote up the PeerConnection proposal. i don't intend to do that with audio.) |
| 00:20 | <Hixie> | roc: true |
| 00:20 | <jamesr> | angry birds would be 100% web platform if there was decent audio support on the web |
| 00:20 | <roc> | actually |
| 00:20 | <roc> | it turns out that most of today's Web games don't need new Web API |
| 00:21 | <annevk> | Hixie, btw, is PeerConnection mostly for client-server still? |
| 00:21 | <annevk> | Hixie, it doesn't really read like it could be client-client as well |
| 00:21 | <Hixie> | annevk: PeerConnection is for peer-to-peer |
| 00:21 | <Hixie> | hence the name |
| 00:21 | <Hixie> | it does NAT traversal |
| 00:21 | <annevk> | Hixie, but the initiation talks about servers |
| 00:21 | <annevk> | Hixie, I missed that |
| 00:22 | <roc> | they just want to play preloaded sounds with low latency, and would be served just fine with a bunch of <audio preload>s and getElementById(id).clone().play() if that was implemented well |
| 00:22 | <Hixie> | annevk: well, yeah, you can't connect two peers without them having some sort of signaling channel to coordinate things |
| 00:22 | <roc> | see the comments from the Zynga guys later in the thread |
| 00:22 | <Hixie> | annevk: and that means a server to coordinate things |
| 00:22 | <roc> | same goes for Angry Birds, AFAIK |
| 00:22 | <TabAtkins> | Low-latency audio with <audio> actually isn't *that* hard to hack around. |
| 00:22 | <TabAtkins> | http://xanthir.com/demos/collidingcircles demonstrates it, frex. |
| 00:23 | <annevk> | Hixie, I see, because the previous idea was that they would connect to each other based on some keys |
| 00:23 | <roc> | There are ... implementation issues: http://www.phoboslab.org/log/2011/03/the-state-of-html5-audio |
| 00:23 | <annevk> | Hixie, and the way the keys were exchanged was up to the site |
| 00:23 | <jamesr> | TabAtkins: is that just keeping a bag of <audio>s around? |
| 00:23 | <TabAtkins> | roc: Yes, thus the "hack around" part. |
| 00:23 | <Hixie> | annevk: it's basically still like that |
| 00:23 | <Hixie> | annevk: you get a callback that gives you the "keys" (actually an SDP payload) |
| 00:23 | <TabAtkins> | jamesr: Yeah, I generate 10 <audio>s per sound, and cycle through them everytime I play that particular sound. |
| 00:23 | <Hixie> | annevk: and you get it to the other side somehow |
| 00:24 | <jamesr> | TabAtkins: doesn't really scale when you are trying to implement quake and have hundreds of sounds that you might want to play and you might need to overlap them dozens of times on top of each other |
| 00:24 | <Hixie> | annevk: and then you pass the "keys" to the other browser |
| 00:24 | <Hixie> | annevk: and so on back and forth |
| 00:24 | <jamesr> | and you need to distance-fade them and adjust some samples to come in one ear more strongly than the other |
| 00:24 | <Hixie> | annevk: with all the video and audio and udp data going straight peer-to-peer |
| 00:24 | <annevk> | I guess I should read it more closely, thanks |
| 00:24 | <jamesr> | or pan the sample from one ear to the other as it plays (and the rocket wooshes by) |
| 00:24 | <Hixie> | annevk: i haven't written a good intro |
| 00:24 | <Hixie> | annevk: it's on my list |
| 00:24 | <jamesr> | which you definitely don't have CPU time to do in javascript |
| 00:25 | <TabAtkins> | jamesr: An actual games company came up with another technique where you concatenate all your sounds and seek. That way you only need to have as many audios as you plan to have *total* sounds playing at once. |
| 00:25 | <TabAtkins> | That requires a bit more expertise, though. |
| 00:25 | <jamesr> | and a really really big continuous audio sample |
| 00:25 | <jamesr> | or rather a few dozen copies of it |
| 00:25 | <roc> | what was my last message? |
| 00:25 | <TabAtkins> | Yes. |
| 00:25 | <jamesr> | roc: There are ... implementation issues: http://www.phoboslab.org/log/2011/03/the-state-of-html5-audio |
| 00:25 | <zewt> | jamesr: can't really do it in javascript anyway if you want to support things like hardware surround |
| 00:26 | <jamesr> | true |
| 00:26 | <zewt> | environmental audio, etc |
| 00:26 | <roc> | 3D spatialization definitely needs new API |
| 00:26 | <roc> | Angry Birds and Zynga games do not |
| 00:26 | <zewt> | roc: well, it's debatable whether 3d audio vs. 2d audio should be separate APIs or not |
| 00:27 | <roc> | they should be tightly integrated |
