🚀 go-pugleaf

>>Engineering is moving towards requiring practicing engineers to be
>>registered witht the state(s) that they practice in (though not
>>required today) It would be analagous (sp?) to the bar exam for
>>lawyers or the boards for doctors.  This is a way for the state to
>>help assure that the people practicing are competent and not outside
>>of there realm of competency.  This type of exam is a _professional_
>>certification.  Programmers are a _professional_ group of people.

I believe that the main difference here is that the average joe/jane
needs some kind of protection from doctors and lawyers.  With doctors
and lawyers you actually see them for a consultation.  You can't just
go down to the nearest Organs 'R Us and pick out a new heart or
go to ContractWarehouse and get a contract off the shelf.  Most of
the software that jane/joe buy now is off the shelf stuff and gets
plenty of reviews and word of mouth to let people get a good idea
of what the programs do.  Most people don't need custom software,
analagous to getting a custom prescription from the doctor or a
custom contract from a lawyer.  The ones (usually business's) that do
need custom software should take the time and energy to talk to a
few consultants and also business associates before they go out and
spend a huge chunk of change.

> Which programmers are you talking about?  Some are `professional'.
> They do valuable work (i.e. someone values it enough to pay them) but
> these people are not going to write the next VisiCalc or Emacs.

> Other programmers are less `professional'.  They dress and behave less
> conservatively.  They are less concerned with coding standards and
> documentation standards than with getting something interesting
> working.

I hope that this wasn't meant the way it reads, but if so:
I really resent the implication that I am not a 'professional' by some
people just because I want to be comfortable while I work.  Or that
because of this I am sploppy in my coding practices or documentation
practices (ok, I hate to document, but I do - self documenting code, right?).
The companies that I have worked for have NEVER had a coding standard in
place - unless you consider the use of a particular language a coding standard.
These same companies have been run by managers/directors/vps that wear
a suit, but have never seen the need to setup a coding or documentation
standard yet I am the non-'professional'.  We (non-'professionals') have had
to push in order to get some of these standards put in place.  The
'professionals' have always seemed to want to just "get it out the door".
BTW, even though I am not a 'professional' (even have a pony tail, but
no ear ring - so I guess I am not THAT wild) I have been told many times
that my manager is very, very happy with my work (as have all the other
non-'professionals' in my group).

>>There is software that is very difficult and has the potential effect
>>of taking several lives.  Mentioned has been medical software and
>>software that is life-critical.  What about cars?  The wrongly
>>programmed software can have detrimental effects, and the amount of
>>hardware and software is not going down.  What about industrial
>>situations?  I have heard horror stories about an industrial
>>controller that went haywire because the 'professional' was not able
>>to correctly write interrupt driven software; compound that problem
>>by writing it in assembly. As a final example, consider aviation
>>electronics.  Very simply put, if it fails, you die.  If there were
>>an error in the software of the primary computer, so would it be in
>>the backups.

Do you really believe that taking an exam and becoming a 'professional'
would make software any more reliable?  Now, just because a 'professional',
licensed software stud wrote the program it want break, huh?  The only
way to increase the reliability of software is to test the software in
anyway that can be thought of - ever hear of alpha-testing, how about
beta-testing?  The programs that my group is working on now are 15500 and 6500
lines of source/comments/blank lines/declarations.  Before any release
we go through a very thorough testing procedure that takes 2-3 weeks.
My manager insists.  She is the only one that I have known that will allow
this thorough testing.  Because of this we have only had one minor bug
in a release and it was found by us in subsequent testing after the release.
During our testing we usually find several MAJOR faults with the other
software and especially systems that we interact with.  In fact, one
of the outside systems was replaced with an in house system because the
company doing the work never seemed to get around to testing even basic
functions (like will it boot).

Was the industrial control software even tested?  How thoroughly?

As far as primary and backup systems go, the only safe way to do this is
the way the only software controlled nuclear power plant (in Canada I think)
did this.  They had two independent systems groups design/write/test
two systems that did the same thing.  This narrows the likelyhood that
the same bug in the same part of a program will occur.  I think that
it took them 3 years or so to finish.  After completion the company said
they were very pleased with the finished results, but if they had to
do it over again they would just do it the standard hardware way.

Most fault-tolerant systems are done that way in case the hardware goes
out, not the software.  After all software doesn't wear out.  If it is
well tested then it should be reliable.

john gay.                      jgay@digi.lonestar.org

"A third theory [on the creation of the Earth] is the one offered by my old
 theological buddy Bobby Joe Dearing, who speculates that the world was
 originally created as a science fair project.  "The problem," says Bobby
 Joe, "is that God made it the night before he had to turn it in, and he's
 probably going to get a 'D' on it."
  - John Anders, Dallas Morning News Columnist