Actually, I think there has been at least one 64-bit Squeak. Didn't 
Georg Gollman do one for the Alpha?

P.S. (for Richard) we aren't programming in C (and we don't want to) 
-- that's one of the large points of the enterprise -- it's just 
being used as an assembler for a higher level generator.

Cheers,

Alan

-----

At 8:54 PM -0800 1/15/01, Duane Maxwell wrote:
>  >Massively worse than all of the above, it is rumored, and
>>experimental evidence seems to support the theory, that Squeak's
>>C code stuffs direct pointers into 32-bit words/ints.
>
>That is correct, sir.
>
>>This is a
>>very serious bug.
>
>... if your integers and pointers are different sizes, and not 32 bits.
>
>>While small integers and pointers may be
>>interchangeable inside the Smalltalk world,
>
>Not at all.  This isn't a Smalltalk issue so much as an implementation
>issue of this particular VM.
>
>>this is a gross, hideous sin in the C language.  Any competent programmer
>>should know better to do that.
>
>This is a well known, understood, and documented limitation of Squeak's
>portability.  Given that C's immediate ancestor treated integers and
>pointers as a single type, it would be disingenuous to suddenly prevent
>anyone from joining the type-fascism country club just for being a little
>base born.
>
>>(This one item has grossly reduced my
>>estimation of the quality and usefullness of the entire Squeak
>>project.)
>
>Well, that's a bit of a shame, when there's so many other, better reasons
>to question the usefullness and quality of Squeak :)
>
>Squeak's heritage can be directly traced back over 20 years, when even
>16-bit machines were rarities.  I think you can expect that as 64-bit
>machines become more common that somebody will tackle the issue.  To date,
>I think you're the only one who's complained so far, but I expect we'll
>hear from others.  In the meantime, please try and take whatever pleasure
>you can from the project as it stands.
>
>-- Duane