Professional licensing

[Daniel Wolstenholme]

>Message: 6
>Date: 25 May 2003 00:54:36 -0700
>From: "Ted Gould" <ted@gould.cx>
>Subject: Re: Professional licensing

>While I see both of these points as close to the same thing
>(although they vary in vulgarity).  Basically: I took the EIT,

You think "crap" is vulgar?  I guess you're pretty offended by
those ultra-vulgar PG movies huh?

>and I don't think that was a good thing.  So be it, but I tend
>to disagree.  I think that there are a lot of Engineers even
>who believe that the entire purpose is to keep people out.
>Many lawyers feel the same way about the bar exams. 

The PE, at least currently, is not a way to keep anyone out of
the electrical engineering profession.  I can assure you that no
large corporation cares one whit whether you have a PE or not. 
And probably not any small companies either.  The PE is just a
way for state governments to make money and feel important.

>There will always be good reasons and bad reasons for taking
>the EIT, and continuing on for a PE.  I think that the best
>reason is for protection of the public, while not every
>project requires that level of insurance, it would be nice if
>every one had that level of quality (I realize that is an
>unrealistic expectation, but I can hope can't I?)

Protecting the public from what?  Poor-quality products?  If you
don't like a company's products, you're free to buy from someone
else.  Ford sure makes some crappy cars, and some downright
dangerous ones (witness Exploder/Firestone), and I sure didn't
hear about any engineers with PEs getting blamed for that mess. 
But as a smart consumer I don't buy products from those
companies.  It's not the government's job to police product
quality, aside from mandating safety standards etc.  I'm already
paying too many taxes for unnecessary government functions (and
I'm not even a Republican so I feel a little strange saying
that); I don't need the state of Arizona making sure my
cellphone works properly or that my microprocessor doesn't have
a FDIV error.  This is something for companies and customers to
work out between themselves.

>And, for the record, my Masters is in Electrical Engineering
>and my title is "Senior Software Engineer".  I'm sure that I
>know nothing about being an Electrical Engineer who writes
>software.  Daniel, you might do a little bit of research
>before attacking someone's credentials.

Since you stated that you fairly recently took the EIT, you sure
didn't sound like anyone with much experience in industry.

>I believe that there are some states that specify for
>individual titles also, but I don't have any examples.

Well there's a whole bunch of engineers working at Intel,
Motorola, Microchip, On Semiconductor, and Honeywell who have
"engineer" on the business card and in their title, and don't
have a PE.  And that's just here in Phoenix.  Maybe they should
force all those engineers to be fired for breaking this silly
law.  That'll be really great for the local economy...

>I think that you'll have to explain to me how software
>development works then.  Is there not an overriding
>architecture and process to go along with that?  Is there not
>design reviews that are handled by more senior engineers?  I
>would agree that in general, software engineering hasn't
>gotten to the level of sophistication that other engineer
>professions have - but I think you can talk to the guys at
>NASA who write software about how it's 'nothing like' standard
>engineering.  Remember, that NASA has one of the most
>sophisticated engineering processes, and software is
>no exception to that.

Well this would certainly depend on the project.  There's a lot
of projects that just aren't that important or mission critical,
so they don't have design reviews, formalized architecture
requirements, etc.  Many times it's just one or two developers
working on it.  Sometimes it's just a simple script.  These
might not be the Space Shuttle control system, but these are all
"software engineering".  Do you think I need to get a PE to sign
off on my 3-line Perl script which calls another program?  Or my
modification to tkCVS to use a different editor?

>I think, perhaps, that you don't understand that level of
>complexity that goes into structural designs.  That many
>engineers work on the design, and craft different parts of it.
>It is then integrated into the larger design.  I see many more
>similarities than differences.  Today an engineering project
>is rarely small enough that one person can work on it oversee
>the entire design.

Ok, since you've just restated one of my original points, why
should any one person accept responsibility for a project so
large that no one person can oversee it?

>For the most part, it is easy to blame one person or group of
>people.

Whoa, hold on here!  Why should any one person accept blame for
something when you can only narrow it down to a group of people?
 I sure as hell am not going to accept responsibility for other
people's mistakes.

>If the problem is properly identified, and found, then you
>know where the error was.  Let's look at the earlier Pentium
>problems, there was one guy who was in charge of the division
>unit for the Pentium, I'm quite sure Intel knows who that was. 
>Sometimes though, problems come through interactions (and this
>is most of the problems in software in my experience) so it is
>an 'understanding' of the specification for the interface. 
>Sometime it is a problem with the definition, and that has
>to be dealt with differently.  Usually both ends are to blame
>in those cases.

The Pentium bug was one of thousands that existed before
validation caught them.  This one made it through, and wasn't
even very important.  So shouldn't the validation teams be also
at fault?  Or should every time a bug is found, the engineer who
erred be identified, fired, and lose their license?  Pretty soon
there won't be any engineers left.  How is licensing going to
help anything here?  And why does the state of AZ have any
business getting involved here?  This is a processor, not a
bridge.

>The approach that I take to solving this problem is having
>engineers on the project that will make sure it is tested in a
>reasonable manner. While I am also impressed by the quality of
>work that Underwriters Labs has done, I think that they will
>never be able to test the 'latest and greatest gadgets' for
>functionality.  Many times the functionality is the key thing
>that needs to be tested, not whether lightning causes it
>to explode.  There are cases where the only people that
>understand it enough to test are the people who built it.

Why does functionality need to be verified by any outside
agency?  UL only exists to verify that products are safe, not
that they work well.  If you buy a UL-listed TV, you can feel
safe that it won't explode and kill you.  If you buy a TV with
no UL, CSA, or TuV certification, you don't have this assurance,
and as a consumer it's your responsibility to look for this. 
You can still buy a TV without these certs though.  But the UL
isn't going to verify the TV menus are easy to use, or that the
remote buttons feel right.  It's your responsibility to do
research before buying stuff, instead of buying some cheap POS
and then being pissed that it's a cheap POS.  Why do we need any
controls on industry to police their product quality?  If people
are buying shoddy merchandise, that's their own fault.

>I think that is the entire point I'm trying to make, there is
>a penalty far greater than bankruptcy for the licensed
>engineer; he/she could loose their license to practice their
>profession.

But the company they worked for is free to continue selling
dangerous products since they can just pass blame to engineers,
hire some replacements, and not bother using proper testing
procedures which would catch these defects?  When you have
projects that require an organization, that organization as a
whole bears responsibility.  Not individuals within the
organization (it's the company's job to hire the right people,
check their work, test products, and fire people when they don't
meet requirements).



__________________________________
Do you Yahoo!?
The New Yahoo! Search - Faster. Easier. Bingo.
http://search.yahoo.com