[rdfweb-dev] RDF all the way on XML.com

Jim Ley jim at jibbering.com
Fri Aug 22 13:41:32 UTC 2003


"Julian Bond" <julian_bond at voidstar.com>
> Libby Miller <Libby.Miller at bristol.ac.uk> wrote:
> >I think the underlying issue is that it's hard to encode a graph in a
> >tree syntax, since trees are a subset of graphs and not the other way
> >around.
>
> In the real world, how much RDF is there out there that is encoding a
> full graph that couldn't be expressed as an XML tree?
>
> I know it's a half way house and not a general solution, but writing
> RDF-XML so that it looks like a normal XML tree let's people use XML
> tree tools to interpret it.

The same can be said if we write XML in a certain constrained format to, so
those of us who think XML is just an extra irrelevant layer for most
applications, that complicates life, I live in a DOM2 world, so I don't have
XPath, I can't use ID's for most XML (most people don't bother
creating/including DTD's )  all I've got is getElementsByTagNameNS - the XML
tools I've got are frankly crap, the RDF tools in my platform are better
than the XML ones.  However rdfweb-dev despite its name, is not the place
for discussing these things, it divides the discussion away from people who
have more to give, and creates noise which gets in the way of foaf
discussion.

> As far as I can see this has no downside for
> people using RDF tools to interpret it. The downside is that data that
> really does need a graph layout will confuse all those people who never
> made the jump to a real RDF tool.

Exactly the XML tool people will not be able to deal with it, if you want to
live in a constrained set of FOAF _for your application_ feel free, you can
write a transformer from a general scutter easily enough (
http://jibbering.com/foaf/cache/35022e505e6a64c05837eccf4beb5d8f981a4e5a.cz )
is a constrained format I use for my application, therefore within your
application your free to work with your XML tools, the only cost is that you
need a transforming scutter, there's lots of scutters out there, and they
can all create such constrained docs with little trouble, most are on BSD or
GPL licences so you've got no trouble using them commercially.

It seems to me that you see a problem for a few tool developers in accessing
FOAF content, and rather than solving the job within those tool developers,
you're suggesting constraining the world.  I'd also appreciate it more if we
could look at more concrete examples of what cannot be done with FOAF as
RDF - there are already lots of discussion on what can't be done with FOAF
as XML.

Jim.




More information about the foaf-dev mailing list