On Wed, 2012-01-25 at 17:52 -0800, Len Ovens wrote: > On Tue, January 24, 2012 8:37 pm, David Henningsson wrote: > > > One of the advantages of Linux is that the line between "single user > > system" and "multi user system" is blurry. E g, I could run a web server > > or other service on the same machine as I use for audio production. This > > is great, it saves hardware. If a malicious user breaks in to the web > > server, I don't want him to be able to use RT prio to lock down the > > entire machine. > > I would think that running a server of any kind (other than something like > netjack) on an audio machine would remove the need for RT anything... In > the radio stations I have worked in, when the mic goes live, lots of > things get turned off. From monitor audio to telephones... some even lock > the door.... though most just light an on air lamp. Recording studios do > the same. I am thinking a DAW should be able to do the same with > extraneous software. Almost like having a recording runlevel. I really > would not want someone to log in remotely while I was recording the 5th > take of something and find out they did something that ruined the best > take of the day. Yet, there are times it is handy to sftp in to look at > files (from behind the the FW). Turning the network off might help, unless > netjack (or other similar application) is using it. > > I think most people want to also be able to use their computer as a > general computer sometimes too. However, the audio/video use is the > priority use for the ubuntustudio distro. One of the reasons US moved to > xfce was something with less overhead, but there is still two screens > worth of process running just for the session (ps x). Top shows 125 > process running overall. Top says 80% mem use while taskmanager says 26% > (which is still a lot). I have 1g ram which is not a lot by todays > standards but I think we could use less anyway. It would be nice if > workflows could turn off some of the extra junk. > > Hmm, just thinking out loud.
An analogy If you lock the door, you could run into trouble, if the studio catches fire and btw. it's a safe bet that somebody will knock at the door. The only safe solution is to ban people from the radio station who have no business in the radio station and ask all people who are needed for the production, to come to work. Accidents will happen easily to everybody of us. For Linux you should ban every piece of software, resp. all settings that could cause issues and start all software with correct settings that are needed. I suspect OOTB no distro will be perfect, but a lot of optimization can be done by scripts to start a session. On another list there's somebody who call people idiots that are unable to follow his work flow, e.g. if people won't add a user to the group audio manually, OTOH he blames a software for killing his speakers, just because he was careless. Taking care does mean to expect something that shouldn't happen. In Linuxland we have a situation where a group of experts will blame users if they just want to use a tool. The approach of Linux is that we need to be toolmakers too. Btw. a lot of stuff that should make live for users easier, made the usage harder instead. A balancing act regarding to the realtime group issue isn't worse the effort. Newbies should use a pre-build audio distro where the first user already is member of the group audio and for a radio station where the DAW has got several user accounts, there should be an audio engineer being able to be admin. 2 Cents, Ralf -- Ubuntu-Studio-devel mailing list Ubuntu-Studio-devel@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-studio-devel
