[SJ Logo]SuperJournal Metadata Specification

Home | Search | Demo | News | Feedback | Members Only


Ann Apps, Manchester Computing, University of Manchester

SuperJournal Technical Report SJMC141

Contents:
1.  Overall Approach
2.  Dublin Core Metadata
3.  Simple HTML Metadata
4.  Resource Description Framework
5.  PDF Metadata
6.  Document Identifiers
7.  Problems and Issues
8.  Technical Appendices
9.  References

1. Overall Approach

This document describes the metadata included in SuperJournal. It also describes various types of metadata investigated during the course of the project.

Metadata is a description of an information resource. It may document the origin of, change of, or use of, data, "meta" deriving from the Greek word for "change". Within the context of the World Wide Web, metadata is used for information discovery by search engines, and so is an important element of any Web page. But metadata may also be necessary for other reasons such as cataloguing and change control of information.

The header data supplied to SuperJournal with all journal articles is effectively metadata for these articles. SuperJournal has added Dublin Core metadata to the HTML pages generated during the data conversion process (described in [SJMC140]) for articles supplied as full text SGML and for these articles' MiniContents. Metadata is also included with the SGML tables of contents files and articles' "cited by" files. All this metadata is generated automatically during the data conversion process. Metadata is not included with the SuperJournal SGML headers, even though these files are used for the article header display, because inclusion of metadata here would have had implications for changes in the SuperJournal application data loading.

In addition, metadata is included with the SuperJournal Usage Statistics HTML pages, generated during the processing of the Usage Logs (described in [SJMC260]), which are available from the "members-only" area of the SuperJournal information Web site (http://www.superjournal.ac.uk/sj/). Metadata has not been included with the information pages on the SuperJournal Web site, other than a basic "title". More effort is involved in including metadata on manually typed Web pages than on automatically generated ones.

Strictly, the inclusion of metadata with the SuperJournal HTML articles and tables of contents may have been unnecessary. The primary use for metadata headings on HTML pages is Web information discovery. The articles within SuperJournal are readable only via the SuperJournal application, following user login. But it was felt that metadata generation was a useful exercise, and it would provide possible future cataloguing of the data.

During the project, SuperJournal has investigated various other means of resource identification. In particular, several types of document identifier and PDF metadata have been considered.

2. Dublin Core Metadata

Dublin Core metadata is a developing standard for metadata specification. It is primarily concerned with the semantics of the metadata rather than the syntax used for its inclusion with an information resource. All the current information about Dublin Core may be found from the Dublin Core home page [1]. Dublin Core metadata was developed through several international workshops, and there is on-going definition through an email list (dc-general@mailbase.ac.uk). It appears that it will become the WWW standard for basic metadata. The definition of Simple Dublin Core appeared to be fairly stable, apart from discussion of some element content, but new proposals for significant change have recently been made. Qualified Dublin Core, which allows further refinement of element content, is still incompletely specified, but little use has been made of qualified Dublin Core within SuperJournal. The problem with trying to use an emerging, but not yet fully defined, standard is that created metadata may require later change.

2.1 Simple Dublin Core

Simple Dublin Core defines fifteen basic elements. Some elements have a defined content vocabulary, from which one term must be selected, others allow free text content. The Dublin Core elements are listed below in the order in which they usually appear (which is the order in which they were developed). They may also be grouped by type:

All Dublin Core elements are optional and repeatable.

2.1.1 Dublin Core Elements

The fifteen simple Dublin Core elements are as follows:

2.1.2 SuperJournal Dublin Core Metadata

The content for the Dublin Core elements used in SuperJournal are listed below. Examples of SuperJournal metadata are given in Appendix 8.1.

2.1.2.1 HTML Article Metadata

Note that this metadata concerns this HTML version of the article, not the original article.

2.1.2.2 HTML MiniContents Metadata

The metadata for an article's MiniContents is similar to that for the HTML article, except that the Title, Subject and Description fields are prepended by "MiniContents for:".

