Introduction: Sergey Melnik

Hi all - I am honored to join the WG as an invited expert.

I am a visiting researcher in the Database Group at Stanford
University and a Ph.D. candidate in Leipzig, Germany.

I got into a liaison with RDF in the end of 1998 when I was working on
my M.S. thesis. I was building a distributed XML-based system, and I
found myself missing and reinventing many features essential for my
application: a uniform way of identifying objects in XML documents,
linking those objects, storing the relations between them in a
database, etc. Besides, I was juggling tons of navigational code
needed to traverse XML trees. At that time, RDF M&S 1.0 was nearing
completion and seemed a panacea for my needs. Thus, it became the data
model for my application, and the liaison turned into a tight bond for
years to come ;)

One of my first experiences with RDF was writing a concise algebraic
spec for RDF (which made it into my M.S. thesis) and working on an API
for RDF. Nowadays, the API is often referred to as the "Stanford RDF
API", but it was actually shaped as a community process on the RDF IG
list, with me as the code maintainer. Starting with the code base of
Janne's SiRPAC, the model and syntax aspects were separated, and the
parser itself became a replaceable module.

In the past couple of years my research was focused on
interoperability between databases and digital libraries. In
particular, I worked on mediation between Web services and on model
management, which is an approach to storing data schemas and instances
as first-class objects and manipulating them using high-level
operators. For my research, I used RDF extensively for prototyping. My
most recent industry experience includes helping a telecom startup
Last Mile Services Inc. to build and deploy an RDF-based
infrastructure for automated service provisioning.

The goal of my work in the RDFCore WG is to help making RDF accessible
and usable for the database community, and the Web community as a
whole. In the recent years, the database community has contributed
significantly to the development of formal models, tools, and products
based on XML and targeted at Web databases and services. However, for
many unfortunate reasons, researchers and developers were reluctant to
adopt RDF. One of the non-technical aspects that I would like the
group to clarify is the relationship between XML and RDF, and the
potential (pragmatic, non-ideological) benefits of using RDF in
realistic application scenarios. Speaking of XML, I am referring to
the whole suite of emerged standards including XML linking, XML
schema, query languages, and database-related efforts. Back to the
spec, I would like to see RDF designed in a layered fashion with a
clean and stable core, equipped with a flexible, well-understood
extensibility path.

I am looking forward to sharing the battlefield with you folks ;)

Best,
Sergey

-- 
E-Mail:      melnik@db.stanford.edu (Sergey Melnik)
WWW:         http://www-db.stanford.edu/~melnik
Tel:         OFFICE: 1-650-725-4312 (USA)
Address:     Room 438, Gates, Stanford University, CA 94305, USA

Received on Tuesday, 12 June 2001 15:57:28 UTC