> Don't misunderstand me: mlj is excellent work if you take it
> academically, but the authors argue that it is also usable for
> practical purposes, and this is the reason why I am amuzed. The
> better conclusion from the facts in the mlj paper is that it is not
> practicable to compile ML to Java bytecode.

I agree with your conclusions, however you're perhaps slightly too
harsh with MLj: for developing applet-sized programs, for instance,
the limitations of MLj (whole-program compilation, size limitations)
are acceptable.

> - Why isn't there an O'Caml plugin for Mozilla? In many environments, you do
>   not need a sandbox, e.g. within intranets, and code signing suffices.

François Rouaix hacked together a Caml plugin for Netscape a few years
ago.  It sort of worked (under Unix and with Caml actually running in
a different process and communicating over a pipe), but he sort of
lost interest in it.  I have doubts on the cleanliness and stability
of the Netscape plugin API :-)  Also, the same technical tricks might
not work under Windows...

At any rate, I believe Web applets are a totally uninteresting
application area.  Most Web designers seem happy with JavaScript
hacks, and in many years of intense Web surfing, I came across an
interesting Java applet only once (Certicom's excellent tutorial on
elliptic curve cryptography).

> - Why isn't there a version of the bytecode interpreter that can dynamically
>   load libraries? (I know that there is a patch, but nothing
>   official.)       

Because I'm lazy.  And it's a somehow more subtle change than you think.  
(The patch you mention misses a number of issues.)

- Xavier Leroy
-------------------
To unsubscribe, mail caml-list-request@inria.fr.  Archives: http://caml.inria.fr