John and anyone, comments
Rich AT EnglishLogicKernel DOT
9 4 9 \ 5 2 5 - 5 7 1 2
[mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of John F.
Sent: Monday, October 18, 2010 5:33 PM
Subject: Re: [ontolog-forum] Oooh, FOL is too
hard to learn.
I have taught predicate
calculus to average engineers, and I believe
that the notation is truly
*abysmal* and *unusable* for anybody who
is not a born and bred
mathematician. The ideas seem to sink in
while they're in the classroom,
but by the next week, their minds
are totally devoid of any
concept from the previous week.
This engineer/computer scientist
agrees. You mathematicians have a culture that uses symbols that literally
have symbolic meaning to you (fine with me), but I use symbols to chunk snatches
of experience, to apply them to new experiences in the present or future, etc.
> I think, John, that you
are here exemplifying the top 20%, who think
> that every other competent
engineer can do more or less what they do.
I most certainly do not believe
that at all. I know seemingly
intelligent people for whom
anything that looks remotely like algebra
is an instant turn off. I
have spoken to publishers who say that
each equation that appears in a
book will cut sales by 50%. That
means 10 equations will reduce
sales by a factor of a thousand.
> Many people use modeling
languages in the same way, to state
> unclear thoughts clearly
and often incorrectly. They don't use
> use the language to mean
exactly what the formal semantics of
> of the language says is
meant by the syntax they used.
> Many people have no
problem with simple syllogisms, but are seriously
> confused about
> And I can tell you first
hand that the first encounter between
> electrical engineering
students and boolean algebra is a filter -- the
> ones who will work in
electronics understand quickly, the ones who don't
> understand quickly will
become radar technicians or something. It is
> not just the notation; it
is the abstraction. Many of the
> simplifications of gating
logic are counter-intuitive.
Again, I have had exactly the
Iterate again here. The math itself
is interesting only as a tool, not as an abstraction in itself, unless it?s the
first time I hear it. But repeating it over and over as though
> FOL is not casual logic;
it is a mathematical discipline.
We have to distinguish here
between notations for FOL,
and the use of FOL as expressed
in ordinary language.
> Many intelligent people
can use logic correctly in their work, but
> they don't have the
discipline, and most of them don't understand
> that there is a
Yes, that is certainly
true. I most definitely do *not* believe
that formal logic is the
foundation for NL semantics. On the
contrary, formal logic is an
*abstraction* from NLs when they
are being used correctly --
even when (or perhaps especially when)
the people aren't aware of the
People who have no education
beyond 4 grade can reason very
accurately about subjects they
know very well. But they're
hopeless when it gets to any
kind of abstraction. Unfortunately,
that point is true of engineers
who have had 16 years of education.
> Conversely, the most
astonishing use of FOL I have ever seen was a
That's not a
counterexample. Three-year-old kids are much
intelligent than the average
engineer. That's because they haven't
been brainwashed by 16 years of
When computers first became
available at large corporations,
scientists and engineers were
among the *last* to use them.
They never learned to use them
until their *children* shamed
them into learning
Summary: I think we agree
quite well on all these issues.
But I still maintain that
they're not an excuse for designing
software and notations (such as
UML diagrams) that do not have
foundation. People can benefit from a well
designed system, even if they
don't know how it was designed.