[H-GEN] Free Trade (was Re: Selections from RISKS DIGEST 19.57)

[Craig Eldershaw]

>> In the case of mission critical systems, we desire more than a code
>> Monkey or clever hacker. We desire a professional.
>
>...
>In my experience a major cause of the shoddy work stems from buyers
>not able to properly specify their needs to the vendor.

OK, I'll jump in here.  I'd argue that since poor user-based specs are
such a common problem (why else is the industry full of jokes and horror
stories on this), then that should be allowed for.  I don't know much
about car-engines, but I expect the mechanic to listen to what I say
and come up with some reasonable specs (possibly after checking back
with me, and possibly even cross-examining me).

Knowing there is this common problem, I beleive the solution has to be
addressed by us (ie. people in the computer industry).  Sure, we can
curse their ignorance (as I'm sure the car mechanics do about me over
their tea break) and have to spend extra time in ensuring that we fully
understand their needs and that they fully understand what we can and
can't do; but in the end, any shoddy work has been done by us, and we
should be to blame.

Now there is the related issue of "cheapo jobs" or "rush jobs".  There
are probably some extreme situations where these can't be avoided, but
in general we should refuse to take on a job where we have insufficient
time/money to fully specify, cleanly design and test our product.  The
user shouldn't really be given the option of "we'll strip the testing
phase down and knock off 30%", since they obviously don't appreciate the
reasons and benefits behind it.  We simply state that the job asked for
cannot be done for that sum/in that time.  Suggest a more restricted
product *could* be produced with that money.

Cheers,
	Craig.


----------------------- HUMBUG General List --------------------------------
echo "unsubscribe general" | mail majordomo at humbug.org.au # To Unsubscribe