| 00:27 | <jamesr> | they could use 2d spatialization, it'd be basically free if the API was designed well |
| 00:27 | <roc> | but we already have an API for playing sound samples |
| 00:27 | <roc> | <audio> |
| 00:27 | <zewt> | angry birds could certainly make use of stereo (whether it does or not I don't know) |
| 00:27 | <TabAtkins> | zewt: They either use or fake stereo on Android, at least - as you scroll away from pigs, their grunts get quieter. |
| 00:27 | <Philip`> | Just wrap OpenAL and call it WebAL |
| 00:28 | <jamesr> | Philip`: i think khronos proposed that :/ |
| 00:28 | Philip` | 's experience with OpenAL has not been universally positive |
| 00:28 | <zewt> | i know environmental audio APIs can be pretty stupidly complex; trying to represent their whole vocabulary in a generic API is probably hopeless |
| 00:28 | <roc> | the Zynga guys didn't ask for spatialization |
| 00:28 | <roc> | they asked for basic sound playback to work |
| 00:28 | <jamesr> | the quake guys did |
| 00:28 | <roc> | and it doesn't |
| 00:29 | <jamesr> | yes, there are QoI issues with existing <audio>, but there are also API needs |
| 00:29 | <Philip`> | (e.g. I work on some game that uses OpenAL, and it randomly hits assertion failures on OS X and nobody has yet figured out why) |
| 00:29 | <roc> | there's both, but the QoI issues get less love, because hey, shiny new API! |
| 00:29 | <zewt> | roc: heh |
| 00:30 | <Philip`> | (and also it breaks with Creative's OpenAL drivers on Windows so we replace it with the OpenAL Soft implementation which defeats the whole point of standards) |
| 00:30 | <zewt> | roc: but if they're built on the same underlying framework, in principle both win |
| 00:30 | <roc> | indeed |
| 00:30 | <zewt> | i seem to remember looking at OpenAL source over a decade ago and it having pages of conditionals 10+ levels deep |
| 00:30 | <zewt> | tried to find the code I saw later on but it seemed to have been scoured from the internet |
| 00:30 | <roc> | but, and here I reveal my biases, Chris Rogers' API and implementation are much not built on the same underlying framework |
| 00:31 | <jamesr> | what do you mean? they use the same backend as our <audio> support |
| 00:32 | <Philip`> | zewt: If I remember correctly, the official reference driver was terrible, and OpenAL Soft is a reimplementation from scratch |
| 00:32 | <jamesr> | the whole stack is the same except for the API bits, and the frontend support for the webaudio features not there in <audio> |
| 00:32 | <zewt> | it was easily some of the worst code I'd ever seen |
| 00:32 | <Philip`> | s/driver/implementation/ |
| 00:32 | <roc> | I mean you play samples using AudioBufferNode, not <audio> |
| 00:32 | <zewt> | then or since |
| 00:32 | <jamesr> | oh, you mean from the API pov? <audio> can be used as source nodes |
| 00:33 | <roc> | that's not very well defined |
| 00:33 | <roc> | it's not the way you're supposed to do it |
| 00:33 | <jamesr> | how so? |
| 00:33 | <jamesr> | http://chromium.googlecode.com/svn/trunk/samples/audio/specification/specification.html#MediaElementAudioSourceNode-section |
| 00:34 | <jamesr> | use one of those as a source node linked to your <audio> or <video or whatever, hook it up to an output, note() |
| 00:34 | <roc> | for example, if the audio element is paused, what happens? |
| 00:34 | <Hixie> | things can definitely be well-defined |
| 00:35 | <Hixie> | last i looked at the audio api, the whole thing was rather under-defined |
| 00:35 | <Hixie> | that's just how things are before they're mature |
| 00:35 | <Hixie> | the question is, is it the right direction at all? |
| 00:35 | <Hixie> | if there's more than one way to do things, it probably isn't (e.g. is there another way of getting audio from a file into an audio node?) |
| 00:36 | <Hixie> | if it requires a lot of code to do simple things, it probably isn't (e.g. do you have to have six lines of boilerplate code to get from <audio> to an audionode?) |
| 00:37 | <Hixie> | if there are things that don't work just because parts of the api have different designs, it probably isn't (e.g. can you go from <audio> to an audio node but not from an audio node to a PeerConnection stream?) |
| 00:37 | <Hixie> | (i don't know the answers to those questions) |
| 00:37 | <Hixie> | and there's also the question "is there another concrete proposal?" |
