On Thu, May 19, 2005 at 01:45:35PM -0400, John Cowan wrote:
Liam Quin scripsit:
Interchange of a database query language over the Web in its own
Internet Type is likely for machine execution or to interchange
files, not for reading by humans, as then text/plain might be
more appropriate... but this is conjecture on my part right now.
FWIW, I think this is a Bad Thing. Programming language content should
go in text/plain files (despite the nasty problem with the encoding
type imposed by text/*), so as to *discourage* browsers from attempting
to execute them, which is a big fat security hole.
Execution of a query in this context could better be written as
evaluation of an expression; the side-effects in XQuery are very
limited, although I agree that whenever code is executed remotely
there are some serious security concerns.
The use of text/css in HTML link elements and XML stylesheet PIs is
essentially a hack so that browsers can decide whether to fetch the
stylesheet, and is not consistent with the intention of IETF media
types, which are designed to specify a minimal mapping from raw
octets to interpretable objects such as characters or pixels.
I think this is a different case -- tect/css is a subsidiary document,
and the "type=" pseudo-attribute in a processing instruction is only
(I believe) there because the work predated widespread adoption of
XML namespaces.
Here, the XML Query document is likely to be the primary object
of transfer, not a subsidiary that applies to something else.
Unless it was by accident that I had John Cowan
offended someone, I never apologized.
jcowan(_at_)reutershealth(_dot_)com
--Quentin Crisp http://www.ccil.org/~cowan
Oh to be on the same page as Quentin Crisp, there can be
no higher honour!
Liam
--
Liam Quin, W3C XML Activity Lead, http://www.w3.org/People/Quin/
http://www.holoweb.net/~liam/