[00:52] persia, do you have a minute to talk about the -lowlatency kernel? [00:52] yes [00:53] i want to make sure that i understood your comment earlier about uploading it [00:53] my understanding was that alessio should upload his -lowlatency kernel to the ubuntu repositories [00:53] If folks are happy with it, I think it ought be uploaded. I'm even willing to sponsor it (assuming it's not too full of packaging warts). [00:54] He doesn't have permission. He applied once, and it all fell apart in confusion. [00:54] I think that's fixable, but not soon. Until then, someone would have to sponsor the uploads. [00:54] can i reply to his email to me that he can work with you on uploading it? [00:57] I'll just catch him when he shows up here later (should be 8-9 hours, based on past history) [00:58] that would be fantastic :) i feel that i still need to respond to his email, so i'll ask him to talk to you then perhaps? [00:59] Did I get this mail as well? I may as well respond directly in that case :) [01:00] it was an email that was sent directly to me, that's why i feel a response is necessary [01:01] Oh, yeah. Go ahead and tell him I said he could contact me, and I'd upload something. I think he knows me well enough to know that I'm not always around if he needs uploads, but let's not block on him not having direct upload today. [01:02] On a separate note, did you ever hear back from diwic? [01:02] certainly, my most pressing concern is getting the kernel in for natty [01:05] i had planned to contact diwic during work, but unfortunately did not have time [01:05] but i will do so tonight when i have a few minutes uninterrupted [01:07] Good luck. I keep not finding him when I have a few minutes to look :( [01:08] my plan was to actually email him directly [01:10] Heh. That might work better than randomly checking his idle status on IRC indeed :) [01:13] okay, email sent and cc'd you [01:17] i feel that there have been some exciting development for ubuntu studio during natty [01:18] but i would also like to begin to look forward to Outrageous Orangutan [01:20] My recommendation would be to make the switch around FeatureFreeze. [01:20] Before then, tight focus on Natty is probably beneficial, only deferring what clearly can't be done into a "future" bucket. [01:21] Then, come FF, review the "future" bucket, and start working towards strategy and goals for natty+1 to be prepared for UDS, whilst watching Natty for must-fix-now bugs or other release showstoppers. [02:11] persia, i still have two more outstanding items that i feel responsible for currently; the gnome-classic default xsession issue and some problems with tasksel [02:11] i have an "appointment" or general agreement to meet with didrocks for the former [02:12] i have code, others have looked at it and "think" it's good, but my testing has been inconclusive [02:12] my concern is that perhaps my testing is flawed :P [02:12] the tasksel issue is that during installation it seems to offer more than the intended options to install [02:13] rather than have a single item for ladpsa, lv2, and dssi, it seems to offer each one as an individual choice, although i'm not sure what they install individiually [02:13] i've looked back at the seeds and it seems that it should do that [02:14] i'll be talking with colin watson about that beginning next week [02:14] [02:14] BUT i had a thought about the future though [02:14] i would like to talk through it here and hopefully others see it and can offer opinions [02:15] troy has bugged me for awhile about ubuntu studio....what IS it? WHO is it for? [02:15] and i've never had a really good idea, and nobody has seen to be able to really, accurately, and concisely write a wiki page to do so either [02:16] i've given thought on it or how ubuntu studio can differentiation itself more from other music distros [02:16] but troy also said something the other night that is pushing me abit [02:16] i had discovered a really good dvd authoring application (dvdstyler) and told troy about it....he was nonplussed, i think he said "yawn" [02:17] he was of the opinion that it's more or less a dead technology and he's more or less right i believe, it's certainly not the excitement for tomorrow [02:17] so, getting video onto the web is where video is at [02:18] and blender can help with that...quite easily [02:18] but what about music...can we help musicians get their music on the web [02:18] can we help musicians become internet savvy with promoting themselves, posting their music/videos, telling others about themsevles? [02:19] instead of a "me menu", can we offer something like a "music me" menu? [02:20] instead of posting a picture to twitter, can we develop an app that would push a song to a server, create the link, and send "listen to my new song - http://shorturl.url" ? [02:20] can we develop good workflows to help musicians make vimeo or youtube ready videos? [02:21] can we cultivate