🚀 go-pugleaf

Submitted-by: mroussel@alchemy.chem.utoronto.ca (Marc Roussel)

In article <193se9INN4h7@ftp.UU.NET> brnstnd@KRAMDEN.ACF.NYU.EDU
(D. J. Bernstein) writes:
>Ah, but you never _have_ to write a standard.
>
>You see that systems vary widely in, e.g., the output format of ``who'',
>and even the type of information stored in /etc/utmp (or whatever file
>``who'' reads). Does this mean you have to apply your imagination,
>specify what the market hasn't specified, merge what the market hasn't
>merged? No.  [...]

     This assumes that "the market" purchases systems based on the
format of the output of who.  This is not the case, but that is not to
say that purchasers of Unix systems would not like their machines to all
behave consistently.

>You shouldn't standardize ``who'' at all. Wait for market convergence.

     This will never happen.  Vendors have no incentive to standardize
who (among other things).  The only real leverage users can exert to obtain
properly standardized environments is through the standards process.
The power of the purse is essentially useless when it comes to little
things like the output of who and the command line switches of cc since
no one (to my knowledge) makes purchasing decisions based on these
criteria.  This sort of thing turns into just another annoyance in a
heterogeneous environment; it is however a pointless annoyance which standards
can essentially do away with.

				Marc R. Roussel
                                mroussel@alchemy.chem.utoronto.ca


Volume-Number: Volume 29, Number 39