[sdiy] Who Needs a Degree?

[Richard Wentk]

You don't need a degree to build stuff that works. You do need experience to build stuff that works, and that experience is always a mix of hands-on and theory. 

But you - usually - need a degree, or at least degree-level knowledge and insight, to *innovate*. 

There are always exceptions, but it isn't usually the technicians who invent completely new algorithms or new technological processes. 

Engineering has different levels of abstraction and innovation. E.g.

1. Maxwell's equations - a complete game changer at the theory level. These happen rarely because the skills needed to create them are very rare.

2. Thermionic characterisation. Fleming realised that Edison's thermionic effect could be used in a radio. Other people refined the diode concept and characterised thermionic components as devices. You need *deep insight* to work at this level, and bridge the gap between theory and possibility.

3. Radio design by rote - eventually you end up with a package of equations, concepts and standardised parts that can be designed and assembled by anyone who's reasonably bright.

Innovation requires a mix of abstraction, general cognitive ability, practical knowledge and creativity.  It's impossible to innovate without good abstract skills - and I'm defining innovation as a process that extends what's possible in a general industry-wide sense, and not as a specific solution that re-implements or marginally refines existing concepts.

In the synth world, analogue voltage control, FM and sampling were innovations. A VCF design that uses modern OTAs instead of 3080s, or an FM synth built from an FPGA is a refinement. 

In computing, stored-program architectures, compilers, pre-emptive multitasking and GUIs were innovations. Any specific language compiler or GUI implementation is a refinement.

In practice the distinction is a sliding scale, not a binary switch. But generally you need *deep insight* to innovate, while you merely need to be competent to build stuff. 

As for degrees - universities and research establishments are there to innovate. Some actually do.

Degrees obviously don't teach basic competence - and almost everyone is confused about that, including businesses and universities. But a good degree suggests an individual has the potential to innovate. It may also imply they can build stuff that works, but in fact academic engineering is really about making innovation possible. 

It's one of the problems with engineering as a discipline that it doesn't make these distinctions explicit. Builders and Innovators have very different skill sets, but both get called Engineers.

Richard

On 28 May 2010, at 22:38, Grant Richter wrote:

> I would like to thank both Karl Dalen and Paul Schreiber for their well reasoned, well written and mature essays covering a number of topics related to the college experience.
> 
> I think there are a few things most respondents would agree on.
> 
> 1. Having an engineering degree does automatically ENABLE you to do talented engineering design.
> 
> 2. NOT having an engineering degree does not automatically PREVENT you from doing talented engineering design.
> 
> 3. The more experience you have doing "technical stuff", the better off you will be.
> 
> A couple of specific responses:
> 
> From Karl:
> 
>> Then there are the superstars, those few who establish a mind-meld with the code or electronics. Have you ever worked with one? When the system doesn't work, mysterious bugs baffle all of our efforts, up comes the guru who sniffs, licks his finger and touches a node, and immediately discovers the problem. We feel like idiots; he struts off in glory.
> 
> He worked on a similar project 9 years ago, had a similar bug and remembered it.
> 
> From Paul:
> 
>> - able to easily any quickly jump from problem to problem without panicing that the specific knowledge is not yet acquired. Many techs excel in a *limited and specific* area of electronics, but cannot easily & seamlessly jump from DSP to RF to FPGA to pc board layout to wiring to firmware.
> 
> Without quibbling, HR would generally classify experienced multi-tasking as a "project management" skill not specifically engineering related.
> 
> I had my first electronics job in 1978 doing Motorola 6800 based numerical machine controls.
> 
> I know at that time the technicians were there to save the engineers ass, by finding the engineers mistakes and finding work-arounds to keep on prototyping while the mistakes were corrected in drafting.
> I know at the time the engineers were there to provide jobs for technicians. In biology they would be called "obligate symbiotes".
> 
> I never did figure out what "Management" was for, except to reprioritize the projects every day, "Good Morning, Today everyone run as fast as you can to the West, thank you." "Good Morning, Today everyone run as fast as you can to the North, thank you."
> 
> I do know that management cooked up the "technician" / "engineer" distinction to keep the two groups at war, which kept them from realizing that they had 100% of the valuable skills and had they co-operated, could have dumped the fat cats in the suits in the blink of an eye.
> 
> 
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy