Phoenix.pm: Programmer-standardized testing - was Parameter p

[Scott Walters]

> Guess it is time to open my mouth ... Wide enough to allow my foot to be
> shoved in it later!

phoenix-pm-list hasn't seen any bloody battles yet (knock on wood).
With matters of opinion, one shouldn't worry about being wrong. Since
we aren't in a position to implement these ideas, it's all theoretical.

> As can my friends who work in completely unrelated, but regulated fields.
> They describe similar stories in their totally unrelated fields.  The point
> being that I do not believe this is isolated to unregulated fields.  It is a
> product of human nature.  Some people are lazier, slower, manipulative,
> etc... than others and licensing will do absolutely nothing to curb that
> problem.  

Most industries have a readily understood, easily observed product.
Production can be quantified. Because of the technical yet abstract
nature of code, a staff of hundreds of developers can get away with doing nothing
for years, asuming no one blows the whistle - not in every case, of course,
but it does happen. I'll grant you that this is human nature - I maintain
that the scale can be larger in some cases in IT/CS ;)

> Thousands of totally unqualified individuals receive degrees every year that
> should never be allowed to professionally practice in whatever field they
> have chosen.  Will another test somehow succeed where the evaluation of an
> educational institution has failed?
> 
> > Code is seen as an investment, and an investment that is not to be made
> lightly, and 
> > managers hearing "this is aweful... I have to rewrite this" one too many
> times has Perl 
> > down for the count.
> 
> Is this not the type of thing one hears when a programmer evaluates
> another's code, regardless of the language used?  Myself I thought it was
> more of an ego issue (I write better code) and not necessarily due to the
> language a program is written in.  I have said those same words looking over
> basic, perl, shell, C, and just about everything else.

Be it myth or fact, Java markets itself as reducing the frequency of this
problem. Perl has so many syntaxisms and idioms, and "dialects of Perl" as
Larry Wall puts it, that it is difficult to know enough of the language to
understand the work of otehrs. Perl has a deserved reputation for being unreadable.
If people use only the features that directly correspond that features in
Java (for instance), there would be a common dialect, but at the expense of
the expressiveness of Perl. Some times other people really do write code
with ignorant choice of idiom and feature, but with Perl it is often just
unfamiliar dialects.

This problem is above and beyond ego associated with one's own code, and
the too often valid suspicion of other people's code.

> My dad used to be a mechanic and he said the same thing about the work of
> some of his fellow  licensed mechanics.  Programming is not the only trade
> to suffer from shoddy workmanship.

Auto mechanics is another rare field where one person has to come in and
do work after another person. Most people elect not to have any repairs 
on their engine done more advanced than replacing a water pump or starter.
With an engine, there is generally one correct way for a repair to be made -
as long as bolts are tentioned correctly, parts are within wear limits,
and it is put together consistent with the service manual, everything is fine.
Programs have no service manual and no fixed structure, but are each ongoing
designs by amature "engineers". I agree with your point, but I think the
scope of the problem is larger with software.

> > Per my first round of suggestions, I see the crux of the problem as
> programmer
> > and employer dysfunction and ignorance.
> 
> Agreed.  If you could develop a test to weed out ignorance then we should
> implement it across society as a whole.  Might not be to many people left,
> but... 
> 
> Face it.  Those of us that excel in our fields only excel because we are the
> rare breed.  Meaning we are surrounded by the average and incompetent (by
> the standards WE defined, of course).  Perhaps if we just accepted that the
> world is full of average people this wouldn't bother us so much.  You cannot
> break the bell curve on this one.

Thanks. Like I said, employers hire a broad range of experience levels - 
some places are happy with a bunch of a budget programmers. 

Attitude has a lot more to do with your success than ability, in my experience,
but that is a different topic. Many shops place "team player" as a virtue
above ability. Others shops have other requirements.

> > Computers make basement dabbling easy. History has shown that sciences
> start off as 
> > things that can be done in the basement and then scale in complexity
> beyond that.
> 
> A lot of innovative science work is still being done in the basement.  Once
> the basement scientist proves his theory and/or completes a prototype it is
> sold off to a huge corporation for refinement.  Yet the creative groundwork
> was still done by the common man in his basement/garage/backyard.  Science
> has most assuredly not been relegated solely to the corporations of the
> world.

Electronics as a hobby has lost popularity lately - these people used to
make their own shortwave radios, do random projects, microprocessor integration,
and so forth. It was a thriving hobby industry - Radio Shack had a large
selection of tools and parts, not just a few odds and ends that they have now.
Several publications were dedicated to it. Now that complex electronic
devices can be had inexpensively, and the complexity of the average circuit
is beyond what a hobbyist has time and energy to understand, and the whole
field is becoming more propreitary, people are losing interest. Circuit
integration has been a hard blow - formerly transparent things are now
cloaked in a black box. I see Digital Rights Management serving a similar
role for the computer hobby. People still design and built circuits, there
are just far fewer of them, using more expensive tools, doing it for money,
not love. The things that can be built at home aren't interesting compared
with what can be bought.

As fields mature, research and practice seperate. I'm not saying major
corporations own science, just that the tools to do interesting work
in a mature field are out of reach for hobbyists. This could well be used
as a definition of "mature" for a field.

> > Companies don't trust programmers - not even good ones - especially good
> ones - so they 
> > *have* to find shirfers to trust that take care of that pesky trust thing.
> 
> Ironically the trust they are looking for is most appealing when sold to
> them by a used car salesman.  His promises and assurances cloak them in a
> fuzzy, warm blanket of trust.

Exactly. When is the last time a programmer said to a client, "Oh, it
doesn't matter what the problem is, don't worry about a thing, I'll have
it all taken care of"?

The willingness to put your reputation on the line at the drop of the hat is
powerful. It only lasts so long, but you can sell a lot of cars before the
used car industry is universally shunned.

(The New Hackers Dictionary, Second Edition published alternate style guidelines,
including rules for punctuating quotations. I'm being a radical and using these
rules, sorry).
> 
> This has dragged on so long I am not really sure what I wrote anymore.
> Summary: I am not against regulation.  It could provide some benefits
> (increased pay), but I seriously doubt it would solve any of our troubles.

Hahaha, I think I'm done too. Thanks for your views. Feel free to refute me,
but I think I've bored PM with my views enough ;)

> - Jeremy "just avoiding doing real work" Elston

-scott