Quality programming in c -- a ra

[jim at ISM780B.UUCP]

> In other words, if those who write Unix utilities, surely the
> top level of C hackers, routinely write bad code --

There is no evidence whatsoever for this; in fact, the evidence that these
people are *not* top level C *programmers*, namely the poor quality of their
code, is overwhelming.  There are many very good C programmers around,
but apparently very few of them are writing code at Bell Labs that is
getting released.  I sure wouldn't waste good programmers on that.
I find programs like tbl or cc to be pure trash from the point of view of
software engineering.  I know the people wrote those are very bright, good
conceptual people, but they have been part of the Bell culture for a long
time, and I doubt that Bell has done much to challenge their habits or train
them in software engineering techniques, or help them develop their *coding*
skills.  Of course, those programs and many of the other utilities were
written a long time ago, and the people who wrote them may be much better
now.  In fact, I would expect people writing utilities to be junior and
inexperienced; experienced programmers go on to do more sophisticated system
design.

It *is* encouraging that C++ could be developed at the labs, and I would hope
it and its author and others with some concept of modern software engineering
will have an effect on what comes out of there, although it will probably
take a long time.

-- Jim Balter, INTERACTIVE Systems (ima!jim)