19:20 < purple_haese> I could pose as moderator, but we need ralph for a theora update (or strike that part of the meeting) 19:20 < purple_haese> We have enough icecast people present. 19:20 < volsung> Okay, well, do we have enough people to talk about Icecast? 19:20 < volsung> Heh. 19:20 < volsung> I say start then. 19:20 < purple_haese> Most other points are Monty, though. 19:20 < Misirlou> Anybody logging? 19:20 < volsung> Let me check if I'm logging. 19:20 < purple_haese> I am logging. 19:20 < Misirlou> 'Kay. 19:20 < oddsock> ok, I'll start with icecast 19:21 < oddsock> icecast is due for a release within the next week or so. 19:21 < oddsock> everything is pretty much ready, we just need a bit of help with binary packages 19:21 < volsung> Which platforms? 19:21 < purple_haese> I can try to build RPMs if nobody else volunteers. 19:22 < oddsock> we have a RPM built, but need it validated 19:22 < volsung> I can test RPMs on Mandrake 9.2 when I get back to Austin. 19:22 < dev0> I can test Debian here... 19:22 -!- katabatic [~um@dsl-206-55-130-250.tstonramp.com] has quit [Read error: 54 (Connection reset by peer)] 19:22 < dev0> but I'm not competent enought o build debs 19:22 < purple_haese> I can test at least RH9 and possibly RH8. 19:22 < oddsock> we need someone who can build .debs 19:23 < volsung> calc is usually our man for that. 19:23 < oddsock> here is the RPM link : http://www.icecast.org/files/icecast-2.0.0-1.i386.rpm 19:23 < oddsock> additionally, we have a new icecast.org website in the works.. 19:23 < Misirlou> I'm guessing that icecast2 is the latest package? 19:23 < Misirlou> If so, the maintainer is Jonas Smedegaard . 19:23 < volsung> ices2 packages too? 19:24 < oddsock> karl ? will we have ices2 in a week or so ? 19:24 < karlH> ices2 is more of less ready, I'm keen on doing beta3 first though 19:24 < oddsock> ok, so ices will probably be a separate release 19:24 < Misirlou> Different guy for ices2, a Keegan Quinn . 19:24 < oddsock> I'd rather not release them together since they really are separate products 19:25 < volsung> I'd be nice to have ices2 out at the same time, but, yes, synchronized releases seem to be our downfall. :/ 19:26 < oddsock> for the new icecast website I am experimenting with a new mailinglist/forum gateway 19:26 < volsung> Ooh. 19:26 < vanguardist> nice 19:27 < oddsock> it will do 2 way communication b/w a web forum and the mailinglist 19:27 < dev0> nice 19:27 < dev0> Built from scratch? 19:27 < oddsock> nope, it uses a slightly modified phpBB2 and a mail filter 19:27 -!- katabatic_ [~um@dsl-206-55-130-250.tstonramp.com] has joined #xiphmeet 19:28 < oddsock> the new icecast website will all be in php... 19:28 < oddsock> I would have liked to wait for the website overhaul project, but..well...I didn't want to wait that long :) 19:29 < oddsock> I think that's pretty much it, icecast-wise... 19:29 < karlH> for ices2, pretty much everything is in place, docs etc 19:30 < karlH> I could do with some verification of some things like solaris live input 19:30 -!- katabatic_ [~um@dsl-206-55-130-250.tstonramp.com] has quit [Read error: 54 (Connection reset by peer)] 19:31 -!- katabatic_ [~um@dsl-206-55-130-250.tstonramp.com] has joined #xiphmeet 19:31 < dev0> oddsock: Is the software section going to be updated for the new icecast site? 19:32 < oddsock> dev0: a bit, if you have something to add, just send me an email about it and I'll add it.. 19:32 < dev0> oddsock: Does it have to be an email, or would it be enough to note that http://muse.dyne.org is a pretty nice program for streaming? 19:33 < pisef> dev0: I couldn't make it work with icecast2, does it support that? 19:33 < oddsock> ah, yes, i've added Muse... http://www.icecast.org/newsite/3rdparty.php is the new page :) 19:34 < Arc> oddsock, is Freeform listed too? 19:34 < dev0> pisef: I have to admit, that I haven't tried yet... 19:34 < oddsock> Arc: never heard of it...what is it ? 