From mboxrd@z Thu Jan 1 00:00:00 1970 Message-Id: <200501311056.j0VAuQq05503@zamenhof.cs.utwente.nl> To: Russ Cox , Fans of the OS Plan 9 from Bell Labs <9fans@cse.psu.edu> Subject: Re: [9fans] playing music (was RSA) In-reply-to: Your message of "Mon, 31 Jan 2005 02:03:33 -0500." References: <200411062237.iA6MbCZ16674@zamenhof.cs.utwente.nl> From: Axel Belinfante Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 31 Jan 2005 11:56:25 +0100 Cc: Topicbox-Message-UUID: 391b5df4-eace-11e9-9e20-41e7f4b1d025 > The audio back end for volume and ajuke is system-specific, > and only Linux is implemented. If you add support for another > system, please mail it to me so I can include it. maybe the following is a weird idea, but anyway: I do my daily work on a sun, where I use drawterm to a connect to a plan 9 cpu server where I run sape's juke player. On the sun I have named pipe's /tmp/dev/audio and /tmp/dev/volume and two small programs, one to read audio bytes from /tmp/dev/audio and shuttle them to the real sun audio device, and another one to implement enough of the plan 9 /dev/volume interface to keep sape's juke happy. (details are at http://plan9.cs.utwente.nl/audio-sun-drawterm/) In drawterm I 'bind -b /mnt/term/tmp/dev /dev' and then run sape's stuff. Would such a scheme fit (could such a scheme be made to fit) with plan9port, such that all plan 9 (-port) programs assume a plan 9 style audio interface, and the system specific backends provide that? I guess that would help to keep the local system audio specifics out of the plan9(port) code... Regards, Axel.