RE: MCL Framework & Directions

[Steve Strassmann <straz@cambridge.apple.com>]

>Date: 25 Jun 92 02:30:05 EDT
>From: Dave Lucky <76557.704@CompuServe.COM>
>To: info-mcl <info-mcl@cambridge.apple.com>
>Subject: RE: MCL Framework & Directions
>
>Folks,
>
>Well, my simple mind is boggled.  I just can't figure out where Apple is 
>trying to lead Mac developers.  You know, the people who are going to create 
>the next "insanely great" application (either for their own organization or to
>sell to others).


Well, these are perfectly good questions. I'll try to answer them, not
as an official Apple response, but just from where I see it, personally.

You mentioned MCL, Dylan, C++, MacApp, and Bedrock, and asked
what the "future direction" is. The problem of course, is that no one
solution will satisfy everyone. There's also the undeniably important
political dimensions as well as technical ones.

 From what I've read about Bedrock, it seems like a natural extension
of MacApp. It's not trivially compatible, but there will clearly be
tools to help current MacApp users migrate to Bedrock. Apple has very
clearly stated that the future for C++ developers is to eventually
migrate to Bedrock.

MCL and Dylan are two very different products, and are expected to fill
the different needs of two very different communities. We have no intention
of forcing people to "migrate" from MCL to Dylan. MCL is a mature, robust,
shipping product which is compatible with an existing, de facto industry
standard. There is a tremendous infrastructure of existing Common Lisp
code, literature, and knowledge, and it would be foolish to waste that
sort of investment. As good as it is, there's plenty of room for growth,
and MCL will continue to grow and be supported. (Again, this is not an
official Apple policy statement, just my personal observation!)

Dylan, however, is not aimed at the Lisp market. It's aimed at the current 
market for static languages. We would definitely consider Dylan a failure 
if it only stole support from the lisp community and had no impact on 
the rest of the world. We think there are compelling reasons for a static
programmer to seriously consider switching to Dylan, and we'll do everything
we can to get them to do so!

>I'd like someone in the know at Apple to tell me what their objectives and 
>strategies are for developers like me.  I've noticed that you are pretty quiet
>when discussions about application frameworks occur.  Please tell us what's 
>happening.  Like Jeff, I'd also like to hear opinions from other developers.

I can really sympathize with this question, but unfortunately, Apple has enough
problems making promises and then switching signals. We're quiet because
we'd rather wait until we can give you a clear message and stick by it, and
I think in the long run you'll agree it's better that way.

Getting back to personal observations, I agree with Moon's recent message
that there's no reason why a future version of CLIM shouldn't be based on
Bedrock. This would give you an excellent combination of portability, performance,
and support.