| 00:37 | <TabAtkins> | Hixie: http://chromium.googlecode.com/svn/trunk/samples/audio/specification/specification.html#AudioElementIntegration-section |
| 00:38 | <Hixie> | TabAtkins: i know, i've talked about this with chris in front of you :-P |
| 00:38 | <TabAtkins> | Ah, I must not have been listening. ^_^ |
| 00:38 | <Hixie> | :-) |
| 00:41 | <jamesr> | strawpoll on another topic: when calling setTimeout(..., 3000), if the system clock is adjusted before the timer fires all browsers except WebKit fire the timer anyway after 3000ms expire. WebKit-based browsers wait until the system clock has advanced to 3000ms past the time the system clock said when setTimeout() was called |
| 00:41 | <jamesr> | is WebKit's behavior here justifiable or just crazy land? |
| 00:41 | <jamesr> | i tested against ffx 8.01a, opera 11.something on mac and ie9/win7 |
| 00:42 | <TabAtkins> | jamesr: So you mean that other browsers use a real time, while webkit goes purely off of Date()? |
| 00:42 | <zewt> | sounds like it'd break everything using timers, especially if the time is set back a lot |
| 00:42 | <zewt> | (which fortunately doesn't happen often, but) |
| 00:42 | <jamesr> | zewt: the delta of Date.now() before the callback and when it fires goes negative in other browsers |
| 00:42 | <jamesr> | actually it happens more than most people think |
| 00:42 | <jamesr> | TabAtkins: yes, for scheduling purposes |
| 00:43 | <TabAtkins> | Webkit's behavior is crazytown. |
| 00:46 | <jamesr> | k i'll just fix it |
| 00:46 | <zewt> | setTimeout should be defined in terms of the time delta API, once it actually exists |
| 00:46 | <jamesr> | yeah, i'm working on a proposal for that too |
| 00:49 | <jamesr> | annevk-cloud: http://dvcs.w3.org/hg/domcore/raw-file/tip/Overview.html#dom-event-timestamp doesn't say anything about leap seconds |
| 00:49 | <jamesr> | annevk-cloud: ECMA-262 defines Date as ignoring leap seconds, although i don't think anyone actually does that |
| 00:50 | <zewt> | (nitpick: "number of milliseconds" is more natural than "amount of") |
| 00:50 | <jamesr> | i think that means the two differ by either 24 or 34 seconds now (i'm not sure what the wikipedia page is trying to tell me) |
| 01:05 | <TabAtkins> | Dammit, these emails I write don't seem so long when I'm writing them. |
| 01:06 | <jamesr> | heycam: webperf uses www.w3.org/StyleSheets/TR/W3C-ED.css |
| 01:06 | <jamesr> | but it's not https so if you load the webperf spec over https you get mixed content errors |
| 01:06 | <jamesr> | is that stylesheet available over https? |
| 01:07 | <heycam> | jamesr, it seems it is |
| 01:07 | <jamesr> | if i just change the url to https:// the connection times out for me |
| 01:07 | <jamesr> | what url are you using? |
| 01:07 | <heycam> | https://www.w3.org/StyleSheets/TR/W3C-ED.css |
| 01:07 | <nessy> | so, hmm, I wonder if I've been doing this correctly: if I want to point out a bug on a part of the spec that's only in the WHATWG version, can I still go through the w3c bug tracker or should I email the whatwg list? |
| 01:08 | <heycam> | maybe the url should be "//www.w3.org/StyleSheets/TR/W3C-ED.css"; so that it uses http: or https: depends on how you view the spec. (that works, doesn't it?) |
| 01:08 | <jamesr> | yeah |
| 01:08 | <jamesr> | (well it should) |
| 01:09 | <jamesr> | either that or just always use the https:// version |
| 01:09 | <jamesr> | i think // is best practice |
| 01:09 | <jamesr> | hm url works for me now, not sure why i was having trouble earlier |
| 01:11 | <TabAtkins> | nessy: There's no practical difference between filing a bug and sending an email, so just send an email (that way it avoids anyone being snotty and closing the bug). |
| 01:11 | <nessy> | TabAtkins: hehe :-) good idea! thanks! |
| 01:15 | <TabAtkins> | nessy: (I'm also prejudiced in that I hate the bugtracker and vastly prefer email.) |
| 01:16 | <nessy> | TabAtkins: aha! I don't really care actually - but conversations in the bug tracker tend to be short and to the point, but also often harsh |
| 01:18 | <heycam> | hmm, where do you actually download the IE10 PP3 from? http://ie.microsoft.com/testdrive/Info/Downloads/Default.html only seems to have PP2 |
| 01:18 | <TabAtkins> | Has pp3 beenr elease yet? |
| 01:19 | <heycam> | http://windowsteamblog.com/ie/b/ie/archive/2010/06/23/internet-explorer-platform-preview-3-released-today.aspx |
| 01:30 | <TabAtkins> | heycam: Hm, I dunno. |
| 01:44 | <jamesr> | TabAtkins: don't you see? corporations are -have already- more likely |
