Was: "Could Haskell be taken over by Microsoft".
Ugghhh.
I propose to change the subject. So, at least, I have changed
the "Subject"...
It seems that the first time when somebody announced the wonderful news
on this list was April the 1. Some people - I was among them - thought
*that*, what we could have thought that day, I have even submitted a posting
about the Fully Lazy debugging of MS products.
Next stage was the personal announcement of Simon P J, with a - not really -
slightly ambiguous statement that Microsoft has nothing against the current
status of the language.
And now people seem really deeply worried.
Now, look around. There are some *languages* which are commercial, i. e.
without decent free implementations, as Lingo, but usually a language is
just a well documented abstraction. Nobody will claim that the Big Gurus
of Scheme had any commercial aims, and we have still some dozens of good
free implementations, but eventually the commercial implementations *must*
arrive on the market, usually bundled with some specific applications, as
Win/3d Scheme of Schemers Inc. The history of Common Lisp repeats itself.
Miranda had never a free implementation, although at FPLE'95 there were
some rumours probably not entirely unjustified...
I don't see any reason why this cannot happen to SML, CAML, Clean, Hope,
SISAL, etc.
At the university of Paderborn in Germany the group of Bruno Fuchssteiner
produced an *excellent* package/language for symbolic computing: MuPAD.
When the development of the Big Ideas was over, and they started to polish
the technical details, the interface, etc. they had to organize that on a
commercial basis, becuse of the resource problems. Again the transmutation
similar to that of (free)Cayley -> (commercial)Magma took place. The same
story applies to another algebraic language: FORM of van Vermaseren. The
free versions of MuPAD and FORM are still available!
***
C'est la vie.
What can *we* really lose? The sane standardisation? Too many people are
already a bit unhappy about the structure of the Standard Prelude, monadic
comprehensions/absence thereof, etc. I think that what really matters is
the fact that the "Real World Applications" might have a chance to become
a reality and that my students will not ask any more: "...but why did you
choose Haskell to teach Compilation? EVERYBODY knows that everybody uses
Lex and Yacc, and it is obvious that you should have use 'C'! We don't care
too much about your 'pedagogical' reasons, we want to combine the learning
of concepts with the practice of really applicable tools. WE leave the
University in 2-3 years..."
Of course I would prefer that the creators, implementors, maintainers
and philosophers of Haskell create their own enterprise instead of "selling
themselves" (no offence meant, sincerely), but it would require from them
a non-functional approach to Life, Universe, and Everything.
Best regards, and Thank You for all the Fish.
Jerzy Karczmarczuk
University of Caen, Normandy, France.
