Re: RSS 0.91 - skips & optional stuff

[Sean McGrath <sean@digitome.com>]

[Gary Teter]
>This seems as good a forum as any to discuss Dave Winer's proposal for RSS
>posted at <http://backend.userland.com/rss091>, so here goes....

>Overall I think it looks fine. I'm not concerned about the lack of a DTD, or
>a formal W3 spec, or whatever, because in my weird world I just have to make
>things work, and it does me no good to say "they're not following the spec."
>:-)

The DTD issue is an interesting one. I can think of many application
scenarios when I would like to blend elements with RSS semantics
into my own documents (an importation of RSS). I can also think
of some applications where I would ideally base my documents
on RSS but extend them in certain ways (an extension of RSS).

The sad fact of the matter is that XML does not support
"importation" or "extension" models well. A document is
either valid w.r.t. a DTD/schema or it isn't.  A
sort of Western European, reductionist view of the
world.

What is needed for apps. like RSS is a more
fuzzy concept of validation.
I find it amusing that this fuzzyness has been
part and parcel of HTML from
day 1 as far as I know.  That is to say, a browser,
on encountering an unknown element does not
crash and burn. It skips over it, attaching
no semantics to its element type and treating any
content as PCDATA.

I think this sort of "fuzzy validation" would be
a very useful facility in XML and would allow
the RSS DTD to be cast in a form that would
support RSS extension and RSS importation without
throwing away the very useful concept of
automatic validation that DTDs/Schemas have
to offer.

regards,





http://www.pyxie.org - an Open Source XML Processing library for Python