2.1.2.3 SGML Tables Of Contents Metadata

2.1.2.4 SGML Article "Cited By" Metadata

2.1.2.5 Statistics HTML Pages Metadata

2.1.3 SuperJournal Dublin Core Metadata Syntax

Dublin Core specifies the semantics of the metadata elements. The Dublin Core elements could be incorporated into any required syntax. The SuperJournal metadata is included within HTML <meta> tags in the <head> section of the HTML document. An alternative, and probably preferable, future syntax would be RDF (see Section 4), but this is a very recent development. Using HTML tags for metadata makes it potentially readable by existing search engines, though few yet understand Dublin Core. One possible problem with including these HTML <meta> tags within an HTML document is that they may not be understood by existing HTML editors like FrontPage. Some of these HTML editors delete HTML tags which they do not recognise from a document.

Examples of the SuperJournal metadata using HTML <meta> tags for the Dublin Core elements are given in Appendix 8.1. Note that the metadata "name"s used are of the form "dc.<type>", e.g. "dc.creator". The current preferred style is "DC:<Type>", e.g. "DC:Creator", but correcting existing metadata was not a viable option.

3. Simple HTML Metadata

Because current search engines probably do not recognise Dublin Core, the basic HTML metadata is also included along with the Dublin Core metadata even though this is a duplication of the information. These basic metadata tags define the title, keywords and description of the resource. They could be used instead of Dublin Core elements to provide simple metadata for Web information discovery. For example:

<head>
<title>An Example Title</title>
<meta name="keywords" content="example, demo, article">
<meta name="description" content="Example Abstract">
</head>

3.1 HTML Authoring Tools Metadata

If HTML authoring tools are used to generate HTML pages, they may include their own metadata tags. Other than the document title, these metadata elements usually define the particular authoring tool used, and possibly the content type. This would seem more related to product advertisement than Web resource discovery. The SuperJournal Web site information pages are created using FrontPage. An example of the metadata generated by FrontPage is given in Appendix 8.1.6.

It is worth noting that some HTML authoring tools will delete HTML tags, including <meta> tags which they do not understand if they are used to edit an existing file. This appears to be the case with earlier versions of FrontPage, though the most recent version is reported to be alright. So care must be taken to preserve Dublin Core metadata if an existing HTML page is updated.

4. Resource Description Framework

An alternative and probably preferable syntax for metadata would be the emerging WWW standard Resource Description Framework (RDF) [2]. RDF is specified using an XML syntax [3], and may be customised via an RDF schema [4].

RDF was not used for metadata specification in SuperJournal. It is an emerging standard still undergoing definition, and was too immature for use during the timescale of the SuperJournal project.

However, an example is given in Appendix 8.2 of a possible RDF metadata specification.

5. PDF Metadata

It is possible to add some metadata to a PDF file using the PDF Document Information fields. This consists of four basic simple text fields: Title; Subject; Author; Keywords. Inclusion of metadata in a PDF file identifies the file for possible use in applications which would extract and display it. For example a search engine could display the Article Title on a list of search hits, rather than the filename, adding to user-friendliness.

Some of the PDF files supplied to SuperJournal included PDF metadata. SuperJournal did not add metadata to any of the remaining supplied PDF files, even though this metadata was available within the corresponding SGML headers. Addition of PDF Document Information is a manual operation which involves typing the text into each separate file using Adobe Acrobat Exchange. A batch method would have been required to import the metadata from the SGML headers into the PDF articles.

It would be possible to write a batch program to add metadata to a PDF file using the Adobe Acrobat Toolkit Library provided with membership of the Adobe Developers' Association (see [SJMC140]). There may also be commercial tools available which could be employed. It is also possible using the Acrobat Toolkit Library to add further metadata fields to the PDF Document Information. These new fields would not be visible when viewing the file in Acrobat Exchange, unless a customisation plug-in was written, but they could be accessible to other software applications. In the future, it is probable that Adobe will extend the capability of PDF metadata.

6. Document Identifiers

During the course of the project, SuperJournal investigated various document identification schemes. The SuperJournal application data load captures any publisher-specific identifiers contained in the supplied SGML header files, including SICI and PII numbers, but these are not displayed to the end-user. SuperJournal defined its own unique article identifier for use in the SuperJournal application implementation.