19:35 < Arc> its a CMS I maintain on savannah. it includes the ability to manage Icecast2 streams as well as YP support 19:35 < Arc> you can setup a schedual with multiple playlists for different times of day in mysql, or change it to relay mode, or live input 19:36 < oddsock> Arc: send me an email about it and I'll add it when I get back home... 19:36 -!- purple_haese changed the topic of #xiphmeet to: Xiph.org meeting and conference channel. January 2004 meeting is in progress. 19:36 < Arc> cool. 19:37 -!- pisef is now known as Kato 19:37 < purple_haese> Anything else for icecast? 19:37 < oddsock> nope..thats it... 19:37 < purple_haese> On to Theora and OggFile, then. Arc? 19:38 < dev0> oddsock: See new page looks slightly deformed with Firebird 0.7+ here. 19:38 * oddsock notes it 19:38 < Arc> well rillian and derf aren't here, but I can speak from what I know 19:38 < karlH> dev0: I'm looking into it 19:38 < Arc> last I heard derf's bitstream changes still need to be added to the code, and derf was still working on his experimental theora encoder 19:39 < purple_haese> Yes, that's what I heard, too. 19:39 < Arc> i added libogg2 support for libtheora, you can set an env variable and cause it to compile for it instead of libogg1. the examples are not yet updated 19:39 -!- volsung [~stan@ip68-228-212-16.ph.ph.cox.net] has quit [Read error: 104 (Connection reset by peer)] 19:40 < purple_haese> Is libogg2 support in libtheora complete? 19:40 < Arc> libvorbis still needs to be updated for libogg2 as well, which since there's a larger dev pool could probobally happen pretty quickly 19:40 < Arc> it should be. It compiles and seems to work, the best I can test it 19:40 < Arc> there may be a few bugs still that won't be found until we try to use it to encode something but it seems to decode fine 19:41 -!- volsung [~stan@ip68-228-212-16.ph.ph.cox.net] has joined #xiphmeet 19:41 < Arc> before the examples can be ported to libogg2 we really need libvorbis to be ported. 19:41 < volsung> Argh. Lost power on my laptop. 19:41 < vanguardist> "always plug in at home, son" 19:42 < purple_haese> Couldn't you try encoding a video only stream? 19:42 < Arc> libwrit, for the text subtitle "codec", is being written to use libogg2 as well, which was my reason for porting theora 19:42 < Arc> yea but I don't know SDL or most of the other code in the examples 19:43 < Arc> I looked at it and figured it'd be easier to wait until libvorbis is ported 19:43 < purple_haese> Just rip out all vorbis calls, how hard can that be? ;) 19:44 < purple_haese> Anyway, no need to get too technical here. 19:44 < Arc> I believe the example encoder needs to also support libogg2, that it's not just a matter of the libraries being ported and then the examples will magically work 19:45 < Arc> which blends into OggFile. OggFile is suppost to be part of libogg2, which means the codecs are going to have to be ported to libogg2 either before or during the phase of writting codec plugins for OggFile 19:47 < purple_haese> Unless Monty comes up with libogg3 before OggFile is complete and everything needs to be redone again. ;) 19:47 < Arc> the porting process actually isn't that difficult, it's alot of search and replace with a few code changes here and there (some of the functions take different arguments) 19:48 < purple_haese> I suggest you write up a small how-to on the wiki if you haven't done so already. 19:48 < volsung> Ooh, very good idea. 19:50 < Arc> I don't know if anyone here works with libvorbis, but could definetly use a volunteer to do the porting 19:50 < Arc> I can assist in the process as I'm now familiar with libogg2's api, but I dont think I can manage it alone. 19:50 < Arc> volunteers? 