[
Lists Home |
Date Index |
Thread Index
]
John Cowan <cowan@mercury.ccil.org> writes:
> Roger L. Costello scripsit:
> > Hi Folks,
> >
> > I have implemented the Game of Life[1] using XSLT.
>
> And since Life is known to be Turing-complete, so is XSLT.
> Neat proof.
>
Hmm, not so sure about that: it appears to me that Roger's code exploits
round trips through XSLT that requires supporting infrastructure above
and beyond XSLT itself (ie; a browser and a server) and as such one
can't directly draw any conclusions about the properties of the XSLT
language? Yes, I am aware of the proofs of XSLT as a language being
Touring complete, it's just that I suspect that without some additional
qualification this isn't another such proof....
As an example, wouldn't one have to show that there are no additional
capabilities passed onto XSLT via external parameter interpretation
(which, for this example, might be rendered moot by showing that Rogers
code doesn't use parameters -- I haven't looked at the code).
|
