Hi! Lots of good points! It was probably a mistake to give me a
soapbox :-)

>hope to dig
>into it at length.

I look forward to your feedback!

>one thing I can assure you is that many,many people
>are listening.

How do you get the lurkers to reveal themselves?! Maybe it's like a
newspaper - lots of people read it, but very few send in letters to
the editor! If that's true, how do you measure whether attitudes are
changing? IMO a test of this would be if application development
topics start to get the attention of academics - any idea if this has
started to happen yet? Your comments about peopleware and workflow
suggest that it may be...

>and I would be quite happy to engage you at
>length, however long you'd like to chat here.

Terrific! Do we have to worry about boring people who don't share this
interest? ;-)


>re: cobol. yes I heard an estimate that SIXTY PERCENT of all lines of
>code in the world is COBOL. seems quite plausible to me, although
>maybe less nowadays.

I've heard even higher numbers. Believe it or not, there was a time
when we were told "even managers will be able to understand code".
Were those more innocent times or what?

>koros

That's probably Khoros - the language is Cantata - sounds like they're
still truckin' along - maybe they'll be the ones to get the revolution
kick-started.

http://www.khoral.com/khoros/

As I describe in the Bibliography in my book, a very large number
of related systems and approaches have been developed over the years,
and one wonders why they never seem to get together! In recent years,
some of them have begun to take advantage of the new and improved
graphics facilities. My own picture-drawing tool (DrawFlow) is pretty
primitive, but it's simple enough that I can take it any direction we
decide is useful. Plus it can be used at any level of granularity -
from applications exchanging data files down to the actual
implementation level of FBP. So I can support stepwise decomposition.

For a total graphical approach, an interesting one is Toontalk. The
graphics are powerful, but IMO the underlying paradigm is a bit
abstruse. I have serious doubts about doing banking apps in Flat
Guarded Horn clauses, but I would be happy to be proved wrong!

http://www.toontalk.com/English/computer.htm

Yes, I have heard of Labview - it makes a lot of sense, especially
for the environment they are addressing, which I assume is naturally
asynchronous. This seems to be the biggest stumbling-block for
prorammers brought up in a single-thread world: the world we live in
IS asynchronous, but we keep trying to force it into a single-thread
strait-jacket.

One other point: like Gelernter (of Linda fame) I believe it is wrong
to try to develop just one language to support all possible
applications. One of the exciting things (for me) about FBP is that
you can combine modules written in many different languages, because
FBP really doesn't care.

Hey, you're right - I admit my naivete. Where do I go for
(re)education?!

>onward and upward!



>
> onward