Open Archives Initiative Object Reuse and Exchange |
DO NOT USE THIS SPECIFICATION, see instead the CURRENT ORE SPECIFICATIONS.
This document was part of an alpha release and has been superseded.
Open Archives Initiative Object Reuse and Exchange (OAI-ORE) defines standards for the description and exchange of aggregations of Web resources. OAI-ORE introduces the notion of a Resource Map, which is a specialization of a named graph that asserts a finite set of resources (the Aggregated Resources), their types, intra-relationships, and relationships with resources external to this finite set (the external resources). A Resource Map Document is a machine-readable representation of a Resource Map. Although multiple serializations of Resource Maps are possible, the initial serialization is done using the Atom Syndication Format [RFC4287]. A detailed examination of the mapping of ORE Data Model concepts to the Atom Syndication Format is known as the Resource Map Profile of Atom [ReMProfileofAtom]. The purpose of this document is to provide guidance to application developers on how to implement and interpret the Resource Map Profile of Atom. This user guide is one of several documents comprising the OAI-ORE specification and user guide.
1. Introduction
1.1 Notational Conventions
2. Populating Elements
2.1 /feed/id and /feed/entry/id
2.2 /feed/link[@rel="self"] and /feed/entry/link[@rel="alternate"]
2.3 /feed/title and /feed/entry/title
2.4 /feed/updated and /feed/entry/updated
2.5 GRDDL crosswalk from Atom XML to RDF/XML
3. Providing Metadata for the Aggregation
4. Common Scenarios
4.1 Multiple Formats
4.2 Mirror Copies
4.3 Versions
4.4 Splash Pages
5. Linking to Other Aggregations
5.1 /feed/entry/source
5.2 /feed/entry/link[@rel="via"]/@href
6. References
A. Acknowledgments
B. Change Log
Open Archives Initiative Object Reuse and Exchange (OAI-ORE) defines standards for the description and exchange of aggregations of Web resources. OAI-ORE introduces the notion of a Resource Map, which is a specialization of a named graph that asserts a finite set of resources (the Aggregated Resources), their types, intra-relationships, and relationships with resources external to this finite set (the external resources). A Resource Map Document is a machine-readable representation of a Resource Map. Although multiple serializations of Resource Maps are possible, the initial serialization is done using the Atom Syndication Format [RFC4287]. A detailed examination of the mapping of ORE Data Model concepts to the Atom Syndication Format is known as the Resource Map Profile of Atom [ReMProfileofAtom]. The purpose of this document is to provide guidance to application developers on how to implement and interpret the Resource Map Profile of Atom. We anticipate that the contents and guidance given in this document to evolve over time to reflect the best practices of the community.
This document does not address ReM validation. Readers interested in validation are referred to Appendix C of [ReMProfileofAtom] for a Schematron Schema of the Resource Map Profile of Atom.
The keywords "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119 [IETF RFC 2119].
The Atom Syndication Format defines many elements and the Resource Map Profile of Atom [ReMProfileofAtom] defines which elements have special meaning when a ReM is serialized as an Atom feed. In this section, the most common and potentially confusing elements are examined and best practices are put forward regarding their population. Throughout document, we will use the following arXiv eprint as the defining example:
http://arxiv.org/abs/astro-ph/0601007
This is an interesting example because it has multiple identifiers, versions, formats, mirrors and external relationships. Although arXiv does not yet support ReMs, we will assume in the examples below a hypothetical ReM available at the following URI ("URI-R" from [ORE Model]):
http://arxiv.org/rem/astro-ph/0601007
The most important thing to address are the nature of
the identifiers of Atom feeds and entries. To begin with,
there is a conceptual difference between identifiers in Atom
(/feed/id
and /feed/entry/id
) and
the protocol-based URIs found in the /feed/link
and /feed/entry/link
elements in Atom:
/feed/link
and
/feed/entry/link
elements in a Resource Map
is explained in the Resource Map Profile of Atom [ReMProfileofAtom]. For
example, the href attribute of a /feed/entry/link
with a rel attribute equal to "alternate" points at
an Aggregated Resource. The URI that is used as the value
the href attribute of the Atom /feed/link
and /feed/entry/link
elements must be
protocol-based. /feed/id
and
/feed/entry/id
) have no meaning in the ORE Model,
but must be provided to make Resource Maps that are valid Atom
feeds. These Atom identifiers can be protocol-based URIs, or
they can be non-protocol-based URIs (i.e., the URI scheme
is not dereferencable). In the examples, to maintain
this distinction, we will use non-protocol-based URIs for
Atom identifiers. Good candidates for non-protocol-based
URIs include uuids [RFC4122],
tag URIs [RFC4151], and
info URIs [RFC4452].
