Re: lazyness

This message is in MIME format.  The first part should be readable text,
  while the remaining parts are likely unreadable without MIME-aware
tools.

--4680531-547-1200067566=:41148
Content-Type: TEXT/PLAIN; CHARSET=X-UNKNOWN
Content-Transfer-Encoding: QUOTED-PRINTABLE
Content-ID: <Pine.WNT.4.64.0801112029331.41148@[EMAIL PROTECTED]
>

On Fri, 11 Jan 2008, rossberg@[EMAIL PROTECTED]
 wrote:

> On Jan 11, 3:12=A0pm, Ingo Menger <quetzalc...@[EMAIL PROTECTED]
> wrote:
> >
> > =A0 =A0foo: jmp foo
> >
> > How would this be better?
>=20
> It does not terminate the whole process merely because one of its
> threads diverges. The remaining threads can continue to do useful
> work.
>=20

Yep. The lazy version with the same optimisation (tail call modulo a=20
slightly odd cons) will, given an arbitrary-size integer type, merely
take=
=20
longer than the likely life of the universe to blow the heap. But an
RTS=20
can catch resource issues - Jon's taking the worst case exhibited by
any=20
haskell implementation and presenting it as the expected behaviour in any=
=20
was what I was taking exception to.

--=20
flippa@[EMAIL PROTECTED]
 does not owe people jobs.
Society owes it to itself to find people jobs.
--4680531-547-1200067566=:41148--