[01:49] hello UbuntuStudio folks [02:29] Kokito: hey [02:36] how is it going holstein ? [02:42] not bad, working on a few blogs [02:42] http://ashevilleoriginalmusicseries.blogspot.com/ for example [02:51] interesting [02:51] * holstein still working on content [03:01] hi Kokito [03:01] sorry i disappeared the other day [03:01] hey ScottL [03:01] that's OK :) [03:01] my computer does these weird locks [03:01] I also disappear quite frequently :) [03:01] i can move the mouse but i can't change windows, pick anything, or switch desktops [03:02] i can't even shut down properly, i end up holding the power down till it shuts down :( [03:02] Kokito, how is the website going? [03:03] at some point i really need to give some serious thought to how/where information is going to be [03:06] Have not had time/motivation to work on it lately; but I am sticking to the 6/16 deadline documented in the wiki :) [03:07] ScottL: tty? [03:08] you really *will* do something to the hard drive like that :/ [03:08] ScottL: discussing/brainstorming what to put where can be done now; in fact, it has to happen now, as how the website evolves will depend on what needs to be supported. [03:11] ScottL: there are a number of questions poised in https://wiki.ubuntu.com/UbuntuStudio/Oneiric/Timeline#Update the website with regards to potential features that can be supported; there has been no discussion/decision on whether we want to implement each of them or not on the new website [03:28] ok, time for my evening coffee [03:28] going out for a while. will be back later [04:39] re [05:02] my Chrome netbook keeps loosing its wifi signal :( [21:12] i would be curious to see who fully understand's ralf's email to the -dev mailing list, i am afraid many of the topics are beyond my current understanding [21:12] i think ailo_ probably has the best chance of explaining it to me ;) [21:15] ScottL: Which emails are you specifically referring to? Granted, I haven't read every piece with thought, but nothing strikes as particularly odd. [21:18] astraljava, https://lists.ubuntu.com/archives/ubuntu-studio-devel/2011-May/003249.html [21:27] * holstein [21:27] i should appologize for that [21:27] i talked him into it [21:27] i just wasnt 100% clear with him === rexbron_ is now known as rexbron [21:28] i didnt mean to join the -dev list and rattle off request [21:28] i was hoping he would ask about CPU scalling being added to ubuntustudio controls [21:28] which is actually a great ida [21:28] idea* [21:32] Speaking of Ubuntustudio-controls, how is that going? [21:33] i want to say ailo_ and paultag are colaborating on that? [21:38] Ralph seems to talk more about his own specific issues I think, but I was interested in the cpu crequency scaling part [21:39] I will code a simple usable controls that should be packaged and ready during June [21:39] But, there are all kinds of ideas for improving the -controls [21:40] rexbron, The current -controls would need to be updated. You're the original author, right? [21:40] rexbron, Just a few changes for Lucid, and Maverick [21:40] ailo_: you can blame it one me ;) [21:41] holstein, i'm glad we have his input...seriously [21:41] too often things get brought out by just one person [21:41] I'm not sure, if it started with Karmic or Lucid, but US-controls should not edit /etc/security/limits.conf anymore, since jackd creates /etc/security/limits.d/audio.conf [21:42] hi rexbron [21:42] ScottL: ralph has C experience [21:42] hey ScottL [21:43] and ralf says he's not a coder :P [21:43] it would be great if he transitioned from whiny list guy, to contributor [21:43] lol neither was I when I wrote -controls [21:43] *im glad to have his input too [21:43] rexbron, My idea is to have a UbuntuStudio-controls that exists in the system-tray or indicator-tray, and at least let's you administer realtime prio for users [21:43] rexbron: w0w, i didnt realize that was yours... cool :) [21:43] ailo_: I think that is a good idea [21:43] rexbron, Since Natty, we don't need to mess with firewire [21:44] holstein: the one any only piece of fuctional code I've ever distributed [21:44] ailo_: that's a plus [21:44] rexbron, It would be great to include some other stuff, like pulse-jack bridge as an option [21:44] I've been out of the loop for a while and am running stock ubuntu [21:45] ailo_: i think thats the only way we can [21:45] * rexbron looks at the redesign page [21:45] i dont think it will ever trickly into the main repos [21:45] pulse-jack* [21:45] pulse-jack being a seperate package? [21:46] trickle* [21:46] or is that getting pulse compiled with jack support [21:46] rexbron: falk made it for KXstudio [21:46] I still don't