The tag and info schemes are good for conveying semantics
in the identifier and uuids are good for opaque identifiers.
Two considerations regarding the minting of these identifiers
are important:
In summary, the URIs in /feed/entry/link
elements
identify the resource that is being aggregated and do not
change based on who is doing the aggregating. The URIs in
/feed/entry/id
elements do change based on who is
doing the aggregating.
For the arXiv eprint described above,
we will assume just 3 Aggregated Resources
(http://arxiv.org/ps/astro-ph/0601007
;
http://arxiv.org/pdf/astro-ph/0601007
;
http://arxiv.org/e-print/astro-ph/0601007
) and
that the ReM is being authored by arXiv itself, thus putting
the burden of minting Atom identifiers on the arXiv itself .
If a third party were to create a Resource Map for the arXiv
eprint, it would have to mint the Atom identifiers. In both
cases, however, the Aggregated Resources would be the same 3
ones. The skeleton of the ReM is shown below (note the
necessary category
element that declares this Atom
Feed Document conforms to the Resource Map Profile of Atom and the
link
element that establishes the "describes" relationship
between this ReM and the Aggregation the ReM describes):
<?xml version="1.0" encoding="utf-8"?> <feed xmlns="http://www.w3.org/2005/Atom"> <id>tag:arxiv.org,2007:astro-ph/0601007v2</id> <category scheme="http://www.openarchives.org/ore/terms/" term="http://www.openarchives.org/ore/terms/ResourceMap" label="Resource Map" /> <link rel="describes" href="http://arxiv.org/rem/astro-ph/0601007#aggregation" /> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:pdf</id> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:ps</id> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:e-print</id> </entry> </feed>
As can be seen, arXiv minted Atom identifiers in the tag URI namespace, and using metadata that pertains to each of the Aggregated Resources of the ReM. It should be noted that for each Aggregated Resource of a ReM, there is a unique combination of values that could be used for the creation of an Atom entry identifier, namely the URI of the ReM itself and the URI of the Aggregated Resource conveyed by the Atom entry. This combination could be used as the seed to achieve a unique re-computable identifier, for example by concatenating the URIs and creating a hash over them.
Now that Atom identifiers are populated, we turn to the
values of the link elements. The ReM asserts its own URI
by means of /feed/link[@rel="self"]
, and it
asserts the URIs of the Aggregated Resources by means of
/feed/entry/link[@rel="alternate"]
, for the remainder
of this section referred to as the feed link and the entry
links, respectively. The feed link, like the Atom
identifiers described above is relative to
the author of the ReM. In particular, it gives the URI where
this particular ReM can be found.
Unlike the feed link and Atom identifiers described above which would change based on the author of the ReM (i.e., 1st or 3rd party ReMs), the URIs in the entry links do not change. They are the Aggregated Resources, regardless of who is doing the aggregation.
Expanding our skeleton ReM, we now have:
<?xml version="1.0" encoding="utf-8"?> <feed xmlns="http://www.w3.org/2005/Atom"> <id>tag:arxiv.org,2007:astro-ph/0601007v2</id> <link href="http://arxiv.org/rem/astro-ph/0601007" rel="self" type="application/atom+xml"/> <category scheme="http://www.openarchives.org/ore/terms/" term="http://www.openarchives.org/ore/terms/ResourceMap" label="Resource Map" /> <link rel="describes" href="http://arxiv.org/rem/astro-ph/0601007#aggregation" /> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:ps</id> <link href="http://arxiv.org/ps/astro-ph/0601007" rel="alternate" type="application/postscript"/> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:pdf</id> <link href="http://arxiv.org/pdf/astro-ph/0601007" rel="alternate" type="application/pdf"/> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:e-print</id> <link href="http://arxiv.org/e-print/astro-ph/0601007" rel="alternate"/> </entry> </feed>
The values for /feed/author
, /feed/title
and /feed/entry/title
are not necessarily
what one would first expect. These elements are not
for encoding metadata about the the Atom feed (ReM) and
Atom entries. This means the /feed/title
is
not "Parametrization of K-essence and Its Kinetic Term",
and similarly the /feed/author/name
is not
"Hui Li, Zong-Kuan Guo, Yuan-Zhong Zhang". The section
below addresses how to
incorporate metadata about the Aggregation into the Resource
Map Profile of Atom. We recommend listing the URI, prepended
with the literal "Resource Map" or "Aggregated Resource" for
/feed/title
and /feed/entry/title
,
respectively. This gives meaningful values for the
titles, but at the same time underscores that they are
titles and not link elements. In this example, the author
of the ReM is the arXiv repository itself. Note that the
/feed/entry/author element
MUST NOT be used in ReMs.