ubuntu studio to be a "web musician" platform? [02:23] are there other items that people can suggest that would help musicians? [02:37] Heh. Now you're starting to think about the possibilities :) [02:37] Personally, I wouldn't focus on differentiation for the sake of differentiation, instead I'd recommend focusing on enabling exciting behaviours. [02:38] If everyone else adopts the same stuff, that's not bad: it just compliments on the direction you've chosen. [02:38] Similarly, if other folk do cool stuff, there's no reason not to borrow. [02:45] right, but i'm not thinking of differentiation just for its sake, i'm thinking of providing something that would cause users to choose ubuntu studio because it was useful [02:46] the open source nature makes it so that everyone will eventually have the same functionality most likely [02:46] and i can accept it, but if we come first then hopefully we can do it right or better than others who will be "catching up" [02:46] Well, there's ways one can make that hard, but I think that following the spirit, everyone will always have the same stuff. [02:47] I think that strategically it makes sense to pick a relatively small area of focus for leadership, and let others lead in other areas. [02:47] For instance, if the focus is "enable simple and seamless audio/video production, promotion, and distribution", that meets the targets you mention above. [02:48] But still provides space for some other distro to push resources into creating the ultimate DAW, or similar. [02:48] And the result is that everyone benefits. [02:59] i don't want to make it hard for others, that's certainly not part of the open source spirit in my opinion [02:59] but innovation should always keep you in front for a period of time [03:00] Well, kinda. Depends where you're headed :) [03:00] hopefully we can maintain some sense of innovation, i'm not sure it's really happening in other audio distributions [03:00] oh, that's true about where you are headed [03:00] But given the resources of Ubuntu Studio, I recommend starting with a fairly narrow definition of "ahead" if you want to have the perception of being ahead of others :) [03:01] I very much support innovation, but I tend to be interested in innnovation towards some goal, rather than in comparison with others. [03:07] right, innovation without a definitive and worthy goal is hollow, doing so to be "better" than others is doubly so [03:11] i would like to help find and develop a defining aspect of ubuntu studio that establish it so that it isn't simply interchangeable with other audio distributions [03:11] But innovation to make Ubuntu Studio better for folk who want to create stuff and post it to the internet is a cool and worthy goal :) [03:12] well, you are correct :) [03:12] i thought of something else [03:12] Yes? [03:13] another possible idea is that a musician could use a "music me" menu to post a song to various music hosting sites, lke bandcamp or soundcloud, with a single click [03:14] sorry for the delay, wife came home and fmaily is going a little crazy behind me currently [03:14] No worries. I'm cooking, so intermittently latent as well :) [03:15] Ought that be contextual per-song: some sort of nautilus plugin? [03:17] my original thought was how you can configure gwibber or the me menu how your posts can send/retrieve from various formats [03:18] click the "push song" button, navigate to the song to select it, and click the "places" (i.e. bandcamp, etc) to send it to [03:18] maybe an option radio button for posting to identica with a message [03:18] So, what is the complete interaction? Audio or video files don't tend to fit in menus as easily as string input widgets :) [03:19] * persia needs to get more patient [03:19] OK. That makes sense. [03:20] My concern is that it feels like an extra step, as compared to navigating to the song via the file manager, and then sending it to the chosen service via the SendTo menu. [03:20] On the other hand, what you describe is more discoverable :) [03:20] that would work as well, i'm not a coder, i just come up with ideas from time to time :) [03:21] but i think it would serve to have others weigh in as well and brainstorm ideas [03:21] It's just software: the implementation is wildly flexible. The hard part is designing the least invasive interaction paradigm so that the user intuitively is able to accomplish their goals and feels the loss when the features aren't available. [03:21] Oh, indeed. [03:21] maybe i should make another blueprint or a ubuntu brainstorm idea as well, try to pull other people into it [03:22] If you want to encourage innovation in the area of creating content and sharing it online, I'd recommend starting a big thread about that, and soliciting people to write specifications (which they would be willing to help implement) once the ideas begin to solidify [03:22] i'm terribly encouraged by paul.tag's involvement