W3C home > Mailing lists > Public > public-lod@w3.org > March 2010

RE: RDF Serializations

From: Rob Vesse <rav08r@ecs.soton.ac.uk>
Date: Fri, 12 Mar 2010 10:41:33 -0000
To: <nathan@webr3.org>, "'Linked Data community'" <public-lod@w3.org>
Message-ID: <EMEW3|e5b11da556bf20ffaff1cfcbe251a9cbm2BAff06rav08r|ecs.soton.ac.uk|000d01cac1d0$95573840$c005a8c0$@soton.ac.uk>
> Hi All,
>
> I've been putting some thought in to RDF Serializations in the context
> of linked data; and ever increasingly I'm questioning why I feel the
> need to offer "the same" RDF graphs serialized in different formats.
>
> I guess a specific questions would be, does anybody operate a linked
> data consuming library that doesn't support a particular serialization?

Yes I expect they do because unfortunately RDF/XML is still the only
officially endorsed W3C standard despite the plethora of other formats - not
every library supports every serialization and then you have the issue of
RDF embedded/implied in other formats - RDFa, micro formats, RSS, GRDDL -
where support is more patchy.

> I'm starting to see this more as a processing / computation load
> transfer between server and client, where most clients could easily
> convert the serialization from one format to another, but prefer to
> offload this to the server where possible.

A good library should be able to do the transformation efficiently whether
at the client/server end, unless you're serving massive RDF dumps where this
is infeasible/ill-advised for the server then there's no reason not to offer
multiple formats

> What I'm gunning for in the end, is to only expose all linked data / rdf
> as static RDF+XML documents within my application - would this in any
> way make the data "less" linked because some clients don't support
> RDF+XML or could I take it for granted that everybody (for instance
> everybody on this list) could handle this serialization.

Yes most clients would support RDF/XML as it's the only official W3C
standard but part of the ethos of the whole LOD movement is that the data
should be as open as possible - restricting it to one format limits the
openness of the data to some degree.  Personally from the point of view of
someone who both consumes Linked Data and writes parsers and serializers for
RDF I'd prefer to get my RDF in a format other than RDF/XML such as Turtle
as other formats are typically far easier (and faster) to parse.

Having it available in formats other than RDF/XML also allows for easy
scripting - someone could quite easily write a script to grab RDF in
NTriples format from some URI and then dump the Triples to the screen
without having to use a full blown RDF library whereas it's just not
possible if you're stuck with getting RDF/XML

I guess in answer to your question it doesn't make the data less linked but
it makes it less accessible i.e. open

> Any other comments or thoughts people may have on this topic are more
> than welcome.
>
> Many Regards,
>
> Nathan


Rob Vesse

PhD Student
IAM Group
Bay 20, Room 4027, Building 32
Electronics & Computer Science
University of Southampton
SO17 1BJ
Received on Friday, 12 March 2010 10:42:12 UTC

This archive was generated by hypermail 2.3.1 : Sunday, 31 March 2013 14:24:25 UTC