19:50 < purple_haese> As I said, write up a how-to, and the volunteers will come. 19:51 < Arc> purple_haese I think I would need help even with the how-to 19:52 < purple_haese> Well, you could start by jotting down which functions have to be changed how, and let it grow from there. 19:52 < Arc> *nod* 19:53 < Arc> i think a step before that even would be to document the new API 19:53 < purple_haese> Feel free to do that as far as you can :) 19:54 < Arc> last I heard from Monty was that he wanted to collaborate on the libogg2 docs, and I offered to document the API the best I could, but never got started 19:54 < Misirlou> I was working on cleaning up a bit in vorbis/doc/xml. 19:55 < Arc> Misirlou do you want to help with documenting libogg2? 19:56 < Misirlou> I suppose so. I don't have any experience programming with it, though. 19:56 < Arc> I have no experience writting documentation, other than just flat out text or html 19:56 < Arc> so maybe collaboration could work there 19:57 < volsung> Unless you have something else in mind, feel free to copy the format of the libvorbisfile docs. 19:57 < Arc> well there was discussion about adopting a real docs format 19:58 < purple_haese> We need less discussion and more action. Go with whatever works. Just my 2 cents :) 19:58 < purple_haese> I suggest Arc writes up the raw docs and Misirlou marks them up. 19:59 < Misirlou> I'm familiar with XHTML more so than DocBook, so I'll go with that for now. 19:59 < Arc> Misirlou how about you write the templates? thats where I think I got stuck last time 19:59 < dev0> but once you have it in DocBook, you can export into virtually everything 20:00 < Misirlou> Correct. 20:00 < purple_haese> Well writtem XHTML can be just as expressive as DocBook. 20:00 < Arc> *nod* and DocBook format is similar to wiki, which is what im familiar with 20:00 -!- jianxin_yan [~yanjx66@218.21.44.14] has joined #xiphmeet 20:01 -!- oddsock [oddsock@66.242.1.25] has quit [] 20:01 < dev0> So either come up with a good XHTML version, which could be formatted using CSS or do it in DocBook... 20:01 < dev0> whatever 20:01 < Misirlou> Arc: Hmm, you don't really have to use a template. You could just write up some raw text and a file and I'll mark it up. 20:02 < purple_haese> Anyway, Arc and Misirlou will work together on libogg2 docs. 20:02 * purple_haese bangs the gavel. 20:02 < purple_haese> :) 20:02 < Arc> :) 20:02 < katabatic_> DocBook similar to Wiki? Har. 20:02 < Misirlou> purple_haese: By the way, I love your new robe. 20:02 < purple_haese> Anything else on Theora and/or OggFile? 20:03 < Arc> I think monty is still working on vorbis stuff, based on the cvs logs 20:03 < Arc> so no 20:03 < purple_haese> Okay, moving along. 20:04 < purple_haese> Helix integration and relationship with Real is a topic for jack and Monty, neither of which are present... 20:04 < purple_haese> Unless anybody else is present who can speak on that topic, we'll move it to next month. 20:05 < Arc> and monty isnt here to talk about vorbis II 20:06 < purple_haese> Well, I can say a little bit on his behalf there. 20:07 < purple_haese> Monty has recently revamped the bitrate management code in vorbis... 20:07 < purple_haese> The changes are on CVS HEAD, and we need people to test the changes. 20:08 < purple_haese> Monty is worried that something is wrong because it worked right off the bat :) 20:08 < volsung> I didn't quite catch the purpose of the change... Speed or quality improvement? 20:08 < katabatic_> No. 20:08 < katabatic_> Real CBR. 20:09 < purple_haese> It doesn't seem faster. The log entry said something about reduced latency and memory usage 20:09 < purple_haese> and real CBR. 20:09 < volsung> Oh, for the chinese folks. 20:09 < dev0> hmmm 20:09 < volsung> Ah, reduced memory usage is good. 20:09 < dev0> If testing is needed I could do some kind of "test experimental vorbis cbr" sticky thread at HA.org... 20:10 < dev0> if that would be of any help 20:10 < katabatic_> Monty did ask for testing. 