currently, especially so at a time when i was beginning to despair and worry about the future [03:22] I think that's the right path. [03:23] "a big thread" at ubuntu forums? [03:23] or on the mailing lists [03:23] I was thinking mailing lists. You could do forums. You probably know which sort of discussion is preferred by our users more than I (I mostly only hang out on IRC) [03:54] i think we have a small, cusping on vibrant, group in the mailing list, with others sometimes joining from time to time, e.g. ailo or ronan [03:55] it's been a bit of a mixed bag with the forums though, it seems to be those who are well intentioned but do not seem to really follow through or contribute [03:58] In that case, I'd recommend the mailing list as the place for discussion intending to lead to work for the participants :) [13:50] persia, david responded to my email (you were included in the email but i know you don't read everything) and he says he hasn't really done anything as he's pretty busy with canonical work [13:51] ScottL: henningson? [13:51] ScottL, Indeed. Hrm. [13:51] holstein, Yes. [13:52] im going to get some better, more factual data [13:52] i'll try and get it later [13:52] cut and past from the JACK window running that generic kernel [13:57] The problem seems to not be the rtprio part, but the memlock part. [13:58] yeah [13:58] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/690010 [13:58] Ubuntu bug 690010 in linux (Ubuntu) "RT_GROUP_SCHED kernel option makes JACK and Rtkit unusable in Ubuntu Natty" [Undecided,Fix committed] [13:58] i'll see if i can catch ailo later [13:58] and we can add to that bug or whatever [14:02] Does that affect pulse as well? [14:02] hmmm [14:02] persia: good question [14:02] If it does, diwic can probably claim it's paid work to work on it :) [14:03] true [14:04] Maybe already sorted: diwic claims "Confirmed fixed in v2.6.38-rc2. Soon coming to a kernel near you :-)" in comment #15 [14:04] And again in #17 with "Confirmed fixed in 2.6.38-1-generic" [14:06] yeah, thats what i need to check [14:06] ailo and i both found an RT error in the JACK messages [14:06] doesnt make JACK not run [14:06] but, you cant use things like rakarrack or soft synths [14:07] So JACK runs, but without realtime? Even with rtprio set in limits.conf? [14:10] persia: thats the issue we found [14:10] and had ScottL confirm [14:10] we were talking about keeping better testing documentation though [14:10] ailo has emailed a lot of the particulars [14:10] so its in writing somewhere [14:13] i'll have to ask ailo though [14:13] right now im seeing JACK server starting in realtime mode with priority 10 [14:13] with a generic kernel [14:13] in lucid [14:14] lots of xruns though [15:09] Let me reboot and post the jack message from using the -generic [15:20] First time I try linux generic 2.6.38-1. Seems to work. How can one see if it isn't handling memlock? [15:20] jackd message http://paste.ubuntu.com/559552/ [15:21] ailo, You are using the limits.conf settings? [15:21] Or rtkit? [15:22] Should I remove limits.conf? [15:22] I'm using it now [15:23] Or, I am using audio.conf to be more exact [15:24] No, juec checking to answer your question :) Do you have a memlock entry in your audio.conf? [15:24] s/juec/just/ [15:25] Yes. It's the default that came with the jackd installation. [15:25] rtprio 95, memlock unlimited [15:26] In that case, your JACK is working with unlimited memlock. You will probably get performance degradation without that setting. [15:26] The outstanding bit of work is 1) to extend rtkit to be able to also provide unlimited memlock, and 2) to have JACK use rtkit instead of relying on the PAM settings. [15:27] So, next update will fix that? [15:27] If someone does the work :) Personally, I think it will be a bit longer. [15:27] Still, just testing the -generic kernel with realtime on Natty for the first time, the performance is no comparable with -lowlatency at least [15:27] That said, it's great news to know that we don't have a regression anymore, although there is still work to do to be able to avoid the "audio" group without exposing the system to DoS attacks. [15:28] That's expected. Hard numbers would be appreciated. [15:28] abogani, May I publish the -lowlatency from your PPA to the archive, or shall I wait for another upload or two? [15:28] I can only do this comparison. -lowlatency 64 frames/period. -generic 256 frames/period [15:29] But I would need to do some more testing too [15:29] persia: I would want wait for next upload I'm building -38 right now. [15:30] abogani, Sounds like a plan. Will that include the fix for the issue discussed above? [15:30] persia: Do you mean the fix for bug 690010? [15:30] Launchpad bug 690010 in linux (Ubuntu) "RT_GROUP_SCHED kernel option makes JACK and Rtkit unusable in Ubuntu Natty" [Undecided,Fix committed] https://launchpad.net/bugs/690010 [15:30] Yes [15:31] persia: Of course. [15:31] Excellent! [15:31] In