Hi Ross,
[ It does my heart good that people read the spec so closely that they
find the bugs in it!]
As John points out, 5.8.2.2 and 5.8.6.2 are in conflict -- the latter
implies that a standard name may be an instrument name (it couldn't
be if it weren't allowed to be a symbol), while the former implies
that it cannot. I prefer, exactly for reasons of bitstream parsing,
the 5.8.2.2 interpretation. This means that 5.8.6.2 is incomplete,
and "standard name" should be listed in the itemization of things
an instrument name shall not be.
Best,
-- Eric
-----Original Message-----
From: Ross Bencina <rossb@audiomulch.com>
To: Saol-dev <saol-dev@media.mit.edu>
Date: Sunday, November 21, 1999 8:12 AM
Subject: symbol vs. ident. vs. token
>Hi, I'm looking for clarification on the following
>
>Since it is legal for a standard name to be used as an instrument name
>(5.8.6.2) would it be fair to surmise that:
>
>A. 5.8.6.2 is in error, a standard name cannot be used as an instrument
name
>
>OR
>
>B. the name would be tokenised as the token corresponding to the token for
>the corresponding standard name, not as a symbol token.
>
>OR
>
>C. all identifiers used to name instruments, opcodes, tables and variables
>are always tokenised as symbols.
>
>Best,
>
>Ross.
>.................................
>http://www.audiomulch.com/~rossb/
>
>
>
This archive was generated by hypermail 2b29 : Wed May 10 2000 - 12:15:47 EDT