Re: Pterodactyl Gnus v0.39 is released

[Hrvoje Niksic <hniksic@srce.hr>]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> Hrvoje Niksic <hniksic@srce.hr> writes:
> 
> > I disagree with your distinction of "really compiled" vs.
> > "byte-compiled" functions.  It makes as much sense for compiled
> > functions to be called just that.  The type of the compilation is an
> > implementation detail.
> 
> I don't think so.  The difference between Lisp and byte-code is in
> many ways as big as the difference between byte-code and native code.
> For instance, you can instrument byte-compiled functions, but you
> can't instrument natively compiled code.

This statement, as it stands, is untrue for a bunch of Common Lisp
compilers.  For example, CMUCL can instrument natively compiled code,
if you use a debug compiler setting.  I was told the same is true for
ACL.  Also, it's not entirely true that byte-compiled code can't
segfault, etc.

But these things are not the point.  The point is that the crucial
difference (IMHO) is whether the code is interpreted or compiled, not
how it is compiled.  I don't buy the notion that a byte-compiled
function is somehow "less compiled" because it's not compiled to
native code.

So, the moment XEmacs developers decided to abstract away a "compiled
function" object, I believe they chose a good name for the accessor
functions.

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
there's a beast upon my shoulder / and a fiend upon my back