Re: comedi as a POSIX standard

Herman Bruyninckx wrote:

> I was in favor of the POSIX idea initially, but I changed my mind:
> POSIX is IEEE-controlled, and these standards are not free (as in
> beer).

(As long as I'm taking the advocacy position....)

The IEEE doesn't control the content of standards. The working group 
that creates the standard controls the content, and the working groups 
are typically open to anyone who wants to contribute. If we convened a 
Comedi working group, we could certainly make it an open process.

It's true that IEEE doesn't publish its standards freely (as in beer), 
but in my opinion, the value added by professional editing and 
production more than makes up for the cost. The POSIX documents are 
*very* good.  As interesting and exciting as the open source movement 
is, it has not up to this point consistently produced good 
documentation. It might be more accurate to say it has not consistently 
produced documentation :-(.

For standards that really matter to application developers, it's rarely 
necessary to buy the standard--the relevant details get incorporated 
into vendor documentation. For example, I don't own a copy of POSIX.1 
recent enough to include pthreads, but I can code to the pthreads spec 
by referring to any number of free sources (e.g., 
http://publib.boulder.ibm.com/iseries/v5r1/ic2924/index.htm?info/apis/rzah4mst.htm). 
It would be interesting to know how copyright issues are handled in 
vendor documentation.

Vendors who want to claim compliance to the standard will buy copies of 
it, but the expense is in the noise for anyone doing commercial software 
development.

Steve

Received on 2003-04-13Z20:15:43