[info-mcl] Pathname and Unicode in MCL 5.1

[Sean Ferguson]

Thanks for this excellent summary.

Another aspect of MCL that I *really* like is Fred. I know that many 
people really love emacs, but there are all sorts of features in Fred 
that make it very easy and efficient to use. And you don't have to 
necessarily learn an entirely new set of keyboard commands than those 
that are used on your platform of choice.

I also have a huge investment in the gui aspects of MCL (eg. the 
quickdraw interfaces) that would be almost impossible to port to any 
other platform.

As far as reuiniting OpenMCL and MCL, I know that is not a 
possibility. I would really love to have Fred on OpenMCL and some 
reasonable path for porting gui code from MCL. If we all chip in 
$1,000 maybe we could hire someone to do this (hahaha...).

At 9:45 PM +0200 10/2/07, Rainer Joswig wrote:
>The state of Lisp on Mac OS X on Intel is both excellent and depressing.
>
>It is excellent because there are now lots good Lisp compilers 
>running on Mac OS X (Intel):
>   CLISP, SBCL, LispWorks, Allegro CL, CMUCL and others.
>
>It is depressing since there is no Lisp that is as good integrated 
>as MCL was for Mac OS 9 (and before).
>
>Okay, we have now more options with Mac OS X, since it has Nextstep 
>and UNIX/BSD heritage.
>A good Lisp on Mac OS X needs to be able to do more than MCL under 
>Mac OS 9. The
>Lisp needs to be a good citizen under Unix and it should have a Cocoa bridge.
>
>What we don't have is a Lisp with a simple, but powerful, IDE, a 
>nice GUI toolkit,
>easy to use OS interfaces, good usage of CLOS for the library and, important,
>'full' source code.
>
>Allegro CL: no IDE, no GUI.
>LispWorks: IDE is not the typical Mac application, no source for the 
>IDE (except some editor code),
>    no source for the GUI Lib
>MCL: uses Carbon, only PPC, no Rosetta, no Intel support, often 
>breaking with the next OS release
>OpenMCL: IDE is a bit ugly, no 32bit Intel (yet?), 10.5 support not 
>yet available (10.5 also not available).
>
>Generally I think that OpenMCL is currently the best bet for a good 
>Lisp on Mac OS X,
>once Mac OS X is out and people use 64bit Intel machines or PowerPC machines.
>OpenMCL needs a lot of work on the IDE side, but it has a lot of potential.
>
>Reuniting OpenMCL and MCL makes no sense, IMHO. OpenMCL and MCL are now
>very different and OpenMCL is already further along on the road to improve
>support of Mac OS X technologies. Some of the IDE tools of MCL would 
>be nice to have
>and maybe a compatibility to Carbon to use it with older code bases 
>would be useful
>
>Users might want to check out OpenMCL on 64bit Intel if they have a 
>Leopard version. But you
>need some help to get the IDE running (worked for me). I plan to 
>move to Leopard when it is
>released and I guess lots of others will. Leopard will be out in one 
>or two months.
>Especially for developers there are many improvements
>in Leopard. With OpenMCL one will not be able to create GUI applications for
>32Bit Intel Macs or for Intel Macs running 10.5. Is that a problem? 
>Somehow, but all
>Intel Macs are sold as 64bit versions and I think the adoption rate 
>of Mac OS X 10.5 will
>be high.
>
>For many a 32bit Carbonized MCL on Intel Macs would be fine. I hear that some
>Lisp developers are still using PPC Macs just to use MCL.
>
>Generally I think that the approach of OpenMCL's IDE is way too conservative.
>I know its work, but I think there is something to be learned from XCode,
>DashCode, Interface Builder and other tools. I can easily imagine
>a nice Lisp IDE that has interface elements from iTunes, iLife and iWork.
>I feel that there is an opportunity. None of the other languages has something
>really cool (atleast I have seen none, maybe people here have seen
>some good (non-Apple) development environments that have  a cool
>Look&Feel under Mac OS X).
>
>Regards,
>
>Rainer Joswig