[Discuss] Another reason government licensing of software engineers is a bad, bad, bad idea (Was: Programming vs Engineering)

[MBR]

Whenever government, be it Federal, state, or local, is given the power 
to decide who is allowed to do a particular thing, that power is 
inevitably abused to satisfy the demands of some powerful lobbying 
group.  In the Constitution, the Founding Fathers gave Congress the 
right to grant copyrights and patents:

    "To promote the Progress of Science and useful Arts, by securing for
    limited Times to Authors and Inventors the exclusive Right to their
    respective Writings and Discoveries."

Copyright is one form of license.  It gives the Federal government the 
power to decide who can and who cannot copy a work.  U.S. Copyright 
originally lasted 14 years with an option of a one-time renewal if the 
author was still alive at the end of the first 14 years. 
http://en.wikipedia.org/wiki/Copyright_Act_of_1790.  Today, as a result 
of Disney corporation buying off Congress, the Mickey Mouse Protection 
Act of 1998 has extended that 28 years to *120 years*, but only if the 
work is by a corporation.  Human beings get only 70 years.  
(http://en.wikipedia.org/wiki/Copyright_Term_Extension_Act)  Copyright 
no longer serves "To promote the Progress of Science and useful Arts".  
As Duke University Law Professor James Boyle has pointed out, "We are 
the first generation to deny our own culture to ourselves. 
<https://twitter.com/#%21/thepublicdomain/statuses/3179288729> ... No 
work created during your lifetime will, without conscious action by its 
creator, become available for you to build upon. 
<https://twitter.com/#%21/thepublicdomain/statuses/3179305900>"

But let's get back to professional licenses.  The argument for why 
government should be able to control who can and who cannot practice a 
particular profession is that they're supposedly protecting the public 
from incompetents.  If a doctor, a lawyer, an electrician, or an 
architect screws up, people may well die!  Of course, in practice, 
licensing boards more often than not protect their own and incompetent 
practitioners generally keep their licenses as long as they don't piss 
off anyone on the licensing board.  So the licensing system doesn't even 
accomplish what it claims to.

Add to that the fact that no politician can resist the temptation to 
control other people's lives in order to pander to powerful lobbying 
groups.  As a result of such pandering to organizations like N.O.W., 
state laws all across the country suspend or revoke men's professional 
licenses if they fall behind on their child support payments.  
(http://www.ncsl.org/issues-research/human-services/archive-license-restrictions-and-child-support.aspx)  
Not only do these laws have nothing to do with protecting the public 
from harm or death if in the course of his profession the practitioner 
does something stupid, they also make no sense whatsoever WRT 
financially supporting children.  If a man can't keep up with his child 
support payments, taking away his ability to make a living, and thereby 
guaranteeing that he won't be able to make the payments, makes no 
sense.  I know of professional men who've never done anything wrong, but 
now have a prison record because they lost their licenses and were 
thrown in jail for being too poor to pay their child support.  But from 
a politician's perspective, giving a well-funded a lobbying group 
whatever they ask for keeps that politician in office.  The result is 
that we have reinstituted debtors prison in the U.S. under another name.

So, when you're arguing that we software engineers should be required to 
be licensed by the state in order to practice our profession, ask 
yourself this: "Do you really want to hand over the power to determine 
whether or not you're allowed to work in your profession to shills like 
Lamar Smith?"  Smith is the Representative in the House who's been 
promoting SOPA, which gives politicians, who couldn't design or 
implement a distributed database of name to IP address mappings to save 
their lives, the power to take a sledgehammer to the carefully crafted 
DNS protocol, thereby destroying the interoperability that makes the 
Internet the Internet, and in doing so, spitting on Jon Postel's grave!

This is not to say that I don't think there's a problem of bad code 
getting out into the world. Over the course of many decades I've 
encountered more than my share of sloppy software engineers.  But I have 
to say that only about half of the ones I've encountered were sloppy by 
nature.  The other half didn't want to code that way, but bad managers 
who pressured them to meet impossible deadlines gave them the choice of 
losing their jobs or hacking stuff together as quickly as possible 
without proper design at the front end and proper testing at the back end.

I don't have a solution to the problem, but I do know that government 
licensing of software engineers will not solve the problem, and will 
actually make things worse.

    Mark Rosenthal
    mbr at arlsoft.com <mailto:mbr at arlsoft.com>


On 1/22/2012 1:55 PM, Drew Van Zandt wrote:
> The trouble is that a PE certification proves:
> 1) You managed to make it through an engineering degree program
> 2) You practiced as an engineer for 4 years or so
> 3) You can test well. (Which, given that you got a degree, is likely.)
>
> I know people who meet all these requirements, and are actively bad as
> engineers.  One of them had his PE certification.
>
> None of the half dozen products he had a key engineering role in during the
> time I knew him ever shipped.
>
> If you want other engineers to take these certifications seriously, you
> need to not give 'em to guys like this.  More importantly, you need to find
> a way to give the certification to the really good engineers out there.  As
> long as you have tens of thousands of guys out there who are better
> engineers than the PE's I have met, I'm going to hire based on skill,
> certification be damned.
>
> For that matter, I'll hire based on skill regardless.  Certifications are
> perhaps handy filters for people in HR who don't know enough engineering to
> filter in other ways.  If you hire based on certifications (or degrees)
> blindly, you'll be sinking your own ship.  I've known too many PhD's who
> were useless for anything outside academia.  (That bitterness you hear is
> the echo of the sound of me cleaning up after a couple of them over and
> over again until management FINALLY realized they were the problem.)
>
> On the other hand, one of the best engineers I know fought her way up from
> being a tech with an associate's degree.
>
> I do know a couple of PE's who are good engineers, but the best engineers
> I've known have not been.
>
> *
> Drew Van Zandt
> Artisan's Asylum Craft Lead, Electronics&  Robotics
> Cam # US2010035593 (M:Liam Hopkins R: Bastian Rotgeld)
> Domain Coordinator, MA-003-D.  Masquerade aVST
> *
> _______________________________________________
> Discuss mailing list
> Discuss at blu.org
> http://lists.blu.org/mailman/listinfo/discuss
>
>
> -----
> No virus found in this message.
> Checked by AVG - www.avg.com
> Version: 2012.0.1901 / Virus Database: 2109/4758 - Release Date: 01/21/12
>
>