----- Original Message -----
From: "p2psmoke" <shelleyp@...>

> Instead of working with those who want a simplified RSS,
> and showing how RDF doesn't have to overly complicate
> the specification [...]

Well, I liked your simplified proposal [1] so much that I decided to
simplify it some more, adding back sequences but in the guise of a "new"
list feature of RDF that is syntactically unobtrusive, and generally making
it easier to manage whilst remaining compatible with RDF.

The two files are in http://infomesh.net/2002/rss2/

ex1.rdf is the main proposal; here are the changes from Shelley's proposal,
which formed the basis of this file:-

* rdf:RDF is gone. You don't need it: "If the content is known to be
RDF/XML by context, such as when RDF/XML is embedded inside other XML
content, then the grammar can either start at Element Node, RDF, [...] or a
production nodeElementList" -
http://www.w3.org/TR/rdf-syntax-grammar/#start If you serve it as
application/rss+xml, you can be sure that the quote above applies, but I
contend that it applies anyway. Think pragmatically: it would actually be
possible to add a version="2.0" attribute to the root of this proposal to
indicate that it's 2.0.
* rdf:parseType="Collection" is in. Chris pointed out to me that Shelley's
proposal didn't contain sequence, so people would object. This feature of
RDF was proposed back in 2002-05; see item 11 of
http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002May/0159 It's
basically the old DAML list construct, but in the RDF namespace; it's much
like Seq, except that the items themselves make up the members, so you
don't have to keep repeating yourself. The order of appearance in the
document is the order that they appear in the list. Note that it is widely
acknowledged that the new-style DAML/RDF Lists are superior to the old
rdf:Seq lists.
* I've put the Dublin Core elements into the RSS namespace. I don't mind if
people use the DC namespace instead, but I don't see why the names can't
just be copied over to save a few bytes. The rationale that people use for
adding the Dublin Core namespace is that Dublin Core-aware agents will be
able to make great assumptions about all of your RSS feeds. That's
nonsense, of course: show me a single program (practical or theoretical)
that is any worse off for moving the elements into the RSS namespace, or
fall in line :-)
* Minor rejigging so that instead of doing <item><Item ...>...</></>, you
can now just do <items><Item/><Item/></items>.

ex1b.rdf is simply a file using the old DAML namespace for the collection
parseType, so that you can check my proposal using the RDF Validator. It
passes, of course:-
http://www.w3.org/RDF/Validator/ARPServlet?URI=http://infomesh.net/2002/rss
2/ex1b.rdf&EMBEDDED_RDF=1

There's also a graph of the output:
http://www.w3.org/RDF/Validator/ARPServlet.tmp/servlet_24159.png As PNG,
but I don't think that the Validator preserves them indefinitely, so check
it out whilst you can :-)

I've taken some notes on http://www.peerfear.org/chump/rss-dev/ It's a
shame about the lack of permalinks there...

Now, I'm about as big an RDF pedant as they come, but I'm also a
pragmatist, and I hope that I've shown convincingly that the two are not
incompatible. I think that Shelley's comments about the future of RSS have
been generally on the mark, and although I'm more of a lurker than a
long-time contributor, I hope that you'll judge this proposal on its
technical, not political, merits.

Oh, and I still prefer Aaron's RSS 3.0 proposal [2]... but that's obviously
going to fall through since it has Aaron as the supreme master of the RSS
namespace. Then again, how does that differ from 1.0? :-)

[1] http://weblog.burningbird.net/archives/000516.php
[2] http://www.aaronsw.com/2002/rss30

--
Kindest Regards,
Sean B. Palmer
@prefix : <http://purl.org/net/swn#> .
:Sean :homepage <http://purl.org/net/sbp/> .