Identifiers may be divided into "dumb" numbers which can be determined only by some look-up procedure and "intelligent" numbers, or numbers with "affordance", which can be constructed from information about a resource, are human-readable, and consequently provide some object metadata.

6.1 Digital Object Identifier

The Digital Object Identifier (DOI) [5] provides a persistent unique identification mechanism, primarily for Web publishing. It is envisaged that the actual DOI, which will probably be a user-unfriendly number, will not be displayed in documents, but will be "hidden" by a "DOI button". The DOI has potential use for electronic trading, copyright management issues and citation reference linking. The DOI consists of three parts:

Because the DOI was still at the prototype stage during the timescale of the SuperJournal project, it was thought inappropriate to incorporate DOIs into SuperJournal. Also SuperJournal was not in a position to take a proactive role in DOI set-up. Individual publishers would have had to register their own journal content. SuperJournal would have accepted, and possibly displayed with a "hot" button, DOIs in any publisher's data, but in fact no supplied data contained DOIs. These DOIs could have been for the articles themselves or for bibliographic references. Given a DOI look-up procedure, it would have been possible to add DOIs to bibliographic references in a similar way to the addition of Medline links (see [SJMC140]). But because of the immaturity of DOIs no such look-up services yet exist, and because a DOI is a "dumb" number a look-up process would be necessary.

6.2 Serial Item and Contribution Identifier (SICI)

The SICI [6] number is a mature unique identifier for serial items, which pre-dates the World Wide Web. It is a US standard and is widely adopted in serials publishing. The SICI is an intelligent identifier which can be constructed retrospectively for a serial item. Because a SICI includes ISSN, publication date, volume and issue numbers and, for version 2, page number, it is effectively providing metadata for serial items. The main disadvantage to the use of SICI as an electronic document identifier seems to be that it is specific to serials, though other similar identifiers, such as a BICI for books, are being developed.

Within SuperJournal generation of SICI numbers both for the articles themselves and for their bibliographic references to serial articles, would have been possible but was not performed.

Where a publisher provided a SICI for articles, this information was captured by the SuperJournal application but was not displayed to the end-user. It was thought that SICIs, which cannot provide a "hot" link to an article, would not provide any extra useful functionality to the end-user.

6.3 Publisher Item Identifier (PII)

The PII [7] is an essentially "dumb" identifier developed for use throughout the serials publishing process. It is generally assigned to an article on receipt, rather than publication, and used for internal production tracking. It does provide some metadata, for example it includes the ISSN, but it cannot be constructed retrospectively.

Within SuperJournal publisher supplied PIIs are captured by the application but not displayed on to the end-user, because they do not appear to add any functionality beyond article identification. It would not be possible to construct PIIs for bibliographic references because there is no central registry and numbers are assigned differently by different publishers.

During the SuperJournal data conversion process, a journal's ISSN was deduced from a supplied PII if it was not supplied separately.

Some publishers supplying data to SuperJournal used PIIs, or partial PIIs, for filenaming, probably reflecting their own production processes. In these cases it was always necessary to rename the files before inclusion in SuperJournal. SuperJournal requires article files to sort in page number order within the issue. This was never the case with "PII-type" filenames because the PII was assigned before the acceptance, or position of the article within the journal, was decided.

6.4 SuperJournal Article Identifier

Every article in SuperJournal has a unique identifier. This is used as a unique key within the object database, and is used for identification by other aspects of the SuperJournal application implementation, for example linking from a bibliographic reference to an article within SuperJournal. Thus it provides some internal SuperJournal article metadata. The SuperJournal article identifier is defined as:

<JournalIdentifier>V<Volume>I<Issue>A<ArticleNumber>

where: <JournalIdentifier> is the SuperJournal journal identifier ; <Volume> and <Issue> are the volume and issue numbers; <ArticleNumber> is a unique article number within the issue.

7. Problems and Issues

8. Technical Appendices

8.1 Example Metadata

Examples of metadata generated for SuperJournal files are given below. Generally they are for hypothetical articles with dummy titles and descriptions.

