Re: [asa] What is exactly is a TE?

Gregory Arago wrote:

> 5) Finally, does
> anything an engineer ever ‘make/build/design/construct/etc.’ count
> as something that ‘has evolved’ (into being or having become)? From
> my pov, engineers do exactly that, they engineer, that is, they
> quite clearly do not ‘evolve’ things. To speak in such a language as
> ‘evolution’ for an engineer makes no sense; however, when engineers
> are trying their hand in biology, then speaking in evolutionary
> language may be suitable.

Gregory

I googled on "cpu evolution" and these are the titles of a couple of the
articles that are found:

Evolution of Intel Microprocessors: 1971 to 2007

  CPU Evolution
Hardware.info has a nice article outlining the evolution of computing
hardware from the CPU/motherboard/bus model to the
Co-processor/Streaming model.

CPU time evolution

Evolve in such cases as above simply means change over time and
frequently implies many many small or even tiny incremental changes in
function and performance. Many of these small incremental changes occur
prior to customer availability as the design iterates towards a more or
less finished product. In general, except in the kind of programming
that Iain was talking about there is little analogy to random gene
copying errors and natural selection and so in that sense it is not
analogous to biological evolution.

Most products have many semi working prototypes prior to market
availability. Sometimes something somewhat akin to random change and
natural selection occurs, usually in large multi team projects. What
happens is that each team may have been working on their own priorities
and sometimes when the whole system is integrated, unexpectedly good
results occur that people spend effort on to try to preserve. Usually
what happens is that the two pieces don't work well together and things
have to be fixed to resolve the problems. Also the market place
provides a kind of analogy to natural selection, think of the Ford Edsel.

In computer language compilers where I worked there always was a great
feeling when the "Hello World" program would run. This program simply
prints out a single line "Hello World", but a lot of function has to
work right for that to happen. The number of test buckets for such
compilers are often in the 50K to 80K range, and it can take months or
years to make them all run properly and feels like forever to the
project leaders who watch the weekly pass/fail numbers.

Of course these analogies are not 100% but neither is there no
similarity. Different scientific and engineering areas appropriate
terms and ideas from one another whether one likes it or not and often
it is fruitful. In an article in the most recent PSCF revised Humpty
Dumpty semantics are defined and that definition seems to have relevance
to this topic.

Dave W (CSCA)

To unsubscribe, send a message to majordomo@calvin.edu with
"unsubscribe asa" (no quotes) as the body of the message.
Received on Sun Sep 9 15:36:21 2007