/srv/irclogs.ubuntu.com/2010/06/13/#ubuntustudio-devel.txt

rlameiroScottL: i think i cant make it to the meeting today12:46
rlameiroScottL: i have to go to the ariport and i am not sure if i get home in time12:47
rlameirojussi: are you there?12:53
rlameiroI am thinking on a implementation of the UBS controls12:53
rlameirowhat do you think if the controls utility, could download from a server a text file with the list of PPA we determine safe for ubuntustudio12:55
rlameirothis way we wouldnt need to upgrade the package each time some PPA changes, or something new is needed to be added12:55
rlameiroif so, where can i put that file? some safe place, also it sgould be GPG signed, but i dont know how, maype gziped or something12:56
ScottLrlameiro, i don't think we are ready for a meeting today, i know i'm not13:13
rlameirook13:13
rlameiroScottL: what do you thing about the file download?13:14
ScottLrlameiro, hmm, i'm not sure, i'd have to think about it13:53
ScottLmy first thought is that i would rather we test the ppa ourselves and then copy the build into our own ppa13:53
ScottLthis keeps a single ppa for user's to worry about13:54
ScottLbut, this does not prevent us from fixing the rtprio and memlock settings, which is actually pretty important13:54
rlameiroScottL: well, but the backpots need to have its own ppa13:55
rlameiroi need to look into the rtprio stuff13:55
rlameiroit needs to be added to the audio.d/config isnt it13:56
ScottLrlameiro, true, but we could include one other for "extended" universe ppa (or whatever name we come up with)13:56
rlameirook13:56
rlameirolike enable backports ppa13:56
rlameiroand another with enable ubuntustudio universe team ppa13:56
rlameiroor something13:56
ScottLhttps://wiki.ubuntu.com/UbuntuStudio/Meetings/2010May1613:57
ScottL/etc/security/limits.d/audio.conf 13:57
ScottLthe link provides alink to mailing list discribing it, but you seee the file also below my link13:57
rlameiroyes that, misspelled it13:57
rlameiroi just dont get that stuff of not adding the user to the @audio group when using PA13:58
ScottLi have misgivings about enabling ppa's that we do not have control over, we cannot guarentee their quality and we would receive the bugs for them13:58
rlameiroScottL: me too13:59
rlameiroI was thinking on separate ppa, just that13:59
rlameirobut all controlled by the team, also the list would need to be approved by the team, but maybe its better to have just one ppa with everything14:00
ScottLthe way i understood dtchen's explanation, is that the active user (the one at the keyboard) has rights to the /dev/snd/* which means you have access to the @audio group14:01
rlameiroi wanted to put the kernel in a diferent one, because, that way, we could warn users about the rt kernel implications14:01
rlameiroeven with jack?14:01
ScottLyou are not explicitly in the group but because you are the person logged on you access to that groups rights14:01
rlameirohumm14:02
rlameirook14:02
ScottLthat's a good idea about the kernel14:02
rlameiroScottL: http://kxstudio.sourceforge.net/index.php?option=com_content&view=article&id=44&Itemid=1214:02
rlameiroScottL: the idea was like this, the user selects the checkbox for the kernels PPA and a Warnig box appears14:03
ScottLaccording to dtchen, this should work, even with jack...of course there may be unforseen situations14:03
rlameirosaying that you may have problems with the propietary kernels etc14:03
rlameirothe things abogani explained on the mailing list14:03
ScottLproprietary drivers?  instead of proprietary kernels?14:03
rlameiroScottL: when you instal the rt kernel, ussually users have some problems with the nvida and ATI propietary drivers14:04
rlameirothis is a defense for us, or elese we would be swamped with bugs and complains14:04
ScottLi meant that you said "saying that you may have problems with the propietary kernels etc"   but i realize you meant drivers not kernels14:05
rlameiroif they want rt kernels, they should know thew caveeats14:05
ScottLexactly14:05
rlameirooops, sorry fail14:05
rlameiro:D14:05
ScottLlol14:05
ScottLi'm hoping this is transitory...if preempt gets in archives and can be shipped by default, many may not want to switch because performance is acceptable14:06
ScottLtransitory = people wanting -rt kernels14:06
rlameiroyeap, but there are still situations whwere rt kernels is better, for instance with laptops14:06
ScottLmy understanding is that we cannot make an -rt kernel without Igno's patch, and that leaves us in a difficult position sometimes14:07
rlameiroas the mobile processor have some weard differences, with the rt you can push it to the limit14:07
rlameiroScottL: no, if we tell the user :D14:07
rlameirowe are not the devs, we just try to get together the best software14:08
rlameirowe are not kernel devs14:08
ScottLrlameiro, i fear that most users think we work in a fantasy land and we have all the knowledge, resources, and time to make things "right" and that somehow we just choose to screw things up for select peoples14:08
rlameirothe problem is that most people think that ubuntustudio makes its own kernel, and all software14:08
ScottLprecisely14:08
ScottL"omnipotent" gods thats fail14:09
rlameiroso, ubuntustudio controls could be a way to WAR users about the caveats14:09
rlameiroand also ask for help14:09
rlameiroit can be naggy, but yes it is needed14:09
ScottLa dialogue box would be nice to tell them of the dangers and a link to an informative website would be a really good thing IMO14:10
rlameiroprecisely my idea14:10
rlameirowith a BIG warning triangle14:10
ScottLenglish is my native tongue and my last sentence was pretty horrible :P14:10
rlameirowe also should push translation14:10
ScottLmaybe I should have an editor and a translator :P14:11
rlameirolol14:11
rlameiroi got you :D14:11
ScottLthat's why i worry about abogani and the way he gets down on himself about his english14:11
