[OSC_dev] 128 bit field type tag?

[Ross Bencina]

Hi All,

Is there any existing usage or defacto standard for 128 bit OSC message
parameters?

I need to transmit IPv6 addresses and it would be useful. I could use a
blob.

Alternatively, maybe there's a good interoperability argument for using
blobs.

Any thoughts on this?

Thanks,

Ross.

P.S. what would be really cool for "2.0" is if a bunch of wasted bits in
the type tag were used to encode short binary field lengths. For
example: if the high bit was set the low 7 bits of the type tags could
encode field lengths from 0 to 127 bytes (or 1 to 128).

[Adrian Freed]

We are thinking about this issue for things like PTP time.
I know this doesn't answer your question but have you considered the human efficiency advantage of coding it as a string?
e.g.

2001:0db8:85a3:0000:0000:8a2e:0370:7334

Post by Ross Bencina
Hi All,
Is there any existing usage or defacto standard for 128 bit OSC message
parameters?
I need to transmit IPv6 addresses and it would be useful. I could use a
blob.
Alternatively, maybe there's a good interoperability argument for using
blobs.

Yes, it comes down to what one might interpret a type to mean.
We didn't write it this way originally but the core OSC types are the most common
"primitive data types" http://en.wikipedia.org/wiki/Primitive_data_type

Post by Ross Bencina
Any thoughts on this?
Thanks,
Ross.
P.S. what would be really cool for "2.0" is if a bunch of wasted bits in
the type tag were used to encode short binary field lengths. For
example: if the high bit was set the low 7 bits of the type tags could
encode field lengths from 0 to 127 bytes (or 1 to 128).

Yes, that is ingenious, but our suggestion for 2.0 is that it be a community development with a new encoding entirely
which would presumably not require such a thing..

[Ross Bencina]

Post by Adrian Freed
I know this doesn't answer your question but have you considered the human efficiency advantage of coding it as a string?
e.g.
2001:0db8:85a3:0000:0000:8a2e:0370:7334

I have considered it. And in combination with the standard rules for
reducing 0 fields it is possibly an option.

My use-case involves sending a list of candidate IP addresses for all
members of a group (in general hosts have multiple IPv6 addresses). So I
would like to keep the message size to a minimum.

Ross.