[Baypiggies] A talk so completely applicable to Python
janssen at parc.com
Thu Mar 28 22:21:37 CET 2013
David Lawrence <david at bitcasa.com> wrote:
> No, it's mostly a comparison of civil engineering to software engineering.
> I don't remember the industrial revolution mentioned even once.
> On Thu, Mar 28, 2013 at 7:55 AM, Ian Zimmerman <itz at buug.org> wrote:
> > I haven't watched; just guessing, is the comparison to the industrial
> > revolution period?
Gist: as engineering proceeds from more concrete objects (bridges, as
with civil engineering) to more abstract objects (by abstract I think he
means "harder to see with the naked eye", as he alleges electrical
engineers build "more abstract" objects than civil engineers), the cost
structure changes as well, from dominated by labor and materials and
time (as with bridges), to dominated by the design process itself. Thus
the "most abstract" branch of engineering, software design, is all about
the design; the actual manufacture is done with automated systems
(compilers) that cost almost nothing to execute. This shift in the cost
structure means that different techniques (such as prototyping plus
testing) is more appropriate for the more abstract branches of
engineering. This is why software designers don't use the same methods
that bridge designers use.
I think the best part of the talk is where he recommends two books:
ENGINEERING AND THE MINDS EYE, by Eugene Ferguson
WHAT ENGINEERS KNOW AND HOW THEY KNOW IT: ANALYTICAL STUDIES FROM
AERONAUTICAL HISTORY, by Walter Vincenti.
I'd personally add Henry Petroski's book to this list, INVENTION BY
DESIGN; HOW ENGINEERS GET FROM THOUGHT TO THING.
More information about the Baypiggies