that case, just let me know when you're happy with the state of the package, and I'll push it. [15:32] persia: Ok. [15:33] Thanks a lot [15:33] persia: Thanks to you. [15:33] * persia hasn't done anything yet :) [15:34] I'm having some problems with nvidia on 2.6.38-1. I suppose that is to be expected? [15:36] By the way, the actual procedure for update the -lowlatency kernel is **really** trivial. So I'm feel confident to let the package in someone others hand. Is there someone would want start his way as kernel maintainer? Volunteers? [15:37] Actually there is no realtime without the audio.conf on the new -generic [15:38] I removed the @audio lines in audio.conf and got this http://paste.ubuntu.com/559558/ [15:39] Still, nice to have realtime at all [15:43] Do I need to do something to get rtkit to work with jackd, aside from installing the latest -generic? [15:50] * abogani is wondering if we could use file capabilities instead of audio.conf, limits.conf and rtkit... [15:52] Is it too rough? [15:55] abogani, i would like to start myway as kernel maintainer [16:01] ScottL: Ok. [16:05] Any thoughts about file capabilities? [16:22] ailo, If you want JACK to use rtkit, it would require source-level changes in JACK. [16:23] abogani, File capabilities aren't enough. We can do rtprio with rtkit easily enough. The part that is yet to be investigated is the memlock=unlimited, which I don't believe we can express with capabilities alone in a way that correctly restricts it to console users. [16:24] The risk of not integrating with consolekit and restricting to console users is that other classes of user would be able to overdemand resources, making the system unresponsive. [16:24] The initial feedback I received from the #ubuntu-hardened folk is that they would much prefer us to limit to console users. [16:25] persia: Ok I understand. Thanks for explanation. [17:06] ah, i think i have the problem fixed for the gnome-classic default xsession issue [17:06] (or, how lysdexlia screwed up my code) [17:10] Will there ever be a problem us using audio.conf on Natty? I assume we will not be able to live without it, since audio apps need to be changed in order to work without it. [17:11] ScottL: So your headache is over? [17:13] I guess I could ask you persia. audio.conf will be required by UbuntuStudio. I assume there will be no problems with that. [17:27] What is the content of audio.conf, and where is it placed? [17:31] persia: /etc/security/limits.d/audio.conf added when choosing yes to realtime while installing jackd [17:31] Oh, instead of modifications to limits.conf? [17:31] Yes [17:32] I very much hope we can find a solution that doesn't require that, but until such a solution is established, yes it is required. [17:32] So, two lines. One for rtprio and one for memlock [17:33] persia: I'm working on UbuntuStudio-controls which will check if that file exists. I guess I won't be able to change that after feature freeze. Don't know what my options are after that. [17:35] Do you think we ought just ship that file configured when JACK is installed, to save the user from confusion? [17:35] The idea is that the -controls will make sure we get good audio performance, and we can't get it without the audio.conf, audio group and -lowlatency kernel [17:35] persia: That would be great [17:36] Then let's do that. In general, I believe in the philosophy of "Just Works": if we think we need to have that file, let's ship it with JACK. [17:36] I'm for that. [17:36] i believe that jack creates the /etc/security/limits.d/audio.conf during installation (even from vanilla ubuntu) if the user choose to enable rt privileges when prompted [17:36] We can recommend the lowlatency kernel, and install it by default (if everything is sorted), but I'd rather avoid forcing the user to manually adjust things, if possible. [17:38] The problem I would think is to add the user automatically to audio group [17:38] ailo, i hope that the headache is over, i just need to wait for a ppa build and then test it [17:38] As for audio.conf, I see no reason to edit that [17:39] And if -lowlatency could be added as a choice while installing either jackd or Ubuntustudio-*(something) that would be great too [17:40] Or will the -lowlatency end up in the repos for Natty, now? I missed that [17:40] I too believe that the user shouldn't need to make too many choices. [17:41] ailo, i think the changes for the -controls should consider that either -lowlatency or -rt kernel could be installed [17:41] ailo, I'm planning to upload -lowlatency to the archives once abogani finishes the update to 2.6.28 [17:41] even if the -lowlatency is installed from a fresh ubuntu studio install we could still install the -rt kernel from the ppa [17:41] Err, 2.6.38 [17:41] persia: great news. [17:42] Is there observable difference in performance between -rt and -lowlatency? [17:42] persia: not sure, but the underlying