[OSC_dev] cnmat max objects for osc

[adrian at cnmat.berkeley.edu]

> Hi,
>
> Sorry for the late reply.
>
> On Dec 21, 2008, at 6:12 PM, Andy W. Schmeder wrote:
>> However, the "built-in" handling of OSC in MaxMSP is woefully
>> incomplete. [...] It does not deal with OSC bundles. [...] most of
>> the rest of the world doesn't (yet) realize the value of time-
>> semantics so they don't care about bundles.
>
> I assume the problem sits a bit deeper: Max itself has no real notion
> of explicit timing and has also very limited support for composite
> data structures. Time is pretty much either "now" or less relevant (I
> am exaggerating, but this is the tendency). I find this fact
> particularly strange for a realtime music system, but it appears very
> many people are quite happy with this limitation. Other systems with
> an OSC interface do not have this limitation, e.g., scheduling the SC
> server works nice with timetags, and so SC does of course support them.
>
> Once you have a scheduler in Max which you can tell to process things
> at a certain time in future I guess you will also soon have support
> for timetags and bundles for OSC :)

Andy has written some Max objects that do this kind of scheduling for Max
so we can now say that
we have at least one complete implementation of OSC. It is described in
our two uOSC papers.