know what pulse-jack is. Isn't it just a script? flaktx has a solution to that anyhow [21:46] holstein, Is this different than the pulseaudio-module-jack we have? [21:46] falktx* [21:46] persia: AFAIK, its very different [21:46] theres a thread on the list that ScottL has been involved with [21:47] i think thats about what we have by default apparently that i wasnt aware of [21:47] OR i dont know how to use [21:47] either are likely ;) [21:48] We don't have pulseaudio-module-jack by default. [21:48] persia: we = ubuntu? [21:48] like, its not in the repos at all? [21:48] And it doesn't have any special interface: if it's set up, pulse tries to register to jack on startup (which doesn't always work as expected, as JACK is often not running at that point) [21:49] ScottL: Oh I see, I missed that one. Whee! :D [21:49] It's in the repos, just not installed by default by any of the flavours (including Ubuntu Studio) [21:49] persia: maybe falk brings that in and uses it* [21:49] That would make sense to me. If it was "very different", I'd worry about duplication of effort, etc. [21:50] ailo_, i'm having trouble getting pulseaudio-module-jack to work as david h. has suggested on the mailing list [21:50] ScottL: im hoping to catch up with that thread [21:50] he made is sound simple [21:51] i personally didnt get it going til falks package, not that i tried terribly hard [21:51] Is there a page for the ubuntustudio-controls redeisgn? [21:51] holstein, aye, that he did....just didn't work for me yet [21:52] rexbron, Are you interested in coding us-controls? I'm supposed to start soon, but I'm guessing you have a lot more experience in coding [21:52] ScottL, holstein Seems to me falktx is the man to talk to about that [21:52] ailo_: lol, I'd be intersted in helping. As for coding skills, they are pretty pathetic [21:52] * rexbron 's BFA counts for something though [21:52] ailo_, but he keeps doing stuff that we simply are not going to get into the repos [21:53] ailo_, i don't think solutions proposed by falktx are going to work for us [21:53] What is the status about per-package uploading/ seeding domains? persia? [21:53] i certainly wished it was otherwise [21:53] ScottL, I think it would be good to pick out the things we can easily handle at least, and see if we can put them in [21:54] rexbron, there is an older page about -controls update, would you like me to find it? [21:55] rexbron, All the infrastructure is in place: it needs a request from the flavour (best if from the leader), and then individuals become eligible to apply. [21:55] ScottL: sure, I think we need to look at what exactly still is needed to be done with -controls [21:57] rexbron: What we absolutely need is probably only administering realtime privilege for users. [21:57] rexbron, there are several to choose from :P https://wiki.ubuntu.com/UbuntuStudio/ControlsRedesign?highlight=%28ubuntustudio\-controls%29 [21:57] https://wiki.ubuntu.com/UbuntuStudio/ReleasePlanning#Ubuntu%20Studio%20Controls [21:57] rexbron, paultag had the idea to make us-controls pluginable. This is something I've thought about too. [21:57] ailo_, That should be rtkit. I thought that the main outstanding bit was memlock control. [21:58] ailo_: plugin able? How so? [21:58] paultag had one but i can't find the webpage [21:58] rexbron, The main application in a system-tray menu would be a system settings application [21:58] rexbron, i have hoping to have ubuntustudio-dev set up soon as a delegated team within a couple of months for upload rights [21:59] rexbron, If someone wants to add features for the controls, just add an application that will end up in the menu. That's my idea. Don't know if it will happen though [21:59] ailo_: simplier to just modify the source of controls [21:59] persia, As far as I understand, rtkit will not be the way we get realtime privilege to jack [22:00] rexbron, I've already made a new app. Just missing some python specific code. [22:00] ailo_: link me to the branch? [22:01] ailo_, Is that because someone actively is opposed, or just because it wasn't done? I thought the main reason it wasn't done yet was that it wasn't enough without a memlock solution. [22:01] rexbron, I don't have the code up right now. I can upload it to gitorious.. [22:01] ailo_: I don't know if this has been decided project wide but using bzr makes things integrate a lot better [22:01] persia, I've only been talking with las. I have no idea who is working on that on the Debian/Ubuntu side [22:02] ailo_: also, there is already a -controls project on launchpad [22:02] ailo_, Was there some decision from las that rtkit should not be used, or only the prior notes about it not