[racket] Expanding frozen Racket core - limitations of universal touring machine

Hi Noel,

> -----Original Message-----
> From: Noel Welsh [mailto:noelwelsh at gmail.com]
> Sent: 17 January 2011 16:38
> To: Jukka Tuominen
> Cc: users at racket-lang.org
> Subject: Re: [racket] Expanding frozen Racket core - limitations of
> universal touring machine
>
>
> On Sat, Jan 15, 2011 at 12:16 PM, Jukka Tuominen
> <jukka.tuominen at finndesign.fi> wrote:
> > Now, imagine the read-only OS was a snapshot of
> Racket+Linux+stuff, a few
> > years ago.
> >
> > - Given the above, I wonder if there are any theoretical limits
> preventing
> > to achieve all the same functionality that are in the current or future
> > Racket versions, but only utilising the external data source? Some
> > optimising issues are obvious, but other than that?
>
> Racket has a back door via the FFI. You can get it to call arbitrary
> code generated at runtime. This code can then extend the runtime in
> arbitrary ways.
>
> N.

That sounds very good :) I wouldn't know how to use it myself, but the main
thing is that it can be done.

Thanks!
jukka