> Sounds like you are still stuck in the rut that "The purpose of
> commercial companies should be to produce high quality
> software"; when in fact their purpose is to maximise investors
> return on investment (which may or may not involve the production
> of quality software).

I agree, the purpose of any commercial company is to make money for its
shareholders.

However, I admit that I still have some of that idealism left in me - but
not nearly as much as most of my colleagues.

Why do I have some of that idealism left?  Because, at least in the case of
mathematical software, the barrier to entry is huge: it takes ~100
person-years to develop a halfway decent piece of mathematical software.  So
'competition' is thin, essentially restricted to existing players only.  And
while 100 person-years is possible to find, especially for operating
systems, browsers, etc, it is much harder to find 100 person-years from
people who combined mathematicians and computer scientists.  [The reason to
go into this is that I am sure that there are lots of other market niches
with similar forces at work].

So customers only have restricted choice, they pick the least-bad choice.
And because of strong lock-in effects, once a choice is made, it sticks for
years (for most people 10+ years).

The question becomes: other than through good will, how do quality
improvements happen in such an environment?  How can customers 'force'
quality improvements?

> I lay a large chunk of the blame at the feet of CS educators who
> would clearly fail an introductory course in business administration.
> Perhaps I was wrong to suggest that software developers should
> be jailed.  Jailing a few university lecturers for teaching nonsense
> might be a more efficient long term solution.

But this is true in many disciplines, not just CS.

However this is why I joined a 'software engineering' department, in a
Faculty of Engineering, with an eye towards getting my PEng.  After which,
if I were to certify that some software does what it is supposed to do, and
it does not, I am liable.  I could be fine, lose my PEng status, or be sued.
I view this as a good thing.  I hope that eventually, through changes in
law, certain software (like certain buildings, electronic products, chemical
plants, etc) will need certification be a licensed engineer, and will not be
allowed to be used/sold/etc otherwise.

Jacques


----------------------------------------------------------------------
PPIG Discuss List ([email protected])
Discuss admin: http://limitlessmail.net/mailman/listinfo/discuss
Announce admin: http://limitlessmail.net/mailman/listinfo/announce
PPIG Discuss archive: http://www.mail-archive.com/discuss%40ppig.org/

Reply via email to