[LAD] OSC: Divide & Conquer, or build a Stronghold?

Paul Davis paul at linuxaudiosystems.com
Fri Dec 18 00:08:21 UTC 2009


On Thu, Dec 17, 2009 at 6:41 PM, Harry Van Haaren <harryhaaren at gmail.com> wrote:
> This would mean that any program that wants to find out which Jack B:B:T or
> Frame were on
> or if there's been XRuns, wouldnt have to be a JACK client, because its been
> abstracted to
> the "Master OSC Host".

i don't many comments on this overall idea, but i do find the notion
that a program wants to know about the internal state of JACK without
being a JACK client, and yet, unlike a simple shell script, is willing
to participate in a 2 way IPC protocol to find out a little .... odd.

also, please keep in mind that the basic semantics of OSC make
query:response interactions really rather tricky to implement in a
robust way. meaning that they will work in 98% of cases, and the other
2% will just either unfathomable and/or unfixable.



More information about the Linux-audio-dev mailing list