Authorship is inherited from the /feed/author
element or from the /feed/entry/source/author
element if that exists. See section
4.2.1 of [ReMProfileofAtom] for a
detailed explanation.
<?xml version="1.0" encoding="utf-8"?> <feed xmlns="http://www.w3.org/2005/Atom"> <id>tag:arxiv.org,2007:astro-ph/0601007v2</id> <link href="http://arxiv.org/rem/astro-ph/0601007" rel="self" type="application/atom+xml"/> <category scheme="http://www.openarchives.org/ore/terms/" term="http://www.openarchives.org/ore/terms/ResourceMap" label="Resource Map" /> <link rel="describes" href="http://arxiv.org/rem/astro-ph/0601007#aggregation" /> <title>Resource Map http://arxiv.org/rem/astro-ph/0601007</title> <author> <name>arXiv.org e-Print Repository</name> <uri>http://arxiv.org/</uri> <email>www-admin@arxiv.org</email> </author> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:ps</id> <link href="http://arxiv.org/ps/astro-ph/0601007" rel="alternate" type="application/postscript"/> <title>Aggregated Resource http://arxiv.org/ps/astro-ph/0601007</title> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:pdf</id> <link href="http://arxiv.org/pdf/astro-ph/0601007" rel="alternate" type="application/pdf"/> <title>Aggregated Resource http://arxiv.org/pdf/astro-ph/0601007</title> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:e-print</id> <link href="http://arxiv.org/e-print/astro-ph/0601007" rel="alternate"/> <title>Aggregated Resource http://arxiv.org/e-print/astro-ph/0601007</title> </entry> </feed>
Whereas RFC-4287 allows updated
to be changed
"when an entry or feed was modified in a way the publisher
considers significant", the Resource Map Profile of Atom
is more strict. The /feed/updated
value SHOULD
match the "Last-Modified" HTTP response header [RFC2616], if that header exists. If
the "Last-Modified" header does not exist (which can happen when a
resource is dynamically generated), the /feed/updated
MAY be the value in the "Date" HTTP response header (the date
the HTTP response was generated). However, in this case of the
"Last-Modified" HTTP response header not being available, it is
RECOMMENDED that the ReM publisher keep track of what the correct
value of /feed/updated
should be and strictly update
its value based on any modification, significant or not. See
section 4.2.15 of [ReMProfileofAtom]
for a detailed explanation.
Populating /feed/entry/updated
is more subtle. Even
in the case of 1st party ReMs (such as the arXiv example explored here),
/feed/entry/updated
is not "Last-Modified" HTTP response header
of the Aggregated Resource, but rather the last time the Aggregation
described by the ReM made an assertion about the Aggregated Resource.
In other words, /feed/entry/updated
is the datestamp of
that particular Atom entry, not the Aggregated Resource described by that
entry.
Finally, the value of /feed/updated
MUST
always be equal to or later in time than the values of all
/feed/entry/updated
elements.
In our example ReM, we have all the /feed/entry/updated
values
being the same (in the case of arXiv, all the resources were aggregated
at the same time), but we show a different datestamp for /feed/updated
to correspond with an edit of the ReM.
