> A hammer is a piece of metal. There's almost nothing that can go wrong.
> I respectfully believe this model is an over simplification of the issue.
> In contrast, a microcontroller circuit is so much more complicated piece
> of equipment. There are many variables, and using such a system without
> understanding these variables is the problem.

I don't follow what you are saying here. If you were to go and attempt to u=
se a piece of cast iron as the hammer head (after all, it is just a piece o=
f metal) then you would get a nasty surprise the first time you attempted t=
o do any real work with it. There is a surprising amount of technology in t=
he metal used for a hammer head.

Perhaps a better example is when building a house. I would go to a speciali=
st joiner to get the roof trusses made, rather than go through the loop of =
designing them, doing the stress calculations (just what is the snow loadin=
g in this area?) and then doing the cutting and fitting.

So it is with microprocessors. I think you are needlessly dissing the folk =
you see using Aurdinos as a hardware base. It may be that the slant on the =
class you are observing is toward software, and the hardware design - if sp=
ecialist hardware is needed - is done by a separate team. That is certainly=
 the way the laboratory I work for does things. When I design the hardware =
I may well do certain things on the basis that I expect certain software co=
nstructs or routines to be used, but the end software is up to the guys tha=
t are trained software engineers.=20

I also do some software, but I certainly don't do it on the basis of normal=
ly doing both the hardware and software. As others have pointed out, there =
will typically be two teams designing the product, the hardware team, and t=
he software team. The software team may well use 'standard' hardware like a=
n Ardino or PIC Explorer16 board to get software development and testing un=
der way while waiting for the hardware to be designed, debugged, and a dece=
nt test bed version of it supplied to the software team. Then once the 'rea=
l' hardware arrives some defines at the beginning of the software get chang=
ed and recompiled to suit the hardware differences.

I suspect the medical equivalent is the anaesthetist in the operating theat=
re - he has a knowledge of how the human body works and uses that to ensure=
 the patient survives. The surgeon however, relies on the specialist knowle=
dge of the anaesthetist to make sure that side of things goes without probl=
ems. Technically the surgeon can do the anaesthetists job, except the latte=
r has gone on to do specialist training in that area that the doctor hasn't=
..

So it is with hardware and software designers - the software designer needs=
 only a basic knowledge of the hardware to properly do his job. Technically=
 the software person may be able to design the hardware, and may get involv=
ed in discussions with what is required on the hardware side to make the ma=
chine work properly, and some people are skilled enough to do both sides - =
but not necessarily everybody. If you do have the ability to do both, count=
 yourself lucky, and you will probably go a long way. In your case you may =
be adding the medical specialty to your toolbox, and that will give you a v=
ery special mix of three important areas of expertise - but don't use your =
abilities to put down others without careful consideration of what may have=
 been the original specification of the job at hand.



--=20
Scanned by iCritical.

--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.