20:10 < purple_haese> I suppose it can't harm, except maybe by attracting trollish comments. 20:10 -!- danx0r [~danbmil@c-24-6-154-203.client.comcast.net] has joined #xiphmeet 20:11 < purple_haese> Hiya, danxor. 20:11 * purple_haese should use nick completion :) 20:11 < danx0r> hi guyz 20:12 < danx0r> is there a log URL? 20:12 < dev0> not yet 20:12 < dev0> the meeting is still going on... 20:12 < Misirlou> dev0: Make sure you append the fact that VBR is recommended over CBR in the majority of cases, just to deter whatever may arise. 20:12 < dev0> Of course. 20:12 < purple_haese> danx0r: No live log this time. 20:13 < Misirlou> One second, I may be able to hack something together. 20:14 < danx0r> np -- just lurking really 20:14 < danx0r> seemed like some Theora discussion last month so I thought I would drop in... 20:14 < danx0r> but where is mont, rill & jack? 20:14 < purple_haese> Your presence is appreciated. 20:15 < purple_haese> They seem to have forgotten about the meeting. 20:15 < danx0r> hmmm.. 20:15 < purple_haese> I am posing as moderator. 20:15 < danx0r> not that present company isn't sufficient, of course 20:15 < danx0r> does anyone have more info on vorbis/theora in AVS (chinese EVD thing)? 20:15 < purple_haese> Okay, I am done with reporting what I know of MOnty's recent work... 20:16 < purple_haese> All we know is that Monty changed the bitrate manager to allow true CBR. 20:16 < Misirlou> 20:16 < danx0r> heh that must have stuck in his craw 20:16 < Misirlou> Apologies about the broken charset; I need to fix that. 20:16 < danx0r> it's completely retarded, but that's how the broadcast industry thinks 20:17 < danx0r> Misirlou -- thanx 20:17 < Misirlou> danx0r: You're welcome. 20:18 < purple_haese> danx0r: Since you're here, do you have an update on your progress with the Theora docs? 20:18 < danx0r> hmmm (blush) 20:18 < danx0r> er, well, umm 20:18 < purple_haese> Don't worry, I'm not gonna throw stones, just look at my Ogg Traffic history ;) 20:18 < danx0r> I took a new job in Sept and moved my family 3000 miles, so... unfortunately, no work has been done 20:19 < danx0r> at one point Mau was going to take over, then maybe Derf 20:19 < danx0r> Derf is most def the Theora maven now, right? 20:19 < danx0r> with his encoder & all 20:19 < purple_haese> So, we are officially in a Help Needed stage? 20:19 < danx0r> I think on the docs, yes, for some time 20:19 < purple_haese> Yes, derf is a Theora wizard, but unfortunately he doesn't know python. 20:20 < purple_haese> (From what I recall) 20:20 < dev0> what exactly needs to be documented? 20:20 < danx0r> purple_haese -- oh, right, I started that concept 20:20 < danx0r> has anyone looked at the limited (keyframe only) version I did? 20:20 < purple_haese> rillian has, most likely. 20:20 < purple_haese> :) 20:21 < danx0r> Derf could learn python in the time it takes me to brush my teeth 20:21 < Misirlou> Is derf around? 20:21 < danx0r> but he's busy I suspect 20:21 < danx0r> and of course the spec doesn't *really* have to be done in Python, that was just a cute idea 20:21 -!- derf_ [~lomeando@planetmath.cc.vt.edu] has joined #xiphmeet 20:21 < purple_haese> Right on cue :) 20:21 < danx0r> hi derf_ 20:21 < Misirlou> :) 20:21 < derf_> Hi. 20:22 < danx0r> I dropped in from the dead, but it appears key ppl are missing today 20:22 < purple_haese> Oh, I think the executable-spec-in-python is a good idea in general, but not if it keeps us from making progress. 20:22 < danx0r> Last month's log looked like Theora was getting some action 20:22 < Arc> danx0r I know Python fairly well. where is your code? 20:23 < danx0r> I think it's linked to from the wiki... let me see 20:23 < purple_haese> How about the people that know Theora (e.g. Derf) write up what they know in whatever (pseudocode), and then someone who knows python translates that. 20:23 < Arc> i thought you were developing it monty-style, where no public looksies until it's ready 20:23 -!