changes are pretty huge [17:42] persia, i believe using the -rt kernel allows users with firewire to adjust irq conflicts [17:42] persia: still remains to be seen, however for firewire devices, -rt is the only option sometimes [17:42] persia: the rt patch really changed the schedule code [17:43] for some users, an -rt kernel, even an older one, is a must to allow their firewire devies to function [17:43] that is what i have been told [17:45] If the -rt will be slightly better, I can believe people will want to use that for extreme low latency too, but if not, it is most likely only suitable for firewire users as a last resort. [17:45] Someone ought investigate that. I've been told by any number of kernel developers that -rt is really only interesting from an embedded viewpoint. I've nothing against it, but I wonder if there isn't some other factor confusing reports. [17:46] I know that -rt sometimes increases overall system load, which can have negative effects for some audio applications, although I generally recommend against anyone loading their system enough to notice this sort of thing. [17:47] I can't see any reason why -lowlatency shouldn't be regarded as the main multimedia kernel, and the -rt more as something experimental. Again, it is the last resort for some. [17:47] ailo, yes [17:48] once the -lowlatency gets into the repos it will included on the dvd as the "default" ubuntu studio kernel [17:48] that doesn't help people who start with vanilla ubuntu and install ubuntustudio packages though [17:48] that would be where the -controls checking script comes in [17:49] it would be nice to offer the choices (lowlatency or rt), giving explanation why each could be chosen, and then allow the user to select [17:49] If we add the option to install it from -controls we'll need to make sure the user is updated on the difference between the two [17:49] and selection option could be "none - stop checking it for me" [17:49] Yes [17:50] on my system i would want to install the -lowlatency, but i don't want the check to keep checking it and telling me the -rt kernel is available when i don't want it [17:50] and "opt out" choice would be very nice [17:50] * ScottL is taking son to lunch then library, be back in a few hours [17:50] I would let it check if either of the two is installed and then recommend the -lowlatency. If any of the two is installed, the user made that choice and we won't bug them about it [17:51] ailo, You'll be using python-apt to check and install the kernel? [17:52] persia: don't know yet. I wrote a script that did that, but I will need to redo a lot if I make a real gui for Ubuntustudio-controls [17:52] My first unfinished script uses zenity to do all that [17:53] The plan was that the script is run at startup to check the system and notify the user if something is missing [17:53] And if the user decides to make changes, -controls is opened, which also can be accessed manually. And that woul be written in python. [17:54] To be clear, the first script was a bash script. [17:56] How are you running it "at startup"? [17:57] As a startup program, which I added manually. I suppose that can be added as a part of the installation. [18:00] There's just one thing that is bugging me when it comes to changing settings and such. Let's say the user installs -lowlatency from Ubuntustudio-controls. Now, the user needs to choose kernel at every boot. [18:01] Why? [18:01] Seems unnecessary, while it would be unusual to uninstall the -generic. Is it possible to make Grub always choose a -lowlatency kernel. [18:01] The standard method is to have the bootloaders boot the last installed kernel unless the user modifes their choices. [18:02] Yes, but then comes an update... [18:02] You will only have this problem when installing from repo and not from the alternate DVD [18:05] I would actually vote for removing the -generic just to save some trouble, even if it is not so much trouble [18:06] There's a good argument for that. The main opposition being that historically we've tried not to remove anything without user intervention. Maybe it is worth investigating the bootloader update scripts to allow one to express a preference for one or another *type* of kernel, regardless of updates. [18:07] I guess making Grub smart would be more standard like. [18:09] Always best to try to improve the code that one finds troublesome, rather than trying to work around it. [18:10] hey there [18:10] In this case, I don't think grub itself needs to be adjusted: just the code that updates grub when the kernels are installed/updated [18:10] I noticed you guys we're looking for a website cleanup/remodelling [18:10] what CMS do you consider best - joomla, drupal or wordpress ? [18:16] persia: I guess I should check it there already is a mechanism for choosing a kernel by name first. Thanks. [18:17] Good luck. [19:37] ScottL: Nice work on the Plymouth theme, but now it's hardly showing