8.1.1 HTML Article Metadata

<head>
<title>An Example Title</title>
<meta name="dc.title" content="An Example Title">
<meta name="dc.creator" content="Bloggs, Fred">
<meta name="dc.creator" content="Smith, John">
<meta name="dc.subject" content="example, demo, article">
<meta name="dc.description" content="Example Abstract">
<meta name="dc.publisher" content="SuperJournal for Journal Publishing Ltd.">
<meta name="dc.date" content="(scheme=ISO 8601)1998-10-08">
<meta name="dc.type" content="Text.Serial.Journal.Article">
<meta name="dc.format" content="text/html">
<meta name="dc.identifier" content="/superj1/Journals/SJ/DEMO/V1I1/010001.html">
<meta name="dc.source" content="SuperJournal DEMO 1:1-5">
<meta name="dc.language" content="en">
<meta name="dc.rights" content="Copyright &copy; Journal Publisher Ltd. 1998">
<meta name="keywords" content="example, demo, article">
<meta name="description" content="Example Abstract">
</head>

8.1.2 HTML MiniContents Metadata

<head>
<title>MiniContents for: An Example Title</title>
<meta name="dc.title" content="MiniContents for: An Example Title">
<meta name="dc.creator" content="Bloggs, Fred">
<meta name="dc.creator" content="Smith, John">
<meta name="dc.subject" content="MiniContents for: example, demo, article">
<meta name="dc.description" content="MiniContents for: Example Abstract">
<meta name="dc.publisher" content="SuperJournal for Journal Publishing Ltd.">
<meta name="dc.date" content="(scheme=ISO 8601)1998-10-08">
<meta name="dc.type" content="Text.Index">
<meta name="dc.format" content="text/html">
<meta name="dc.identifier" content="/superj1/Journals/SJ/DEMO/V1I1/010001.minc.html">
<meta name="dc.source" content="SuperJournal DEMO 1:1-5">
<meta name="dc.language" content="en">
<meta name="dc.rights" content="Copyright &copy; Journal Publisher Ltd. 1998">
<meta name="keywords" content="MiniContents for: example, demo, article">
<meta name="description" content="MiniContents for: Example Abstract">
</head>

8.1.3 SGML Tables Of Contents Metadata

<head>
<title>Table of Contents for SuperJournal DEMO 1-1</title>
<meta name="dc.title" content="Table of Contents for SuperJournal DEMO 1-1">
<meta name="dc.creator" content="SuperJournal">
<meta name="dc.publisher" content="SuperJournal for Journal Publishing Ltd.">
<meta name="dc.date" content="(scheme=ISO 8601)1998-10-08">
<meta name="dc.type" content="Text.Index">
<meta name="dc.format" content="text/sgml">
<meta name="dc.identifier" content="/superj1/Journals/SJ/DEMO/V1I11/DEMOV1I11.toc">
<meta name="dc.source" content="SuperJournal DEMO 1-1,pp 1-149">
<meta name="dc.language" content="en">
<meta name="dc.rights" content="Copyright &copy; Journal Publisher Ltd. 1998">
</head>

8.1.4 SGML Article "Cited By" Metadata

<head>
<title>SJ/DEMO/V1I1/010001 Cited By</title>
<meta name="dc.title" content="SJ/DEMO/V1I1/010001 Cited By">
<meta name="dc.creator" content="SuperJournal">
<meta name="dc.publisher" content="SuperJournal">
<meta name="dc.date" content="(Scheme=ISO 8601)1998-09-08">
<meta name="dc.type" content="Text.Index">
<meta name="dc.format" content="text/sgml">
<meta name="dc.identifier" content="/superj1/Journals/SJ/DEMO/V1I1/010001.cit">
<meta name="dc.language" content="en">
<meta name="dc.rights" content="Copyright &copy; 1998 SuperJournal">
</head>

8.1.5 Statistics HTML Pages Metadata

The following example is the metadata for an HTML page which shows the number of registered users at the end of September 1998.