<?xml version="1.0" encoding="utf-8"?> <feed xmlns="http://www.w3.org/2005/Atom"> <id>tag:arxiv.org,2007:astro-ph/0601007v2</id> <link href="http://arxiv.org/rem/astro-ph/0601007" rel="self" type="application/atom+xml"/> <category scheme="http://www.openarchives.org/ore/terms/" term="http://www.openarchives.org/ore/terms/ResourceMap" label="Resource Map" /> <link rel="describes" href="http://arxiv.org/rem/astro-ph/0601007#aggregation" /> <title>Resource Map http://arxiv.org/rem/astro-ph/0601007</title> <author> <name>arXiv.org e-Print Repository</name> <uri>http://arxiv.org/</uri> <email>www-admin@arxiv.org</email> </author> <updated>2007-10-10T18:30:02Z</updated> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:ps</id> <link href="http://arxiv.org/ps/astro-ph/0601007" rel="alternate" type="application/postscript"/> <title>Aggregated Resource http://arxiv.org/ps/astro-ph/0601007</title> <updated>2006-05-31T12:52:00Z</updated> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:pdf</id> <link href="http://arxiv.org/pdf/astro-ph/0601007" rel="alternate" type="application/pdf"/> <title>Aggregated Resource http://arxiv.org/pdf/astro-ph/0601007</title> <updated>2006-05-31T12:52:00Z</updated> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:e-print</id> <link href="http://arxiv.org/e-print/astro-ph/0601007" rel="alternate"/> <title>Aggregated Resource http://arxiv.org/e-print/astro-ph/0601007</title> <updated>2006-05-31T12:52:00Z</updated> </entry> </feed>
The up-to-date version of the GRDDL-compliant
XSLT to transform an Atom-based serialization of a
Resource Map into a serialization in RDF/XML is available at http://www.openarchives.org/ore/atom-grddl.xsl.
In order to allow Atom-based Resource Maps to be readily
transformed in this manner, it is RECOMMENDED to add GRDDL
namespace and transformation information as attributes to the
/feed
element.
<?xml version="1.0" encoding="utf-8"?> <feed xmlns="http://www.w3.org/2005/Atom" xmlns:grddl="http://www.w3.org/2003/g/data-view#" grddl:transformation="http://www.openarchives.org/ore/atom-grddl.xsl"> <id>tag:arxiv.org,2007:astro-ph/0601007v2</id> <link href="http://arxiv.org/rem/astro-ph/0601007" rel="self" type="application/atom+xml"/> <category scheme="http://www.openarchives.org/ore/terms/" term="http://www.openarchives.org/ore/terms/ResourceMap" label="Resource Map" /> <link rel="describes" href="http://arxiv.org/rem/astro-ph/0601007#aggregation" /> <title>Resource Map http://arxiv.org/rem/astro-ph/0601007</title> <author> <name>arXiv.org e-Print Repository</name> <uri>http://arxiv.org/</uri> <email>www-admin@arxiv.org</email> </author> <updated>2007-10-10T18:30:02Z</updated> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:ps</id> <link href="http://arxiv.org/ps/astro-ph/0601007" rel="alternate" type="application/postscript"/> <title>Aggregated Resource http://arxiv.org/ps/astro-ph/0601007</title> <updated>2006-05-31T12:52:00Z</updated> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:pdf</id> <link href="http://arxiv.org/pdf/astro-ph/0601007" rel="alternate" type="application/pdf"/> <title>Aggregated Resource http://arxiv.org/pdf/astro-ph/0601007</title> <updated>2006-05-31T12:52:00Z</updated> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:e-print</id> <link href="http://arxiv.org/e-print/astro-ph/0601007" rel="alternate"/> <title>Aggregated Resource http://arxiv.org/e-print/astro-ph/0601007</title> <updated>2006-05-31T12:52:00Z</updated> </entry> </feed>
With few exceptions, typical bibliographic metadata about either the Aggregation or the Aggregated Resources is conveyed with elements outside the Atom namespace. In this section we look out how to handle common metadata requirements.
We show three ways to add metadata to and about the Aggregation.
First, the aggregation can have additional identifiers
such as DOIs or Handles. The arXiv example has an info URI
associated with the aggregation and we encode it with the
feed/link[@rel="related"]
element. This element
can be repeated for additional identifiers as well. Second,
we provide some minimal bibliographic metadata (title, creator)
by importing from another namespace, in this case from the Dublin
Core namespace. To facilitate conversion to RDF, we encourage the
use URIs instead of literal values (e.g., "Zong-Kuan Guo"), but we
acknowledge that often these URIs are unavailable. As described
in section 2.3, the /feed/title
and /feed/author
elements MUST NOT be used for
this purpose. Finally, we add a fourth Aggregated Resource,
which is the OAI-PMH metadata available from the arXiv E-print
repository. Although not required, in this example, we modify
the /feed/entry/title
to add a human-readable hint
that this Aggregated Resource is metadata. To unambiguously mark
it as bibliographic metadata, we import an element from the RDF
namespace (although in the example below, it uses an Info URI
namespace (eu-repo) that is pending registration). We also show
a different /feed/entry/updated
value.
