[Shootout-list] Directions of various benchmarks
Robert Seeger
Robert Seeger <rhseeger@gmail.com>
Fri, 20 May 2005 07:58:04 -0400
------=_Part_1625_14850784.1116590284149
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
I disagree with this approach. Personally, I think benchmarks should be=20
specified in algorithms stating what needs to be done, rather than the step=
s=20
to do it. Every generally useful language (that I have seen) can have a=20
program written for it from a "what needs to be done" description.
Rob Seeger
On 5/20/05, John Skaller <skaller@users.sourceforge.net> wrote:
>=20
> On Thu, 2005-05-19 at 21:36 -0700, Brent Fulgham wrote:
> > On May 19, 2005, at 6:56 PM, John Skaller wrote:
> >
> > > On Thu, 2005-05-19 at 13:30 -0700, Brent Fulgham wrote:
> > >> --- Bengt Kleberg <bengt.kleberg@ericsson.com> wrote:
> > >
> > >> How can we judge if a declarative language meets
> > >> an imperitive specification?
> > >
> > > By the output.
> >
> > What if the output was precomputed?
>=20
> Then the test code passes. If you don't like that
> you change the benchmark to avoid the problem.
>=20
>=20
> --
> John Skaller, skaller at users.sf.net <http://users.sf.net>
> PO Box 401 Glebe, NSW 2037, Australia Ph:61-2-96600850
> Download Felix here: http://felix.sf.net
>=20
>=20
> _______________________________________________
> Shootout-list mailing list
> Shootout-list@lists.alioth.debian.org
> http://lists.alioth.debian.org/mailman/listinfo/shootout-list
>
------=_Part_1625_14850784.1116590284149
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
I disagree with this approach. Personally, I think benchmarks should be
specified in algorithms stating what needs to be done, rather than the
steps to do it. Every generally useful language (that I have seen) can
have a program written for it from a "what needs to be done"
description.<br>
<br>
Rob Seeger<br><br><div><span class=3D"gmail_quote">On 5/20/05, <b class=3D"=
gmail_sendername">John Skaller</b> <<a href=3D"mailto:skaller@users.sour=
ceforge.net">skaller@users.sourceforge.net</a>> wrote:</span><blockquote=
class=3D"gmail_quote" style=3D"border-left: 1px solid rgb(204, 204, 204); =
margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
On Thu, 2005-05-19 at 21:36 -0700, Brent Fulgham wrote:<br>> On May 19, =
2005, at 6:56 PM, John Skaller wrote:<br>><br>> > On Thu, 2005-05-=
19 at 13:30 -0700, Brent Fulgham wrote:<br>> >> --- Bengt Kleberg =
<
<a href=3D"mailto:bengt.kleberg@ericsson.com">bengt.kleberg@ericsson.com</a=
>> wrote:<br>> ><br>> >> How can we judge if a declarativ=
e language meets<br>> >> an imperitive specification?<br>> >
<br>> > By the output.<br>><br>> What if the output was precomp=
uted?<br><br>Then the test code passes. If you don't like that<br>you chang=
e the benchmark to avoid the problem.<br><br><br>--<br>John Skaller, skalle=
r at=20
<a href=3D"http://users.sf.net">users.sf.net</a><br>PO Box 401 Glebe, NSW 2=
037, Australia Ph:61-2-96600850<br>Download Felix here: <a href=3D"http://f=
elix.sf.net">http://felix.sf.net</a><br><br><br>___________________________=
____________________
<br>Shootout-list mailing list<br><a href=3D"mailto:Shootout-list@lists.ali=
oth.debian.org">Shootout-list@lists.alioth.debian.org</a><br><a href=3D"htt=
p://lists.alioth.debian.org/mailman/listinfo/shootout-list">http://lists.al=
ioth.debian.org/mailman/listinfo/shootout-list
</a><br></blockquote></div><br>
------=_Part_1625_14850784.1116590284149--