solving the problems, so not worth doing? [22:04] persia, I asked him if the way we access realtime privilege will change, for example using rtkit. He said no. He also doesn't think adjusting memlock is needed. He thinks memlock unlimited is fine. [22:05] memlock unlimited *is* a change. [22:05] That's why the configuration file is needed. [22:06] If we had a way to do per-application memlock, it would make sense for JACK to request it. [22:06] persia, I would like to find out if adjusting memlock has any positive effects, though. Some say they will have system lockups with memlock unlimited [22:06] and if that was happening, switching to rtkit at the same time makes sense. [22:06] Yes. [22:06] That's why ideally memlock should only be unlimited for jackd. [22:06] But that requires writing memkit, or some similar thing. [22:07] ailo_: hit me up when you have the link, I'm curious to see what you've done :) [22:07] (which I believe nobody has done) [22:07] But now is a better time for ubuntustudio-controls: we're not going to fix the memlock-is-only-configurable-by-user/group issue within Ubuntu Studio. [22:08] rexbron, Ok :P. It uses scripts to execute the commands, and I was not thinking of using it as a base - more as inspiration for a UbuntuStudio Controls system settings application. [22:08] ailo_: Scripts are fine, just import them into the app indicator [22:15] ailo_: I'm waiting ;) [22:16] Working on it [22:16] Just had to check it works first [22:21] rexbron, git clone git://gitorious.org/old-uscontrols/old-uscontrols.git [22:21] rexbron, If you want to see it in action, add the files ending with *.sh to /usr/lib/ubuntustudio-controls [22:24] To get the memlock value updated at each boot, I think it would be good to have a script that does that in /etc/init*. Just add a percentage value in /etc/ubuntustudio-controls and have the file /etc/security.limits.d/audio.conf be updated if needed. I'm thinking of the possibility that someone adds or removes memory from their system. [22:27] ailo_: that strikes me as a rather rare occurance [22:27] some general comments [22:27] having seperate bash script files is less idea I think [22:27] I guess what I don [22:27] rexbron, This application was a hack, so only to be regarded as a design idea, nothing more [22:28] I guess what I don't have is a grasp of what exactly needs to be done [22:29] people still wish to set memlock [22:29] rexbron, For me, I think administering realtime privilege is the firstmost important feature. The other to adjust memlock. Then, there are further ideas that can be explored and added. That is from what base I was intending to work from in the coming weeks. First code a simple but well functioning app, then add features if wanted or needed [22:30] ailo_: may I suggest a different design? [22:30] rexbron, Sure [22:31] if you go the appindicator route, I feel that it would be more accessable to users and more clearly focused in it's intent [22:31] rather than a full fledged application [22:32] rexbron, That's the way I want to go. Make an app indicator. I also have been working on a system check script in usermode that checks if user has realtime privilege and if user is running a lowlatency kernel [22:32] I mean, make an indicator menu, with an item called "system-settings" where you can adjust some things [22:34] Back to the system-check script: If user does not have the goodies for realtime audio, it announces that. [22:34] ailo_: shouldn't they have those by default after installation? [22:34] rexbron, Not if they use Ubuntu Vanilla as base [22:35] ailo_: I wouldn't consider those supported users [22:35] ailo_, rexbron , i wouldn't mind if -controls even expanded to help with turning pulse off/turn jack on, transport for jack, connections for jack, and/or even install new applications based on work flows [22:35] but this is "pie in the sky" stuff, not just mundane [22:35] ScottL: there are simpler, existing solutions for all of those imo [22:36] ScottL: at least with my vanilla ubuntu 11.04, pulse shuts off when jack is started [22:36] then restarts when it's closed [22:36] Is that not the case in studio? [22:39] I believe, since US is in the main repo, it should be directed to all Ubuntu users, not just Ubuntu Studio users. It's not a big stretch to let non technical users adjust their system for realtime audio by allowing us-controls to handle that. [22:40] I would argue that if they choose to go that route, then they are responsible for configuring their system but that's me [22:40] * rexbron dislikes handholding [22:42] I'm also not in favour of handholding, but I don't believe we can safely rely on installer behaviour