New brainfart for threaded VFS and data passing between threads.

Chris Pressey cpressey at catseye.mine.nu
Mon Apr 5 18:32:03 PDT 2004


On 05 Apr 2004 20:31:53 GMT
Rahul Siddharthan <rsidd at xxxxxxxxx> wrote:

> I wrote:
> >>I have a theory that Forth is more readable for someone whose native
> >>language has SOV (or OSV) word order.
> >
> >I'm not sure word order matters: lisp is often much more readable
> >than C but no human language I'm familiar with has a VSO word order.
> 
> I should take that back.  English of course has that order in the
> imperative mode, which is what one uses in programming: 
> "add 2 to 3" is like (+ 2 3)
> 
> But I'm still not convinced this has anything to do with readability.

Well, word order determines the minimum amount of punctuation required
for disambiguation.  The fact that parentheses aren't necessary with RPN
is often touted by its proponents.  I've yet to hear one explain how
this helps readability, however - and in my opinion, it hinders it.

To try to drag this back on-topic - I'm of the opinion that Forth's
great strength is bootstrapping.  It's quite easy to write a minimal
Forth (and hand-assemble it if necessary!), and then use it to improve
itself piecemeal, without needing help from a bulky toolchain etc.

Perhaps it's not a surprise, then, that the boot sequence is the place
in DragonFly where Forth is used.  I'd like to think we're exploiting it
appropriately.  I'm sure it could be used for other parts of the system,
but in those parts we wouldn't see as much of an advantage over the
alternatives.

-Chris





More information about the Kernel mailing list