On Wed, 21 May 2003, Martin Stecher wrote:
How about just calling the command with its full name and then only
refering to the abbreviation as the code in the technical
description.
Example
7.1 The connection-start command
command code: CS
anonymous parameters: none
...
Would you use "CS" or "connection-start" in the examples? Full names
make examples much more readable and make readers feel like they
immediately know what is going on...
Looking from the other point of view, what are the technical reasons
for NOT allowing full names on the wire? We can probably convince most
developers to use abbreviations while supporting full versions. The
amount of extra code or memory needed to support both versions is
minimal. Are we concerned with extra bandwidth and CPU cycles spent on
handling full versions if some broken implementation sends them?
Or are we primarily concerned that some implementations will end up
recognizing one version (full or abbreviated) and not the other?
Thanks,
Alex.