- Misirlou changed the topic of #xiphmeet to: Xiph.org meeting and conference channel. January 2004 meeting is in progress. Log available at . 20:23 -!- ChanServ changed the topic of #xiphmeet to: Xiph.org meeting and conference channel. January 2004 meeting is in progress. 20:23 < Misirlou> GAH 20:23 < danx0r> Arc -- no, I released something in like july or aug 20:23 < purple_haese> conference channel. January 2004 meeting is in progress. Log 20:23 -!- purple_haese changed the topic of #xiphmeet to: Xiph.org meeting and 20:24 < danx0r> it's in CVS under docs I think 20:24 < purple_haese> Ugh. 20:24 < Misirlou> purple_haese: We have no mojo. 20:24 < Arc> I'd also love the chance to try out the py-ogg2 that I just wrote 20:24 -!- purple_haese changed the topic of #xiphmeet to: Xiph.org meeting and conference channel. January 2004 meeting is in progress. Log available at . 20:24 < purple_haese> Misirlou: Speak for yourself ;) 20:24 < danx0r> the log seemed to be working 20:24 < purple_haese> I just needed the right copy-paste mojo 20:25 < danx0r> Arc -- in the Alpha-2 release, theora/doc 20:26 < danx0r> spec.html 20:26 < purple_haese> That's the latest version in existence? 20:26 < Arc> wait spec.html contains python code? 20:26 < danx0r> spec.html *is* python code 20:27 < Arc> *blink* that's just wrong, man! 20:28 < danx0r> well spec.html and spec.py are textually identical minus formatting 20:28 < danx0r> I generated them both from the same file 20:29 < Arc> ok 20:29 < danx0r> er, spec2html creates spec.html from spec.py 20:29 < katabatic_> Arc: have an URL to spec.html/.py? 20:29 < danx0r> but it looks a bit wrong actually -- I think there's a missing file, spec.txt 20:30 < danx0r> anyway... read the doc, you'll get the picture 20:30 < purple_haese> derf_: Thanks for coming. Would you like to give us a Theora status update from your perspective? 20:30 < Arc> katabatic_ its on my system 20:30 < derf_> Sure. I've made no progress in the last month. 20:30 < danx0r> wierd, dunno why spec2html is there when the original file is missing 20:30 < Arc> danx0r can you join me in #theora? 20:31 < purple_haese> Heh. 20:31 < danx0r> sure sorry to noise the meeting 20:32 < derf_> I discovered I had flubbed up the supra-threshold contrast sensitivity curves, which I need to generate quantization matrices, which I need to get bitrate statistics, which I need to do mode decision, which I need to do P frames. 20:32 < purple_haese> Yikes. 20:32 < derf_> So until that's resolved, my experimental encoder is still I frames only. 20:32 < derf_> And I haven't had time to look at it. 20:33 < purple_haese> No prob. 20:33 < derf_> Yeah, exams, projects, part time job work, holidays... December is not a good month. 20:35 < purple_haese> Alright... 20:35 < purple_haese> jmspeex: Any news from Speex? 20:36 < purple_haese> (Or anybody else who wants to give us status updates on anything?) 20:36 -!- xercist [xercist@dsc02-ari-co-204-32-206-61.rasserver.net] has joined #xiphmeet 20:37 < dev0> seems like there a no further updates. 20:38 < purple_haese> In that case, I move to adjourn the meeting, unless anybody really wants to discuss the MP3 Newswire article that somebody put on the draft agenda. 20:38 < volsung> Not particularly. 20:38 < Misirlou> Nope, looks like trolling to me. 20:38 < volsung> We can grumble about it in private. :) 20:39 < Misirlou> Oh, did anyone want to create a #flac channel? 20:39 < purple_haese> Is there demand for it? 20:39 < jmspeex> purple_haese: I'm still mostly concentrating on the fixed-point 20:39 < dev0> has anybody looked at the article? 