at all. At boot-up I only see it for less than a second, after a few seconds of purple. The purple I'm assuming is a part of the Grub theme [19:37] If we want to have a unified Ubuntustudio-look I would suggest someone had a look at changing the Grub theme as well. [19:38] Or, is it scott-upstairs? :) [19:42] Also, I had to reinstall it after some updates I did. Will that be a problem later, when the theme is in the repo? (I guess not). [19:43] The theme was replaced by a blue screen, that is why I had to reinstall the Ubuntustudio Plymouth theme. [20:09] ailo, thanks on the plymouth theme (i actually thought it might have been just a touch too big now), but i don't believe i can control how long it stays on the screen [20:10] i had mentioned to holstein before about using the same graphic possibly for plymouth, grub, gdm, and background [20:10] ailo, also check with paultag about the -controls update, he's working on it as well with doctormo [20:11] i suggested that paultag get involved because i believe it too be extremely important to get the udpate done in time for testing so it can be included with natty to appreciate the user's experience [20:23] ScottL: glad to hear that paultag is working on that. I should follow their work instead of pursuing my own. Still only learning the gui stuff, since I've had too little time. [20:24] ScottL: Didn't try your update on the plymouth theme, though. So, I haven't seen the new size. [21:43] * persia notes that in an ideal situation the boot is too fast to even display plymouth: one only sees it for slow boots. [21:52] ailo: I want to hand this off to you after it's stable [21:52] ailo: you seem like you'd make a good maintainer :) [21:53] paultag: I would like to see what is done, since I already got involved. Can't predict what will happen for the next release. [21:54] ailo: OK [21:54] I'm ready to do some work on details if necessary. At least have the opportunity to slip in an idea [21:59] paultag: Do you guys have a list of features for it? Ubuntustudio-controls that is. [22:02] ailo: on the wiki somewhere. doctormo and I are fleshing it out [22:03] paultag: This one? https://wiki.ubuntu.com/UbuntuStudio/ControlsRedesign [22:04] ailo: https://wiki.ubuntu.com/UbuntuStudio/SettingsApp/Redux [22:11] paultag: Thanks. I discussed with persia about maybe letting jackd install the audio.conf by default. Don't know who can make this change. I would vote for that file to be added in the same fashion, if needed by -controls. No need to adjust values, just add the file as a part of adding user to audio group, since those go hand in hand.. [22:12] paultag, ailo , i love to learn more about the updates as well that you are working on, and if no one else is there i will volunteer to be the maintainer [22:12] ScottL: oh, no, I'm here for the long haul -- if no one really wants to take it over, I'll maintain it no problem [22:13] ScottL: I'm just looking to give anyone at all their shot [22:13] ScottL: and you're project leader, so stuff it ;) [22:13] paultag: (The file /etc/security/limits.d/audio.conf being the file that includes rtprio and memlock settings.) [22:14] ailo: yeah :) [22:17] paultag: so far, I at least need rtprio set for me to use realtime on the latest -generic, so perhaps this will be needed in the future too? Seem like a lot of things are a bit uncertain. Anyway, that's my two cents. [22:18] ailo: we'll chat, but I need to run to a wine tasting :) [22:19] paultag: Ok. Have fun :) [22:19] ailo: talk with you soon :) [22:30] ScottL: As much as I would like to learn more about these things, I can't promise much for the next release, so I'll just focus on this one. At this point I think I should just focus on the testing documentation. If my startup-script is needed for anything, just use it. It will need to be adjusted to work with -controls, if used at all. [22:32] paultag, that link for doctormo's image didn't bring anything up :( just an imagebin page with no image [22:32] ailo, that's understandable about the next release and i appreciate you being forthcoming about it :) [22:33] ailo, and you would aboslutely ROCK if you made any progress with the testing documentation [22:34] paultag, i'm an idiot sometimes, it's the logo for -controls [22:34] ScottL: Thanks. Shouldn't have to be too in to detail. Just nice to have a list of things to go through. [22:34] paultag, i was looking for a dialogue box type configuration page [22:34] ailo, that's true :) [22:50] paultag, kids were keeping my attention (even when looking at doctormo's image), looks pretty cool although the center reminds me of kde [23:14] i have SUCCESS with my gnome-classic xsession patch! i'm so bloody happy :D [23:14] fresh install of natty, installed my ppa build, and new users default to gnome-classic [23:33] ScottL, Excellent work! [23:39] thank you, persia [23:39] i'll see if i can't get a patch uploaded tonight to the bug report