<head>
<title>SuperJournal Results: Registered Users , September 1998</title>
<meta name="dc.title" content="SuperJournal Results: Registered Users , September 1998">
<meta name="dc.creator" content="SuperJournal">
<meta name="dc.subject" content="SuperJournal Results, Registered Users">
<meta name="dc.description" content="SuperJournal Results: Registered Users , September 1998">
<meta name="dc.publisher" content="SuperJournal">
<meta name="dc.date" content="(scheme=ANSI X3.30-1985)19981006">
<meta name="dc.type" content="Data.Statistical">
<meta name="dc.format" content="text/html">
<meta name="dc.identifier" content="regu9809.htm">
<meta name="dc.language" content="ENG">
<meta name="keywords" content="SuperJournal Results, Registered Users">
<meta name="description" content="SuperJournal Results: Registered Users , September 1998">
</head>

8.1.6 SuperJournal Web Site Information Pages

The following is an example of the metadata contained in the information HTML pages on the SuperJournal Web Site which were created using FrontPage.

<head>
<title>SuperJournal Information</title>
<meta HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=windows-1252">
<meta NAME="GENERATOR" CONTENT="Microsoft FrontPage 3.0">
</head>

8.2 Example RDF Metadata

Please note that these RDF metadata examples may not be exactly correct, because the specification of RDF, and particularly RDF schemas, is not yet fixed, in addition to inexperience with RDF. They are included in this document to give a flavour of RDF use.

8.2.1 Example RDF Metadata for an HTML Article

It is preferable to keep RDF metadata in a separate file from the HTML article, possibly in a separate "rdf" directory. Within the <head> section of the HTML file would be a link to the corresponding RDF file:

<LINK rel="MetaData" src="rdf/010001.rdf" name="010001" type="text/rdf">

The RDF metadata file would contain:

<?xml:namespace ns="http://www.w3.org/TR/WD-rdf/" prefix="RDF" ?>
<!--Note that the two following URLs are hypothetical-->
<?xml:namespace ns="http://midas.ac.uk/sj/RDF/DC/" prefix="DC" ?>
<?xml:namespace ns="http://midas.ac.uk/sj/RDF/SJ/" prefix="SJ" ?>
<RDF:RDF>
<RDF:Description RDF:HREF="http://midas.ac.uk/sj/SJpdf/SJ/DEMO/V1I1/010001.html">
<DC:Title>An Example Title</DC:Title>
<SJ:Creator>
<RDF:Description>
<SJ:FNMS>Fred</SJ:FNMS>
<SJ:SNM>Bloggs</SJ:SNM>
<SJ:AFF>Manchester University</SJ:AFF>
</RDF:Description>
</SJ:Creator>
<SJ:Creator>
<RDF:Description>
<SJ:FNMS>John</SJ:FNMS>
<SJ:SNM>Smith</SJ:SNM>
</RDF:Description>
</SJ:Creator>
<DC:Subject>example, demo, article</DC:Subject>
<DC:Description>Example Abstract</DC:Description>
<DC:Publisher>SuperJournal for Journal Publishing Ltd.</DC:Publisher>
<DC:Date>1998-10-08</DC:Date>
<DC:Type>Text.Serial.Journal.Article</DC:Type>
<DC:Format>text/html</DC:Format>
<DC:Identifier>/superj1/Journals/SJ/DEMO/V1I1/010001.html</DC:Identifier>
<SJ:Source>
<RDF:Description>
<SJ:JTL>DEMO</SJ:JTL>
<SJ:VID>1</SJ:VID>
<SJ:IID>1</SJ:IID>
<SJ:PPF>1</SJ:PPF>
<SJ:PPL>5</SJ:PPL>
</RDF:Description>
</SJ:Source>
<DC:Language>en</DC:Language>
<DC:Rights>Copyright &copy; Journal Publisher Ltd. 1998</DC:Rights>
</RDF:Description>
</RDF:RDF>

8.2.2 Example RDF Schema

In the above example, the Dublin Core elements Creator and Source have been specialised to include their constituent elements. These are specified according to a SuperJournal specific RDF schema given below. Dublin Core itself will also be defined by a Dublin Core RDF schema.