20:39 < purple_haese> jmspeex: Right. How's that going? 20:39 < Misirlou> dev0: I did. 20:39 < jmspeex> It's progressing slowly though (mostly because of the free time I've got) 20:40 < dev0> it was basically a Pro-RIAA rant... 20:40 < dev0> nothing to care about. 20:40 < Misirlou> purple_haese: There might be some demand, and for awhile I created one and a few people popped in and out, but I never got enough people. 20:41 < volsung> I second the motion to adjourn. :) 20:41 < purple_haese> Misirlou: I think you answered your own question :) 20:41 < dev0> Misirlou: I'd join :) 20:41 < jmspeex> About the article, I read part of it and though it's not very accurate, it's true that talking about Vorbis II (hadn't heard about it before), makes it sound like those who implemented Vorbis I wounldn't be happy... 20:42 < dev0> jmspeex: That's not even in that article 20:42 < jmspeex> Sorry 20:42 < OggZealot> I'd join too ... sometimes 20:42 < dev0> the article is just a mindless rant about what has been commercially succesful and what not 20:42 < jmspeex> Oh, nm then. 20:42 < dev0> But I agree about your remark on Vorbis II though 20:43 < purple_haese> Well, porting from Vorbis I to Vorbis II is less of a step that implementing I from scratch... 20:43 < purple_haese> especially because MOnty plans to make Vorbis II more hardware-friendly. 20:43 < jmspeex> Of course, if Vorbis II gets started today, it'll take quite a while before it's done, so it probably won't clash with Vorbis I... 20:43 < Arc> monty will probobally release tremor with support for both 20:44 < Arc> and monty said OggFile is a higher priority 20:44 < purple_haese> And WMA9 is AFAIK incompatible with its predecessors (but I could be wrong) 20:44 < dev0> WMA9Pro 20:44 < dev0> there are 2 WMA9s 20:44 < dev0> Std. and Pro 20:44 < jmspeex> purple_haese: MS can affort to release incompatible stuff every 6 months. We can't (I'm not saying that's what we do either). 20:45 < dev0> Std is still compatible with the old v2 decoders (and has the same old horrible WMA quality) and Pro is a completely new codec. 20:46 < dev0> As an end-user I'd think it would be good to have some development happening on Vorbis I... 20:47 < purple_haese> And development is happening. See bitrate management changes, for example. 20:47 < dev0> yeah. 20:47 < Arc> as monty pointed out, vorbis is the oldest "modern codec" 20:47 < dev0> and still performing reasonable competetive quality-wise... 20:48 < purple_haese> Anyway, this is all just idle chatter without Monty to chime in, so I renew my motion to adjourn. 20:48 < purple_haese> Seconds? 20:48 < Misirlou> I move to adjourn. (That sounds cool.) 20:48 * Arc seconds 20:48 < purple_haese> going... going... 20:48 < purple_haese> gone :) 20:49 < purple_haese> So, thanks for coming, everybody. 20:49 < purple_haese> Happy new year, by the way. 20:49 < Misirlou> purple_haese: Are you going to post a more formal log on x.o? 20:49 < purple_haese> Next meeting will be February 7th/8th 20:50 < Misirlou> I mean, you're free to use mine, but yours might be better. 20:50 < purple_haese> I don't know about 'formal', but I'll post my log on x.o 20:51 < purple_haese> I still have to crank out the minutes for December, too :( 20:51 < purple_haese> Holidays. 20:51 -!- dev0 [tobias@p508A5C19.dip.t-dialin.net] has left #xiphmeet ["Leaving"] 20:52 < purple_haese> Anyway, log will be posted within the next 24 hours. Minutes whenever I get around to writing them. 20:52 < Arc> purple_haese how about an ogg traffic when icecast is released? 20:52 < Misirlou> Happy new year to you too! (I'll keep my log up just in case anyone's interested.) 20:52 -!- Misirlou [~aew@c-24-125-118-27.va.client2.attbi.com] has left #xiphmeet [] 20:52 < purple_haese> Arc: Yes, that sounds like a plan. 20:53 < purple_haese> Okay, I'm leaving. See you later.