Eachus@MIT-MULTICS.ARPA (12/08/85)
Now that the problems caused by moving Info-Ada have been fixed, I
would like to say some things in defense of Ed's efforts. The need for
professional software engineers is here. There are a lot of areas where
poor software could be much more harmful than the recent problem with
this mailing list. Responsible people are going to insist that this
work be done, and supervised, by professionals. There are four ways to
do this:
1. Use professionals from other engineering fields.
2. Impose existing standards on Software Engineers.
3. Develop new professional standards for Software Engineering.
4. Ignore the problem and hope it goes away.
For software engineers using Ada solutions one and two are
unacceptable, and in many areas (flight software, banking, SDI, etc.)
the fourth alternative is too horrible to contemplate. This leaves a
choice between developing standards ourselves, and having others impose
them. I hope that the standards that do evolve are similar to those for
other professional engineers, or those for CPA's and actuaries, where
only those who want or need to be recognized as professionals are
required to comply with the standards and ethics of the profession.
Robert I. Eachusdgary@ecsvax.UUCP (D Gary Grady) (12/13/85)
In article <851208185443.106615@MIT-MULTICS.ARPA> Eachus@MIT-MULTICS.ARPA writes: > >. . . The need for >professional software engineers is here. There are a lot of areas where >poor software could be much more harmful than the recent problem with >this mailing list. Responsible people are going to insist that this >work be done, and supervised, by professionals. You miss the point. The argument is not against competence or professionalism, but against the usual idiot, ineffective efforts aimed at ensuring such. I don't doubt enthusiasts of multiple choice certifying exams, degree requirements, and continuing education programs are well-intentioned; I just think they're nuts. I don't object to the end but to the means. Is anyone really stupid enough to think that sitting in a classroom for x number of hours is necessary OR sufficient to learn something? That two people with the same amount of education and experience are necessarily even similar in competance? The only way to determine if someone is a competent, conscientious programmer is to look at the results of their work, WHICH IS PRECISELY WHAT COMPETENT EMPLOYERS DO RIGHT NOW. Replacing this with blind trust in some sort of certification program is going to make things much worse, not better! There is one case where certification might make sense: Independent consultants [disclaimer: I am one]. But I would hope such certification would be based on demonstrated ability, not on a silly mix of education and experience requirements and a multiple choice test (as the worthless CDP is set up, for example). This is not sour grapes: I have been programming since 1970 and I usually do in the 99th percentile on standardized tests. I just recognise how utterly irrelevant that is to any real-world performance, with the exception of quiz shows. (Maybe we should certify contestants? :-) -- D Gary Grady Duke U Comp Center, Durham, NC 27706 (919) 684-3695 USENET: {seismo,decvax,ihnp4,akgua,etc.}!mcnc!ecsvax!dgary
friesen@psivax.UUCP (Stanley Friesen) (12/17/85)
In article <910@ecsvax.UUCP> dgary@ecsvax.UUCP (D Gary Grady) writes: > >You miss the point. The argument is not against competence or >professionalism, but against the usual idiot, ineffective efforts aimed >at ensuring such... I don't object to the >end but to the means. > >Is anyone really stupid enough to think that sitting in a classroom for >x number of hours is necessary OR sufficient to learn something? That >two people with the same amount of education and experience are >necessarily even similar in competance? The only way to determine if >someone is a competent, conscientious programmer is to look at the >results of their work,... > An example. I have gotten high ratings on technical performance and programming skill from every employer I have ever worked for. I learned programming, and every language I use regularly, entirely on my own, from reading books and "playing" on computers! I got my first job from a small firm that was willing to gamble on my being able to handle the job, I was soon thier top programmer. Formal training and experience are, at least in my case, almost meaningless. (I tend to use the same method of learning in all areas of intertest to me.) -- Sarima (Stanley Friesen) UUCP: {ttidca|ihnp4|sdcrdcf|quad1|nrcvax|bellcore|logico}!psivax!friesen ARPA: ttidca!psivax!friesen@rand-unix.arpa
rcd@opus.UUCP (Dick Dunn) (12/17/85)
> ...I would like to say some things in defense of Ed's efforts. The need for > professional software engineers is here. There are a lot of areas where > poor software could be much more harmful than the recent problem with > this mailing list. Responsible people are going to insist that this > work be done, and supervised, by professionals... The objections to the "Ada professionalism" posting were not objections to the idea that software should be built in a professional fashion, nor that poor quality code can be harmful. They objections concerned two apparent misconceptions: - that constructing software in Ada is qualitatively different, to an extent that requires Ada-specific solutions - that concerns of professionalism in programming can be addressed by formal/bureaucratic/paperwork-oriented solutions In particular, I think that the first misconception is the more dangerous. To the extent that the Ada community allows or even encourages the idea that it is qualitatively different from the rest of the software world, that community is isolating itself to its own detriment. From a different view: There is a LOT of critical software in existence today. A TINY (miniscule; insignificant!) amount of it is in Ada. -- Dick Dunn {hao,ucbvax,allegra}!nbires!rcd (303)444-5710 x3086 ...Are you making this up as you go along?