[Beowulf] Java vs C++ for interfacing to parallel library

Joe Landman landman at scalableinformatics.com
Sun Aug 20 12:43:31 PDT 2006

Jim Lux wrote:
> At 06:08 AM 8/20/2006, Joe Landman wrote:


>> Hey ... I like that +/vector is a sum reduction over a variable named
>> vector, and it can do it on the outermost index.
> But you need that special APL keyboard with the greek letters...

They weren't greek so much as ... er .. something.

>> Don't knock APL, it was designed for calculation.  That it was also as
>> arcane in its syntax as possible, as well as being insanely terse in its
>> implementation.
> And if you think the classic IBM implementation on the "first PC" (the
> 5100) was funky, you should see Iverson's book

I did read Iverson's book.  And I used one of them IBM terminals, you
know, the one with the ball-o-characters?  And I switched them from
normal to APL.  And back again.  I seem to remember one of those high
speed 110 or 300 baud acoustic coupler modems on it.

Later on I used APL on the PC, and it wasn't quite the same.

> That said, I spent an entire summer of my youth doing all sorts of
> interesting things in APL.
> Like any arcane and esoteric knowledge, if you use it regularly, it's
> second nature.  If you only use it occasionally, it's hard to do stuff.

It was fun at the time ...

>>  Still, any language that allows you to model a physical
>> system in 4 lines of code and gives you meaningful answers cannot be all
>> bad.
>> I ... just ... dont ... want ... to  ... use ... it  ...... again.  Ever.
> Come now.. think: a hybrid of APL and Matlab?

I like Matlab (and Octave), and I do think its syntax could be improved
a bit.  I am just not sure APL is the right way to go on this ...

> And doesn't APL elegantly represent inherently parallel constructs.  It

Yes it does.  And thats scary.

> might be the most appropriate language for parallel computing (at least
> Iverson might think this).

Yes, I am sure Iverson thought that.  The concept of a language which
allowed you great freedom to accurately construct the models that you
needed was wonderful.  Just this little learning curve....

Joseph Landman, Ph.D
Founder and CEO
Scalable Informatics LLC,
email: landman at scalableinformatics.com
web  : http://www.scalableinformatics.com
phone: +1 734 786 8423
fax  : +1 734 786 8452 or +1 866 888 3112
cell : +1 734 612 4615

More information about the Beowulf mailing list