<?xml:namespace ns="http://www.w3.org/TR/WD-RDF-Syntax/" prefix="RDF"?>
<?xml:namespace ns="http://www.w3.org/TR/WD-RDF-Schema/" prefix="RDFS"?>
<!--SJ:Creator-->
<RDF:RDF>
<RDFS:Class ID = "SJCreator">
<RDFS:comment>Fine-grain creator</RDFS:comment>
<RDFS:allowedPropertyType>
<RDF:PropertyType ID = "FNMS">
<RDFS:range RDF:href="http://www.w3.org/TR/WD-RDF-Syntax#String"/>
<RDFS:necessity RDF:href="http://www.w3.org/TR/WD-RDF-Schema#ZeroOrOne"/>
</RDF:PropertyType>
</RDFS:allowedPropertyType>
<RDFS:allowedPropertyType>
<RDF:PropertyType ID = "SNM">
<RDFS:range RDF:href="http://www.w3.org/TR/WD-RDF-Syntax#String"/>
<RDFS:necessity RDF:href="http://www.w3.org/TR/WD-RDF-Schema#ExactlyOne"/>
</RDF:PropertyType>
</RDFS:allowedPropertyType>
<RDFS:allowedPropertyType>
<RDF:PropertyType ID = "AFF">
<RDFS:range RDF:href="http://www.w3.org/TR/WD-RDF-Syntax#String"/>
<RDFS:necessity RDF:href="http://www.w3.org/TR/WD-RDF-Schema#ZeroOrOne"/>
</RDF:PropertyType>
</RDFS:allowedPropertyType>
</RDFS:Class>
<RDF:Description ID = "Creator">
<RDF:instanceOf RDF:href="#SJCreator"/>
<RDFS:necessity RDF:href="http://www.w3.org/TR/WD-RDF-Schema#ZeroOrMore"/>
<RDFS:Comment>The creator of the resource</RDFS:Comment>
</RDF:Description>
<RDF:Description ID = "FNMS">
<RDF:instanceOf RDF:href="http://www.w3.org/TR/WD-RDF-Syntax#PropertyType"/>
<RDFS:necessity RDF:href="http://www.w3.org/TR/WD-RDF-Schema#ZeroOrOne"/>
<RDFS:Comment>First names</RDFS:Comment>
</RDF:Description>
<RDF:Description ID = "SNM">
<RDF:instanceOf RDF:href="http://www.w3.org/TR/WD-RDF-Syntax#PropertyType"/>
<RDFS:necessity RDF:href="http://www.w3.org/TR/WD-RDF-Schema#ExactlyOne"/>
<RDFS:Comment>Surname</RDFS:Comment>
</RDF:Description>
<RDF:Description ID = "AFF">
<RDF:instanceOf RDF:href="http://www.w3.org/TR/WD-RDF-Syntax#PropertyType"/>
<RDFS:necessity RDF:href="http://www.w3.org/TR/WD-RDF-Schema#ZeroOrOne"/>
<RDFS:Comment>Address/Affiliation of Creator</RDFS:Comment>
</RDF:Description>
<!--SJ:Source-->
<RDFS:Class ID = "SJSource">
<RDFS:comment>Fine-grain source</RDFS:comment>
<RDFS:allowedPropertyType>
<RDF:PropertyType ID = "JTL">
<RDFS:range RDF:href="http://www.w3.org/TR/WD-RDF-Syntax#String"/>
<RDFS:necessity RDF:href="http://www.w3.org/TR/WD-RDF-Schema#ExactlyOne"/>
</RDF:PropertyType>
</RDFS:allowedPropertyType>
<RDFS:allowedPropertyType>
<RDF:PropertyType ID = "VID">
<RDFS:range RDF:href="http://www.w3.org/TR/WD-RDF-Syntax#String"/>
<RDFS:necessity RDF:href="http://www.w3.org/TR/WD-RDF-Schema#ExactlyOne"/>
</RDF:PropertyType>
</RDFS:allowedPropertyType>
<RDFS:allowedPropertyType>
<RDF:PropertyType ID = "IID">
<RDFS:range RDF:href="http://www.w3.org/TR/WD-RDF-Syntax#String"/>
