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

Re: Paul Graham's PyCon Keynote & The Programmer's Apprentice




> Date: Mon, 31 Mar 2003 15:50:18 -0500 (EST)
> From: Eli Collins <edc206@cs.nyu.edu>
> 
> On Mon, 31 Mar 2003, Shriram Krishnamurthi wrote:
> 
> > Christopher Barber wrote:
> >
> > > Of course, Moore's law will not hold.  The rate of doubling of the
> > > amount of transisters you can fit on a chip has slowed since Moore's
> > > original observation.  It is quite possible that the growth could
> > > hit a brick wall at some point in the future.  Of course, computers
> > > could become faster for other reasons, such as increased
> > > parallelism.
> >
> > Or they could choose to not become faster but instead more useful.
> > We're seeing this with the new Intel chipset -- I forget its name, the
> > one that sounds like a brand of spaghetti.  It is somewhat clever of
> > Intel to recognize that customers notice features, not innards, and
> > therefore market it as if it were about WiFi.
> >
> 
> Anyone have more ideas about what hardware makers could do to make chips
> more useful from the PL POV? Any relevant lessons learned from Lisp
> machines?
> 
> I'm not sure how Centrino makes things more useful beyond the typical
> advances (fewer chips, lower power consumption). I agree that the WiFi bit
> is a red herring.
> 
> Eli
> 

There's an interesting recent thread on the squeak smalltalk mailing list
about how to use hardware to boost the performance of smalltalk:

http://groups.yahoo.com/group/squeak/message/58516?threaded=1

Warning!  This mailing list gives a whole new meaning to the term "high
bandwidth".  Squeakers are a very talkative lot ;-)  One example of what
hardware might be useful for is extremely fast method lookups.  In more
general terms, I can imagine that FPGAs might be used to make custom
processors for different programming languages.  Cross-platform portability
is less significant when the platform itself is mutable.  Of course, then
you have the question of how my lisp processor can talk to your java
processor...

OT: Someone mentioned that Intel's chip won't run Linux.  Why is this?

Mike