[Prev][Next][Index][Thread]
Re: Origins of Dylan
It was never the objective of Common Lisp (and CLOS) to make it
very efficiently compilable, nor was it a goal to allow Lisp apps to
be compiled down into, e.g., DLL's.
It was explicitly among Dylan's goals to:
- allow very efficient compilation
- allow the generation of shared libraries such as DLL's
- have a "pay as you go" strategy, whereby unused "advanced"
features don't penalize people who don't use those features
Dylan is not "Lisp done right". It is instead a dialect of Lisp that
was designed to meet some different constraints.
I would still use Lisp to do highly dynamic, speculative development
in fields like AI or designed very complex domain-specific langauges.
If I want to write "ordinary" complex programs (e.g., compilers,
editors, devel environments, web servers), I would use Dylan.
qtom@my-deja.com wrote in message <8hv3tm$qde$1@nnrp1.deja.com>...
>Hi,
>
>I've been reading Object-Oriented Programming: The CLOS Perspective, ed
>by Paepcke. It's a really informative comparison of CLOS with some other
>major OO languages.
>
>Something that has struck me in reading the book is how much the
>developers complain about the compromises they had to make in order to
>keep backwards compatibility with all the old flavors of Lisp. I
>remember such complaints being made about Common Lisp in general.
>
>Recently, I was looking at some material on Dylan and notice that the
>developers were a lot of the same people who were involved with CLOS.
>That made me wonder, is Dylan what they really wanted to make when they
>made CLOS?
>
>Does anyone know to what degree Dylan was intended to be a better CLOS?
>Or was it actually focused on a different range of needs?
>
>Cheers,
>Tom
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.
Follow-Ups:
References: