Re: Practicality of functional and logic languages?

maniattb@cs.rpi.edu (Bill Maniatty) writes:
>Has there been research in creating architectural support for
>functional programming languages?


Well, there are (were) Lisp Machines, and there have been object-oriented
architectures. Probably your question is more closely answered by things
like the SKIM machine for hardware graph reduction.


In short, yes.


These machines tend to suffer from the same problems as Lisp Machines,
though: because the funding for development and implementation is many
orders of magnitude less than that for conventional architectures, it is
always better to wait a year and buy a conventional machine programmed to
imitate (say) a graph reduction machine.


      What about the power of the functional programming paradigm? Is there an
      important class of problem which cannot be done by Functional Programming?
      Could you write an Operating System using Functional Programming
      techniques for example?


Yes, you could. In fact, we are working on that here at CMU, extending SML
for doing OS work (in particular, comms suites and drivers). We're not
abandoning the paradigm (of course, purists might suggest that SML doesn't
really adhere to the functional language paradigm in any case).


You'd have more trouble with real-time software, but most languages have
trouble with that. The paradigm doesn't prevent you, you just need a
language (and a runtime) that can handle asynchronous events.


Nick Haines nickh@cmu.edu
--

Return to the comp.compilers page.
Search the comp.compilers archives again.