I was recently using Google Scholar to try and find out more about the impact of my peer-reviewed publications.  Initially I was looking at papers published since 2004, but I then thought it would be interesting to see how far back the citation data might go. 

So I used Google Scholar to find out about links to my paper on The Evolution of Web Protocols which was published in the Journal of Documentation in 1999 (Vol. 55, No. 1 January 1999, pp. 71-81).

I discovered two citations to this paper: one in course material for a course on Organization of Information written by the School of Library and Information Studies at The University of Alabama and, much more interestingly, one in a US Patent claim!  The title of the patent is “System and method for discovering information about web resources “. And, as can be seen from the Google Patent Search, the patent was filed in February 2002 and issued in August 2007, with the assignee being Microsoft Corporation! 

The first part of the patent states that the claim is based on:

A computer-implemented method for identifying metadata about a first resource identified by a first Uniform Resource Identifier (“URI”), the method comprising:

issuing a request for the first resource identified by the first URI;

receiving a response document from the first URI;

parsing the response document received in response to the issued request, wherein the response document includes a second URI for accessing a second resource, wherein the response document includes an indication that metadata about the first resource exists on the second resource, wherein the indication indicates a metadata format;

generating a request to retrieve the metadata from the second resource, wherein the generated request is formatted to support the metadata format identified by the indication; and

retrieving the metadata from the second resource.

The patent goes on to describe how this will be implemented:

The computer-implemented method of claim 1, wherein the response document comprises an HTML document and the indication comprises a LINK tag.

 Yes, the patent is based on use of the HTML LINK tag to link to a metadata description.

As my colleague Pete Cliff has pointed out to me;

OAI-ORE says you can include a resource map (which describes the agreggation of resources that make up (for example) a document – an article in the form of a Web page that includes images say)

<link rel="resourcemap" href="http://www.ukoln.ac.uk/importantdoc/map.xml" mce_href="http://www.ukoln.ac.uk/importantdoc/map.xml" />

The resource map is metadata. Does this mean that doing this now will require paying a fee to Microsoft?

How can this patent claim have been granted? And why was my paper cited in the patent?

Looking back at my paper I find that I stated that:

Metadata can be described as the missing architectural component of the web.

I went on to say that:

Work in this area included Netscape’s proposal on “Meta Content Framework Using XML” [32] which provides a specification for describing information structures (metadata) for collections of networked information using XML and Microsoft’s “Web Collections using XML” [33] proposal for providing a metadata framework which can be used for a variety of applications, such as sitemaps, distributed authoring and content labelling.

Both of these proposals recognised the importance of XML for representing the syntax of the metadata. The proposals, together with other related work, led to the development of RDF, the Resource Description Framework, which provides a framework for metadata giving interoperability between applications that exchange machine-readable information on the Web [34].

At the time of writing (July 1998) work in developing RDF is still at an early stage. However RDF does seem to provide a mechanism for pulling together the various related metadata components and adding a new architectural component to the Web.

It seems the patent claim cites my work as evidence that use of the <LINK> tag to embed metadata was not envisaged back in 1998. However my paper was never intended to do provide a complete description of the architecture of Web. And I am sure that there will be examples of use of the <LINK> tag for this purpose prior to the submission of this patent in 2002.

My paper clearly has had an impact which I hadn’t expected! However rather than flaming me for helping Microsoft to patent use of metadata in Web pages 🙂 I’d much rather the readers of this blog provided examples of prior art and suggested ways in which nthis patent can be overturned.