dwp@willett.UUCP (Doug Philips) (02/15/90)
I've just started catching up on the old issues of Forth Dimensions,
starting with Volume I. What to my hungry eyes did I find, in Vol 1
Issue #6 but a transcript of a talk Charles Moore gave in October 1979
at the FORTH Convention in San Francisco. What is scary, reading it
over ten years later, is how some of things he talked about are
*still* true. The following quotes are from the reprinted transcript.
Any typos and other mistakes are mine. I've extracted pieces that seem
interesting to me.
We developed a number of systems at NRAO and encountered the
issue of patenting software. Programs cannot be patented;
ought not to be patented; would be very expensive to patent.
... FORTH seemed like something that perhaps should be patented,
so we spent a year writing proposals, investigating and getting
lawyer's opinions. The conclusion was that maybe it could be
patented, but it would take Supreme Court action to do it.
NRAO wasn't interested. As inventor, I had fallback rights but I
didn't want to spend $10,000 either, so FORTH was not patented.
This probably was a good thing. I think that if any software
package would qualify for patenting FORTH would.
...
I forsee no fundamental changes in the design of the language
except for the accomodation to the standards which are becoming
increasingly important...but there has never been a demand for
portability. In fact I know very few programs that portability
has ever been seriously attempted with. The time has clearly
come to change that.
...
It may be that FORTH is not merely a programming language. It
may be saying something much more important about communication
-- between people, between computers, between animals. This is
startling! It had never occurred to me that anyone would really
"speak FORTH" in an attempt to communicate with anything else
than a computer; it is not any longer clear that that is the
case. There may be concepts embodied in FORTH of greater utility
to the basic problem of communication.
...
I would estimate that there are now 1,000 FORTH programmers,
which is 2 to the 10th power and comes out nice and round -- a
doubling time of one year. Actually, I think the doubling time
is slightly shorter than that -- 10 times in three years and that
comes out to 2,000 as some people would prefer. What we conclude
is that next year there will be twice as many programmers...
We don't know how it is going to come out. FORTH, Inc. can't
train twice as many people next year -- well, maybe we can. But,
somehow the FORTH community as a whole has got to train twice as
many people next year and thereafter. ... I have fairly great
confidence that 1) the doubling time is one year, and 2) it is
going to continue.
...
[Lots of history of Forth elided...]
The first thing to be added to what had already existed was the
return stack and I don't remember why that was. I don't remember
why I didn't just put the return information on the parameter
stack. It was an important development to recognize that there
had to be two stacks -- exactly two stacks, no more, no less.
...
[Task of coherently extracting quotes describing the capabilities
of Forth is left for another time...]
I think that completes the capabilities that I think of a FORTH
today and I think you can see how they dribbled in. At no point
did I sit down to design a programming language. I solved the
problems as they arose. When demands for improved performance
came along I would sit and worry and come up with a way of
providing improved performance. It is not clear that the process
has ended, but I think it is clear that that process has now got
to be carried into the hardware realm.
...
CONCLUSION
To close on a philosophical note: power to the people.
This is the first language that has come up from the grassroots.
It is the first language that has been honed against the rock of
experience before being cast into bronze. I hesitate to say it
is perfect. I _will_ say that if you take anything away from
FORTH, then it isn't FORTH any longer, that the basic components
that we know are all essential to the viability of the language.
[Comments all? -dwp]
---
Preferred: willett!dwp@gateway.sei.cmu.edu OR ...!sei!willett!dwp
Daily: ...!{uunet,nfsun}!willett!dwp [in a pinch: dwp@vega.fac.cs.cmu.edu]