to dictate system behaviour, especially as the number of methods of installation keeps growing. [22:43] fair enough [22:43] but where exactly do we draw the line? [22:44] I guess I don't see do it once actions as being something to include in controls [22:44] I don't like the idea of having all-in-one applications [22:44] for things like that, perhaps a gnome-system-controls item could be used [22:45] but that won't be included in xfce [22:45] if us-controls would be pluginable, you could add apps like qjackctl to the menu [22:45] I don't think we want to stuff everything into ubuntustudio-controls. [22:46] but why? [22:46] why would you want or need qjackctl in that meny [22:46] menu rather [22:46] If you want qjackctl to be more tightly integrated, there are useful alternate tools to manage that. [22:46] * rexbron mentions lash [22:47] Well, lash is deprecated these days in favour of ladish. [22:47] But I was thinking of some of Hans Baier's work. [22:47] e.g. jackpanel [22:48] Another idea is having a multimedia app-manager, that has a different database of info, than the standard one [22:48] http://www.hansfbaier.de/wordpress/jackpanel/ [22:49] (this isn't in the repos mostly because not enough people reviewed/advocated Hans' work, not because it isn't any good: Hans was even on this channel for a while) [22:51] One thing that I liked about KXStudio was the list of applications that showed what they supported - ladspa, vst, etc [22:51] That sort of oversight is often missing [23:03] rexbron, re: pulse shuts off when jack is started...yes, that is currently the case, qjackctl.dekstop runs pasuspender first [23:05] I believe jackd2 disables pa too, but not jackd1 [23:05] If run from a terminal [23:06] That's a bug, or rather a leftover workaround from a bug. [23:06] JACK should suspect pulse if pulse is using the sound card JACK is using, and ignore it otherwise. [23:07] For extra points, if pulseaudio-module-jack is available, and JACK wants to use the sound card pulse is using, it should reconnect pulse as jack [23:07] * persia gets tired of shaking and hides in a doorway [23:08] persia, i'm trying to work with david h. and get this resolved, but i'm not sure it works like everybody thinks it does [23:08] Right. [23:08] It doesn't :) [23:09] * ScottL thinks of princess bride "i don't not think that means what you think it means" [23:09] Continuing, reconnect pulse as JACK IO points, in case the user wants to do something, and to not annoy other sound-generating applications. [23:09] persia, weird, are you prescient? because you answer that statement almost immediately after i hit enter :P [23:10] I live 15 or 16 hours in the future, so it's not quite as fast as you think it is :) [23:11] oh, well that explains it and i feel much better now , prescient...pttthh! you live in the future, now that i can believe :P [23:11] ailo_, how do you find kxstudio listing supported applications helpful? [23:12] So, I've reviewed the relevant thread. [23:12] There's a few things that are messy. [23:12] ScottL, Just seeing what sequencer works with what plugins and instruments make life a lot easier. [23:13] 1) The volume controls most folk use don't necessarily have a good view into JACK. [23:13] ailo_, ahhh, i can wrap my head around that...yes, i think that would be helpful [23:13] 2) Just creating inputs/outputs in JACK doesn't always do what the user expects [23:13] 3) We don't install integration by default [23:14] 4) Our helper tools don't invoke integration with the typical configuration. [23:15] 5) I've yet to see anything written on the issue that provides a generic solution for both the single-audio-interface and the multiple-audio-interface model. [23:15] ailo_, I actually don't much like the idea of giving the users a big list of applications and telling them which bits work with which other bits: I'd rather try to pick a set of functionality to support, select apps to support that functionality, and make sure they can all use the same sorts of formats. [23:16] Yes, this means we don't get all possible functionality, but it also makes it much more accessible to just get things done. [23:16] persia, to which thread are you referring? [23:16] And this reduces the number of folk who get frustrated and just go write their own tool from scratch (e.g. jokosher) [23:17] persia, You mean select applications for a standard US install that supports the functions we want to use? [23:17] ScottL, Subject "PulseAudio and Ubuntu Studio 11.10" starting from ckontros writing "As it is so tightly woven into Ubuntu, I suspect it will. We are however looking to better integrate JACK with PA. (seems like we've been