[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Future Plans for MCL

Dear Mr. Fleischman:

Like Henry Lieberman, and no doubt many others,
I too am saddened and puzzled by your message on Apple's decision on MCL.
I agree with Lieberman that it is a big mistake.

Here at the University of Waterloo's Statistical Computing Laboratory
our application area is possibly quite different from the usual MCL
We have developed a system (Quail) for data analysis and statistical
modelling, that involves extensions to CL and a statistical graphics
package that we believe to be second to none.
There is no question that MCL is the development environment of choice.

We too have been active MCL users since the Coral days.
Indeed, we originally purchased Apple hardware, and continue to do so,
because of MCL! --- not the other way around.

For statistical computing, data analysis, and I venture to say the 
great bulk of scientific computing, features like speech recognition,
movie making, etc. are of relatively little serious interest.
However, a programming environment where one can quickly prototype new
ideas or adapt existing methodology to new problems is extremely
important and valuable to many.
If Apple would like to make inroads into the scientific community,
I think it would be well advised to enthusiastically promote MCL,
rather than virtually hide it as seems to have been its policy so far.

I should mention that we also have several SGI machines whose graphical
capabilities are beyond anything Apple has to offer.
But because of the richness of the programming environment of MCL over
that available on the SGI (or any other box so far) we continue to use MCL
as our primary environment (development and delivery).
Indeed, I have just refurbished the lab with quadras  --- again
only because of MCL.

While we do not have the popular profile that the MIT media
lab has,  students regularly use our MCL-based software in quantitative courses 
and that use is likely to expand.
A number of students have used MCL as the programming environment for
their graduate research in Statistical Computing and Graphics.
I will echo Lieberman's comments when I say that we at Waterloo
                                                   "...teach courses
that train many future innovators, and our use of MCL later influences
choice of the Macintosh as a platform for future projects in our
sponsors and employers of our students, even if MCL is not used in final
deployed versions"
Moreover, as is the case elsewhere, the results of research in this
laboratory are discussed and demonstrated at scientific conferences around
the world.

It would be a shame to lose MCL.  To us, and no doubt others in the
general scientific computing community, it would mean dropping the
Macintosh as well.


R.W. Oldford
Associate Professor
Dept. of Statistics and Actuarial Science
University of Waterloo
Waterloo, Ontario N2L 3G1