It is up to the author of the ReM to decide what best approach for including Aggregation metadata: importing from other namespaces, aggregating resources that are metadata, or some combination of both.
Providing bibliographic metadata for the Aggregated Resources
is done similarly, but the elements in the imported namespaces
are enclosed in the appropriate /feed/entry
elements.
In this example, we annotate the last Aggregated Resource with
a free text description that the Aggregated Resource is LaTeX source
files.
<?xml version="1.0" encoding="utf-8"?> <feed xmlns="http://www.w3.org/2005/Atom" xmlns:grddl="http://www.w3.org/2003/g/data-view#" grddl:transformation="http://www.openarchives.org/ore/atom-grddl.xsl" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:dc="http://purl.org/dc/elements/1.1/"> <id>tag:arxiv.org,2007:astro-ph/0601007v2</id> <link href="http://arxiv.org/rem/astro-ph/0601007" rel="self" type="application/atom+xml"/> <category scheme="http://www.openarchives.org/ore/terms/" term="http://www.openarchives.org/ore/terms/ResourceMap" label="Resource Map" /> <link rel="describes" href="http://arxiv.org/rem/astro-ph/0601007#aggregation" /> <title>Resource Map http://arxiv.org/rem/astro-ph/0601007</title> <author> <name>arXiv.org e-Print Repository</name> <uri>http://arxiv.org/</uri> <email>www-admin@arxiv.org</email> </author> <updated>2007-10-10T18:30:02Z</updated> <link href="info:arxiv/astro-ph/0601007v2" rel="related"/> <dc:title>Parametrization of K-essence and Its Kinetic Term</dc:title> <dc:creator>Hui Li</dc:creator> <dc:creator>Zong-Kuan Guo</dc:creator> <dc:creator>Yuan-Zhong Zhang</dc:creator> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:oai-opmh</id> <link href="http://export.arxiv.org/oai2?verb=GetRecord&metadataPrefix=oai_dc&identifier=astro-ph/0601007" rel="alternate"/> <title>Aggregated Resource Dublin Core Metadata</title> <rdf:type>info:eu-repo/semantics/DescriptiveMetadata</rdf:type> <updated>2006-09-27T00:00:00Z</updated> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:ps</id> <link href="http://arxiv.org/ps/astro-ph/0601007" rel="alternate" type="application/postscript"/> <title>Aggregated Resource http://arxiv.org/ps/astro-ph/0601007</title> <updated>2006-05-31T12:52:00Z</updated> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:pdf</id> <link href="http://arxiv.org/pdf/astro-ph/0601007" rel="alternate" type="application/pdf"/> <title>Aggregated Resource http://arxiv.org/pdf/astro-ph/0601007</title> <updated>2006-05-31T12:52:00Z</updated> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:e-print</id> <link href="http://arxiv.org/e-print/astro-ph/0601007" rel="alternate"/> <title>Aggregated Resource http://arxiv.org/e-print/astro-ph/0601007</title> <dc:description>LaTeX Source Files</dc:description> <updated>2006-05-31T12:52:00Z</updated> </entry> </feed>
In this section we review the appropriate techniques within the Resource Map Profile of Atom for handling common scenarios for Aggregations.
In the example we have been working with, the PostScript, PDF and
the page containing links to other formats are all considered
separate Aggregated Resources and not simply the same Aggregated
Resource realized in multiple MIME types. This differs from the
typical Atom Syndication Format idiom, where multiple MIME types
are often grouped in the same entry
element.
We can also express that Aggregated Resources share a format
relationship. In this example, we link the PostScript and
PDF formats together by placing hasFormat
elements
(from the dcterms
namespace) in each Atom entry,
referencing each other.