trying this forever)" [23:18] ailo_, Rather, the functions we intend to support as first-class use cases. I'm not convinced that this will precisely match the functions any one of us wants to use. [23:18] The idea is to pick sensible defaults for the majority case, and then make everything in those defaults work together. [23:18] Those with more specialised needs can select from a wider software set, with the expectation that other things may not work together quite as well. [23:19] We can't fix all of Linux Audio (nor do I think this is a good idea: attempting to do so would stifle innovation), but we probably can help make sure a few applications to achieve some goal work together well, achieve common workflows, and are well documented. [23:22] (mind you, these are my ideas: I don't expect everyone to agree :) ) [23:25] my largest desire is to clean up the mess between jack and pulse....i think that will alleviate many headaches [23:26] persia, I agree with you. I would rather have a small set of applications in a standard US install, that we know will work for the most common use cases, than have a full variety. [23:27] ailo_, Heh, OK. Sorry to jump on you: I just worry that we'll end up solving the problems that users complain about and miss solving the problems that cause few people to bother being users. [23:27] i don't expect this, but i would like it for canonical to hire someone just to sort this out [23:27] s/like it for/appreciate if [23:28] as an aside, i started getting my nanokontrol set up :) [23:28] ScottL, I think that if we patch qjackctl, and have it recommend pulseaudio-module-jack, and add an autoconnect script to pulseaudio-module-jack, and use jackdbus by default, we have a chance: in my mind the unsolved piece is that I don't believe the current setup works well for folks with multiple audio interfaces (like yourself). [23:29] If waiting for Canonical to do things was how we did things, there would never have been an Ubuntu Studio. [23:29] heh, very true persia [23:29] To inform users what they can use and how to use it, documentation/info is the first and most important way to do that, but if a multimedia app-manager would make things even easier, I wouldn't mind using that. [23:29] As much as I appreciate all the contributions Canonical makes to Ubuntu, I feel very strongly that we can't rely on them to make sure our goals are met (meaning each of our individual goals, not just Ubuntu Studio goals) [23:30] ailo_, It might, but my fear is that we'd be deluged with requests to add information about new/different application mixes, and it would become a support nightmare. I could be wrong (and I typically don't do support) [23:30] well, if canonical would hire me for ubuntu studio, maybe i _could_ make them all happen :P [23:31] Do you think that Canonical would be markedly better at gaining money from your work on Ubuntu Studio than you could be yourself? [23:33] persia, It would need to be a separate application no doubt. But, it would make sense having it a part of controls, if it were available in Debian repo for instance. Just entertaining the idea so far. [23:34] persia, no i do not think so [23:35] ailo_, I suppose. I think it's a mountain or work, but we all need to do the things that we want to do, so if you think it's important, go ahead :) [23:35] ScottL, In that case, I'll say you'd get paid better to work on Ubuntu Studio if you did it yourself. [23:35] * persia tries to come up with three business models for Ubuntu Studio developers [23:35] persia, I'm only concerned with the design of us-controls and what it could be used for in a longer perspective while making it very simple at first [23:36] 1) Provide installation/training/support services to "basement" studios in your city. [23:36] 2) Run a "basement" studio [23:36] 3) Use Ubuntu Studio and "found sounds" to generate soundfonts for use by foley artists [23:37] ailo_, Personally, I think that the long-term goal should be to not have ubuntustudio-controls. [23:38] persia, That would work too [23:38] For packages we aren't sharing with other flavours, we should just patch them to do the right thing (sometimes this is hard, like JACK and memlock) [23:39] For packages we are sharing with other flavours, we should try to find common default configurations: if this is impossible, we should extend the package to allow drop-in default snippets (for examples, look at how gconf-defaults works, or gdm) [23:40] And in certain special cases where we can't come to agreement, we should fork (examples are the variety of linux-* packages in the archive, or having both "mutter" and "metacity") [23:40] * saidinesh5 really wanted