| 01:44 | <TabAtkins> | jamesr: It's... it's all so obvious now. |
| 01:44 | <jamesr> | you have no time to rebut make your accessibility |
| 01:46 | <TabAtkins> | Take off all SVG. For great justice. |
| 02:03 | <TabAtkins> | 40% of the new email in my inbox from the last 10 minutes is a11y. |
| 02:04 | <TabAtkins> | Ah, to be fair, one of those was the public-a11y reminder that I'm not registered and the list hates me. |
| 02:14 | <Philip`> | Does 10 minutes provide a statistically significant sample? |
| 05:23 | <boblet> | damn, writing specs is hard. I’m trying to think of a good alternative *sentence* to propose… and failing <_< |
| 05:31 | <zewt> | heh, g+ feedback button sends a screenshot of the page ... i wonder what they're doing to implement that |
| 06:53 | <boblet> | Anyone following along with my problems with the blockquote spec, I wrote an article http://oli.jp/2011/blockquote/ found real-world examples that would be non-conforming with current spec http://oli.jp/example/blockquote-metadata/ and emailed WHATWG list |
| 06:53 | <boblet> | so do I file a bug report now or is that overkill? :) |
| 07:44 | <annevk> | boblet, email should be sufficient |
| 07:44 | <annevk> | boblet, unless you want to escalate it within the W3C if you disagree with Ian, in which case you need to file a bug report |
| 07:48 | <annevk> | jamesr_, I mentioned your comment on leap seconds in the source |
| 07:48 | <boblet> | annevk: thanks. yeah heard email is enough from Hixie too |
| 07:48 | <annevk> | jamesr_, I guess we want to be consistent with new Date() |
| 07:53 | <MikeSmith> | it's really great to see the latest IE10 platform preview has an HTML5 that intends to conform with the spec |
| 07:54 | <MikeSmith> | a lot of nice surprises in that latest release |
| 07:55 | <MikeSmith> | is http://gsnedders.html5.org/html5lib-tests/runner.html up to date? |
| 07:56 | <MikeSmith> | and/or it would be great to get some data on how much of the html5lib test suite the IE10 parser is passing at this point |
| 08:00 | <MikeSmith> | annevk: dunno if you saw http://enable-cors.org/ yet |
| 08:01 | <Rik`> | yeah, happy morning without conditional comments \o/ |
| 08:01 | <annevk> | Yeah some time ago |
| 08:01 | <MikeSmith> | annevk: seems mhausenblas made that page |
| 08:03 | <MikeSmith> | Rik`: funny that most responses to that IEblog post were about "no more conditional comments" rather than "IE now has a real HTML5 parser" |
| 08:04 | <annevk> | We should sign up for https://plus.google.com/105923173045049725307/posts/E3mVj6nskaX |
| 08:05 | <annevk> | http://goo.gl/zq95C |
| 08:06 | <Rik`> | MikeSmith: the HTML5 parser is only interesting if you don't write valid code so web developers that care about the IE announcement are not writing invalid code hopefully |
| 08:06 | <MikeSmith> | I see |
| 08:07 | <annevk> | Rik`, conditional comments are invalid |
| 08:08 | <Onderhond> | I'll miss them though. |
| 08:08 | <nlogax> | man, i really can't figure out how to cancel a drag&drop action at the time it is dropped |
| 08:08 | <nlogax> | like what happens when you press escape mid-d&d |
| 08:15 | <MikeSmith> | maybe Google Circles should be called Google Cells instead |
| 08:16 | <MikeSmith> | in the sense of terrorist cells where there's only guy who knows who's actually in the cell, and none of the members of the cell necessarily know they are even in it |
| 08:18 | <Onderhond> | Wouldn't it still be wise to forsee some kind of standardized mechanism proving fall-back code (css) for unsupporting browsers? |
| 10:01 | <foolip> | Hixie, I'm here |
| 10:01 | <foolip> | Hixie, I was looking at this just yesterday |
| 10:01 | <foolip> | you're probably sleeping, will email you |
| 10:33 | <krijn> | (Server will be down a bit today, getting a new fiberglass connection) |
| 11:02 | <MikeSmith> | http://groups.google.com/group/es-operating-system/browse_thread/thread/5146403c9623eeec?pli=1 |
| 11:03 | <MikeSmith> | "Today I pushed a brand-new web browser source code to the ES operating system's code base. It's been built from the ground up at Esrille Inc., which I founded in Kyoto just a year ago." |
| 11:04 | <MikeSmith> | bravo Shiki |
| 11:04 | <MikeSmith> | his company has a great logo too |
| 11:04 | <MikeSmith> | http://www.esrille.com/ |
| 11:05 | <MikeSmith> | that's based on his actual dog |
| 11:05 | <MikeSmith> | named Jiminy |