Stackless/microthreads merge news

[Jeff Senn]

ejr at lotus.CS.Berkeley.EDU (Edward Jason Riedy) writes:
...
>  - 'simple-unified-VM' that supported continuations (if not full
...
> Like a Virtual Virtual Machine? See http://www-sor.inria.fr/projects/vvm/ .

Hm... I'll look.

> Other people might say that the x86 is the unified VM.  It doesn't
> necessarily need to run on x86 hardware, as Transmeta and IBM
> have shown (www.transmeta.com, http://www.research.ibm.com/daisy/).
> Bochs is kinda in this space, too.

Ah -- but the x86 is very complicated (i.e. not portable and tiny) --
the Transmeta stuff is great but I heard about it many years ago --
why hasn't it taken over the world?  Too big.  What are we going to
run on the "smart-dirt"?  Maybe x86 code, but probably not.

> And yet other people might eschew virtual machines for intermediate
> representations that can be compiled quickly while keeping the

Indeed -- my goal is not performance.  It is to factor out the
complexities of software development to make it more accessible to
more people on more types of devices/platforms.  To scale from tiny
devices up to big servers.  There is something to be learned from the
apparent success of environments like Python in amplifying developer
productivity.

> And then there are those folks who have been working on a Scheme
> interpreter to serve precisely this role, Guile.

Yes.  But again I wonder why Lisp/Scheme/... never took over the world
and things like Python, the other-P, VB, and (perhaps) Java, do seem to
take hold...

> So lots of people have thought about this.  Which of these approaches
> make sense?  I'm not at all sure.  Having bytecode-like format that
> can be efficiently compiled and optimized (and recompiled and 
> reoptimized) while still being ok to run directly would be nice.
> Once upon a time, some of the Guile developers pondered a threaded
> tree coding, one that might be like crossing byte-code with the
> slim binary approach, but I don't know what became of that.  If I
> weren't a lazy bastard, I'd look into something like that myself.

In past days I've explored many of these issues myself -- some of it
in research that has never seen the light of day (so I'm either a lazy
bastard too or just have too much to do :-) ) -- my best guess is that
Python is as close as anything I've seen to being the right balance of
language features and power with user-accesibility -- the VM
implementation is quite good/simple and Chris's stackless makes it
even better -- and if only I could run the core of it on a small, raw
microprocessor with modest memory (or a SmartCard ... or even my
Palm!) I believe it would take over the world.

-- 
-Jas