sw dev

[Jean-David Beyer]

On 09/18/2016 12:18 AM, ed65love wrote:

>>
>> Now the company I worked for no longer exists, though another company
>> uses its name and trademarks.
> 
> And I bet they all got paid well, too.

The top management got bonuses. The rest of us got no raise at all my
last year there.
> 
> Ah, the old maxim: folks either manage what they don't understand, or 
> understand what they don't manage :(
>>
>> The worst slogan was 10x quality improvement every 5 years. I asked how
>> quality would be defined and measured. They had no clue. And, at least
>> in software, they had no process for developing it; every thing they did
>> was started over fresh, so none of the classical quality control
>> techniques could be used and stepwise refinement was impossible.
> 
> It's funny how people think computing is a science :)
> 
But it could become engineering and a science (two different disciplines).

What is very sad is that it could be mostly a science. Victor Vyssotsky,
who was the manager of the Safeguard project at our company., wrote a
great paper, "What I Want Computer Science To Tell Me" that outlined
what he found lacking in computer science, as taught, as practiced, and
what he needed from it and was not getting. Now Safeguard was a major
anti-ballistic missile system with radars and stuff as input, and
anti-aircraft missiles and the Strategic Air Command as output devices
(among other things). You could not just have a room full of hobbyist
boffins improvising that. One of the problems was that the software was
just too complex, so it could not be validated to be correct.

This is not his paper, but he was one of the authors; He was no longer
at the company I worked for when he contributed to this.

https://books.google.com/books?id=55G1I5RYg7oC&pg=PT3&lpg=PT3&dq=Victor+Vyssotsky+computer+science&source=bl&ots=wyeCzkFe-n&sig=-U52E0tqcQtjyhq8GGX8j5zxO_w&hl=en&sa=X&ved=0ahUKEwiTueDn5pjPAhUk_4MKHUJQBpIQ6AEIRzAJ#v=onepage&q=Victor%20Vyssotsky%20computer%20science&f=false

Dave Parnas wrote an early paper, for example, on how to write better
software.

https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=8&cad=rja&uact=8&ved=0ahUKEwjareGt6ZjPAhWr1IMKHQhICD0QFghMMAc&url=http%3A%2F%2Frepository.cmu.edu%2Fcgi%2Fviewcontent.cgi%3Farticle%3D2979%26context%3Dcompsci&usg=AFQjCNEizGxh1eAtoOSIZ_PuJE61Ik2pGg&sig2=D-65HVaiNcOmjGzPTgVBzg&bvm=bv.133178914,d.amc

The sad thing about this was even though it was written in 1971, we were
not using it at the end of 1989. So our software, written by very bright
people, was disorganized and incomprehensible even when it worked; it
could not be maintained.

-- 
  .~.  Jean-David Beyer          Registered Linux User 85642.
  /V\  PGP-Key:166D840A 0C610C8B Registered Machine  1935521.
 /( )\ Shrewsbury, New Jersey    http://linuxcounter.net
 ^^-^^ 07:20:01 up 25 days, 5:29, 2 users, load average: 4.65, 4.49, 4.29