something like GetHotNewStuff for Soundfonts ....... download and share all the freesound fonts [23:41] saidinesh5, Last I knew, there were about 3 free soundfonts. The tools need a bit of work, and the soundfont community needs a fair bit of cultural adjustment to generate free artifacts. [23:42] persia: over the last few years i ve collected a lot of nice and free soundfonts in the public domain [23:42] Unfortunately, "the public domain" is meaningless in most countries. [23:43] as in? [23:43] Which then causes such soundfonts to revert to regular copyright models, which makes it a crime to distribute them. [23:43] ohh [23:43] I believe that the UK is the only EU member that recognises "public domain", just to pick a few examples. [23:43] but the creators themselves say "do whatever you want" [23:43] They need to do that in a way that is binding. [23:44] Creative Commons has some good licenses. [23:44] oh [23:44] persia, for business models, troy_s had a good one of getting ubuntu studio into educational institutions and then charging for support [23:45] well ScottL are we really ready for that ?? [23:45] saidinesh5, Don't get me wrong: I'm *really* happy with the growth in the zero-cost soundfont community: there are some tools bugs, but lots of the rest can be solved with education. [23:46] ScottL, I know several people who do educational support to fund their work in Ubuntu: apparently it's highly personally satisfying, although dealing with the School Boards to get in can be annoying. [23:46] saidinesh5, it the person knows the ins-and-outs of ubuntu studio and can do after installation setup :) [23:46] persia: and especially with musicians have 0 tolerance for any fuss [23:47] saidinesh5, Depends on the musician, but often, yeah, one doesn't bother learning about contracts until one has an agent who does that for one. [23:47] i tried to get my friends try things out and installation itself is the end of the story [23:48] I'm not gonna chime in on the technical stuff on the list (as I here for 1 thing really) but VST support from packages in the official repos won't happen. [23:49] but there are a lot of free VSTs which work natively.. arent they? [23:50] Well, at least not before 13.10 [23:51] saidinesh5, Steinberg would have to extend a license to someone to be able to ship a working host. There's debate in the area, but most folk try to avoid violating known patents. [23:51] saidinesh5: IIRC simply having VST support means having packages like Ardour with support for ALL VSTs. Unless something has changes lately. [23:51] no but the free implementation of VST standard? [23:51] *changed [23:52] * ckontros will let persia handle this. [23:52] ckontros, Well, there was VST 3.5 in February: I'm not sure that any of the hosts we have support 3.5 yet. [23:52] a good support for VST 2.x itself is a big deal though [23:54] ardour 3 does midi too right? [23:54] persia: Has the implementation been "freed" in some way? A new free version of VST? [23:54] saidinesh5, My understanding is that there are some implementations that attempt to support some versions of VST, for which there is free code, a couple issues occur: [23:54] * saidinesh5 has really lost touch with the multimedia apps in foss world for the last 2 years ....... hoping to get back to all of them now that he has graduated [23:55] 1) There are some patents (starting in 1996) on VST technologies, which would need to be licensed to sell a product (and Ubuntu Studio must be able to be sold). [23:55] saidinesh5: Are you Ralf? (from the list) [23:56] ckontros: nope... just got interested in ubuntu studio (once again) just a few days ago [23:56] 2) There are some dependencies in many VSTs that require them to be run in Windows, or in WINE. Since WINE is virus-compatible with Windows, we probably want to think several times before installing it by default (note that we did install it by default last I looked). [23:56] saidinesh5: Ok. I was wondering. [23:57] persia: iirc there are some native linux vsts too [23:57] we can start with them ? [23:58] saidinesh5, Sure. Just get a license from Steinberg to distribute a VST host under a license that allows those to whom you distribute it to to redistribute it. [23:58] then grant that license to your favorite open source VST host, and watch it become everyone's favorite within a month. [23:59] holstein: Mention to Ralf that saying things like "I can try to help a little bit, but I'm not a coder and I don't want to spend too much time with testing etc. any more." really isn't the best way motivate anyone to help work on things. ;) [23:59] persia: how about Vestige? [23:59] the way the lmms folks do it