ScottLi really do think he types it better than a lot of native english speakers14:12
rlameiroScottL: that is normal with us, because our native language isn't english, but we do spend most of our time reading stuff in english14:13
ScottLonce we get an ISO build that works i should be updating the seeds to include the new lv2 stuff and hopefully moderating the menu for them as well14:13
rlameiroso sometimes qe feel that we should do better at it14:13
ScottLeven if something is spoken (or written) not quite correctly but the intent is understood, then that is a win14:13
ScottLbecause sometimes even for those born in US and speak english natively, it is difficult to understand some spoken dialogue because of slang and poor word choices14:14
ScottLit's not what you would call "proper english" but it gets understood the same more or less14:14
rlameiroyea, true, sometimes i cant get some english expressions ou there :D14:15
ScottLoh, and when we get an ISO build that installs I really want to test JACK/Pulse Audio together and see what's different14:16
rlameiroScottL: do you know another ubuntu channel for developres, besides #ubuntu-dev?14:16
ScottLi do not...i wasn't even directly aware of #ubuntu-dev until now14:16
rlameiroScottL: about that,, http://kxstudio.sourceforge.net/index.php?option=com_content&view=article&id=44&Itemid=1214:16
rlameiroi think we could implement it on controls14:17
rlameiroas enable it or disable it14:17
ScottLwhich part?14:17
rlameiroalso warning that when jack is stoped pulse crash :D14:17
rlameirothe pulse/jack14:17
ScottLyou mean the bridge?14:18
rlameiroso you can get pulse audio into jack and vice versa14:18
rlameiroyeap14:18
ScottLi agree, but probably not this cycle14:18
ScottLi am expected to have some troubles still with maverick in terms of getting jack and pulse enabled and working in a stable manner vai dbus14:19
rlameiromaybe14:19
ScottLnot that i have great or expansive knowledge of the subject14:19
ScottLbut when we change infrastructure like this it seems to have "unforeseen consequences" which take a cycle to rectify14:20
rlameiroit depends on how can i make it or not14:20
rlameiroi have some spare time now14:20
ScottLi'm not talking about -controls14:20
rlameiroi am starting to reading stuff about python-apt etc14:20
rlameiroi need to make a mock-up and implement it14:20
ScottLbut first can we fix the rtprio and memlock situation, that is clearly defined and pretty critical14:21
rlameiroalso i will meet in person RNCBC in 15 days so i can ask him directly about rtirq14:21
ScottLhey, that's pretty cool!14:22
rlameiroScottL: AFAIK the issue is easy to fix, just changing the file path in the python script14:22
rlameirothat is the first thing i will do14:22
rlameirothat and remove nice14:22
ScottLrlameiro,  i would even suggest that we fix that first, test it, and then push the changes now to improve usability14:23
rlameiroScottL: should i develop it on the alpha?14:23
ScottLand then worry about other improvements14:23
rlameiroScottL: for lucid?14:23
ScottLyou could develop on lucid i think without any problems14:23
rlameiroi dont know if i can14:24
ScottLyou don't know if you can update -controls on lucid?14:24
rlameirothe problems with rtprio came because the change was made between karmic and lucid14:24
rlameiroso never was detected fixed in time14:24
ScottLif you can update -controls on a lucid install, it should be able to be merged into the code for maverick and then released14:25
rlameirobut, of course it can be developd in lucid, but needs testing on MM14:25
rlameiroyeap14:25
ScottLthen an SRU can be filed to get it released into lucid still during maverick's development14:25
rlameirothats tha14:25
rlameirothats it, so we can push it to lucid also :D14:26
rlameirothats what i wanted to know :D14:26
ScottLoh good :)14:26
ScottLhopefully you can make the rtprio, memlock, and nice changes, they get reviewed and testing in lucid, and then an MM ISO is available to test as well14:27
ScottLthen we can merge the code for MM14:28
ScottLafk a few minutes14:28
rlameirook14:28
rlameiroScottL: ok, i just removed nice from UScontrols14:58
ScottLrlameiro, sorry had to finish a blog post before the family woke up    http://dullass.blogspot.com/14:58
ScottLgood :)14:59
rlameiroi changed the file were writ the memlock, but there is already info on mine14:59
ScottLrlameiro, did you just remove the code or did you comment them out?  i think i would advocate commenting them out for the time being15:00
rlameirothe instllation process of jack already does some stuff, so i need to make something to check for the actuall state of the file, or it will addin one more line15:00
rlameiroi removed it15:00
ScottLyes, that is true15:00
rlameirobut its not important15:00
rlameiroits only mu branch15:00
ScottLwe have the old code still so it's not that big of a concern15:00
rlameiroafter i make a diff and commnet out the original15:00
ScottLrlameiro, it's time for me to go, i have to wake up son and get him ready for us to go out into town15:01
rlameiroalso, there is a commented line with a nice setting /etc/security/limits.d/audio.conf15:01
rlameirogo go 15:02
ScottLrlameiro,  ciao, i always enjoy talking with you15:02
rlameirohave a good time15:02
rlameiromee to :D15:02
rlameirohi again20:13
rlameiroScottL: i just reconfigured jack, using $dpkg-reconfigure -p high jackd and it added memlock unlimited, and rtprio 9920:14
rlameiroboth to @audio20:14
rlameiroso, i dont know, why do need ubuntustudio controls for the moment...20:14
rlameirobesides enabling firewire20:15
rlameirocrimsun_: I would love if you could clarify some things avout the audio setup at the moment20:16
rlameiroas i want to push ASAP a modified version of ubuntustudio controls for the 10.04.120:17

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!