<RDFS:necessity RDF:href="http://www.w3.org/TR/WD-RDF-Schema#ExactlyOne"/>
</RDF:PropertyType>
</RDFS:allowedPropertyType>
<RDFS:allowedPropertyType>
<RDF:PropertyType ID = "PPF">
<RDFS:range RDF:href="http://www.w3.org/TR/WD-RDF-Syntax#String"/>
<RDFS:necessity RDF:href="http://www.w3.org/TR/WD-RDF-Schema#ExactlyOne"/>
</RDF:PropertyType>
</RDFS:allowedPropertyType>
<RDFS:allowedPropertyType>
<RDF:PropertyType ID = "PPL">
<RDFS:range RDF:href="http://www.w3.org/TR/WD-RDF-Syntax#String"/>
<RDFS:necessity RDF:href="http://www.w3.org/TR/WD-RDF-Schema#ExactlyOne"/>
</RDF:PropertyType>
</RDFS:allowedPropertyType>
</RDFS:Class>
<RDF:Description ID = "Source">
<RDF:instanceOf RDF:href="#SJSource"/>
<RDFS:necessity RDF:href="http://www.w3.org/TR/WD-RDF-Schema#ExactlyOne"/>
<RDFS:Comment>Source information</RDFS:Comment>
</RDF:Description>
<RDF:Description ID = "JTL">
<RDF:instanceOf RDF:href="http://www.w3.org/TR/WD-RDF-Syntax#PropertyType"/>
<RDFS:necessity RDF:href="http://www.w3.org/TR/WD-RDF-Schema#ExactlyOne"/>
<RDFS:Comment>Journal Title</RDFS:Comment>
</RDF:Description>
<RDF:Description ID = "VID">
<RDF:instanceOf RDF:href="http://www.w3.org/TR/WD-RDF-Syntax#PropertyType"/>
<RDFS:necessity RDF:href="http://www.w3.org/TR/WD-RDF-Schema#ExactlyOne"/>
<RDFS:Comment>Volume Number</RDFS:Comment>
</RDF:Description>
<RDF:Description ID = "IID">
<RDF:instanceOf RDF:href="http://www.w3.org/TR/WD-RDF-Syntax#PropertyType"/>
<RDFS:necessity RDF:href="http://www.w3.org/TR/WD-RDF-Schema#ExactlyOne"/>
<RDFS:Comment>Issue Number</RDFS:Comment>
</RDF:Description>
<RDF:Description ID = "PPF">
<RDF:instanceOf RDF:href="http://www.w3.org/TR/WD-RDF-Syntax#PropertyType"/>
<RDFS:necessity RDF:href="http://www.w3.org/TR/WD-RDF-Schema#ExactlyOne"/>
<RDFS:Comment>First Page</RDFS:Comment>
</RDF:Description>
<RDF:Description ID = "PPL">
<RDF:instanceOf RDF:href="http://www.w3.org/TR/WD-RDF-Syntax#PropertyType"/>
<RDFS:necessity RDF:href="http://www.w3.org/TR/WD-RDF-Schema#ExactlyOne"/>
<RDFS:Comment>Last Page</RDFS:Comment>
</RDF:Description>
</RDF:RDF>

9. References

[1] Dublin Core home page: http://purl.org/dc

[2] Resource Description Framework (RDF): http://www.w3.org/RDF/

[3] Resource Description Framework Model and Syntax Specification: http://www.w3.org/TR/PR-rdf-syntax/

[4] Resource Description Framework Schemas Working Draft: http://www.w3.org/TR/WD-rdf-schema/

[5] Digital Object Identifier (DOI): http://www.doi.org

[6] Serial Item and Contribution Identifier (SICI): http://sunsite.berkeley.edu/SICI/

[7] Publisher Item Identifier (PII): http://www.elsevier.nl/homepage/about/pii/


This web site is maintained by epub@manchester.ac.uk
Last modified: July 07, 1999