Object-oriented serialization (Was Re: Some questions)

[Gabe Beged-Dov]

Matthew Gertner wrote:

> Let's say I have an arbitrary object structure that I want to serialize
> and send down the pipe. Serializing a bunch of object attributes in XML
> is a no-brainer, and representing arbitrary references between objects
> is also fairly trivial if something like XLink is used (and we need
> XLink, there's surely no controversy about this).

XLink is explicitly intended to support hyperlinking rather than linking, i.e. you have an
instance level title on each object reference :-!  RDF is explicitly intended to support
linking rather than hyperlinking. You can specify a title for you object reference but you do
it at the class level rather than the instance level.

Even if you try to use XLink for  OO linking you will find that you end up with the
equivalent of void* pointers.  Let's call these kinds of links properties of the source
object. RDF allows you to specify the type of the property using a URI and (using RDF Schema)
specify the base type of the property value.  This is what you would expect to be able to do
for strongly typed pointers in OO interchange.

XLink doesn't even allow you to use a namespace qualified name for the "role" (this may have
been fixed but it will be done as a new attribute value type like qname).  It certainly
doesn't touch being able to specify a type for the property value.  The XML Schema group may
end up supporting strongly typed references but I wouldn't be surprised if this fell off the
plate.

In short, RDF (and RDF Schema) support OO interchange in a pretty straightforward manner
TODAY. David Megginson's work on the DATAX toolkit shows how straightforward it can be to use
RDF.  As part of my work at Rogue Wave, I participated in the development of several
alternative C++/XML frameworks. We didn't use RDF because of the lack of tools. If I had to
do it over again and choose between RDF + RDFSchema today and XML + XLink + XMLSchema
tomorrow for OO interchange I know which way I would go.

Cordially from Corvallis,

Gabe Beged-Dov
http://www.jfinity.com/gabe


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To unsubscribe, mailto:majordomo at ic.ac.uk the following message;
unsubscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)