<?xml version="1.0" encoding="utf-8"?> <feed xmlns="http://www.w3.org/2005/Atom" xmlns:grddl="http://www.w3.org/2003/g/data-view#" grddl:transformation="http://www.openarchives.org/ore/atom-grddl.xsl" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:dcterms="http://purl.org/dc/terms/"> <id>tag:arxiv.org,2007:astro-ph/0601007v2</id> <link href="http://arxiv.org/rem/astro-ph/0601007" rel="self" type="application/atom+xml"/> <category scheme="http://www.openarchives.org/ore/terms/" term="http://www.openarchives.org/ore/terms/ResourceMap" label="Resource Map" /> <link rel="describes" href="http://arxiv.org/rem/astro-ph/0601007#aggregation" /> <title>Resource Map http://arxiv.org/rem/astro-ph/0601007</title> <author> <name>arXiv.org e-Print Repository</name> <uri>http://arxiv.org/</uri> <email>www-admin@arxiv.org</email> </author> <updated>2007-10-10T18:30:02Z</updated> <link href="info:arxiv/astro-ph/0601007v2" rel="related"/> <dc:title>Parametrization of K-essence and Its Kinetic Term</dc:title> <dc:creator>Hui Li</dc:creator> <dc:creator>Zong-Kuan Guo</dc:creator> <dc:creator>Yuan-Zhong Zhang</dc:creator> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:oai-opmh</id> <link href="http://export.arxiv.org/oai2?verb=GetRecord&metadataPrefix=oai_dc&identifier=astro-ph/0601007" rel="alternate"/> <title>Aggregated Resource Dublin Core Metadata</title> <rdf:type>info:eu-repo/semantics/DescriptiveMetadata</rdf:type> <updated>2006-09-27T00:00:00Z</updated> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:ps</id> <link href="http://arxiv.org/ps/astro-ph/0601007" rel="alternate" type="application/postscript"/> <title>Aggregated Resource http://arxiv.org/ps/astro-ph/0601007</title> <updated>2006-05-31T12:52:00Z</updated> <dcterms:hasFormat>http://arxiv.org/pdf/astro-ph/0601007v1</dcterms:hasFormat> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:pdf</id> <link href="http://arxiv.org/pdf/astro-ph/0601007" rel="alternate" type="application/pdf"/> <title>Aggregated Resource http://arxiv.org/pdf/astro-ph/0601007</title> <dcterms:hasFormat>http://arxiv.org/ps/astro-ph/0601007v1</dcterms:hasFormat> <updated>2006-05-31T12:52:00Z</updated> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:e-print</id> <link href="http://arxiv.org/e-print/astro-ph/0601007" rel="alternate"/> <title>Aggregated Resource http://arxiv.org/e-print/astro-ph/0601007</title> <dc:description>LaTeX Source Files</dc:description> <updated>2006-05-31T12:52:00Z</updated> </entry> </feed>
arXiv maintains many mirrors throughout the world. For example, the the same e-print can be accessed at the mirror repository in Japan with this URI:
http://jp.arxiv.org/abs/astro-ph/0601007
We treat this the same as adding an additional
identifier (such as the arXiv info URI) and use the
feed/link[@rel="related"]
element.
<?xml version="1.0" encoding="utf-8"?> <feed xmlns="http://www.w3.org/2005/Atom" xmlns:grddl="http://www.w3.org/2003/g/data-view#" grddl:transformation="http://www.openarchives.org/ore/atom-grddl.xsl" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:dcterms="http://purl.org/dc/terms/"> <id>tag:arxiv.org,2007:astro-ph/0601007v2</id> <link href="http://arxiv.org/rem/astro-ph/0601007" rel="self" type="application/atom+xml"/> <category scheme="http://www.openarchives.org/ore/terms/" term="http://www.openarchives.org/ore/terms/ResourceMap" label="Resource Map" /> <link rel="describes" href="http://arxiv.org/rem/astro-ph/0601007#aggregation" /> <title>Resource Map http://arxiv.org/rem/astro-ph/0601007</title> <author> <name>arXiv.org e-Print Repository</name> <uri>http://arxiv.org/</uri> <email>www-admin@arxiv.org</email> </author> <updated>2007-10-10T18:30:02Z</updated> <link href="info:arxiv/astro-ph/0601007v2" rel="related"/> <link href="http://jp.arxiv.org/abs/astro-ph/0601007" rel="related"/> <dc:title>Parametrization of K-essence and Its Kinetic Term</dc:title> <dc:creator>Hui Li</dc:creator> <dc:creator>Zong-Kuan Guo</dc:creator> <dc:creator>Yuan-Zhong Zhang</dc:creator> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:oai-opmh</id> <link href="http://export.arxiv.org/oai2?verb=GetRecord&metadataPrefix=oai_dc&identifier=astro-ph/0601007" rel="alternate"/> <title>Aggregated Resource Dublin Core Metadata</title> <rdf:type>info:eu-repo/semantics/DescriptiveMetadata</rdf:type> <updated>2006-09-27T00:00:00Z</updated> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:ps</id> <link href="http://arxiv.org/ps/astro-ph/0601007" rel="alternate" type="application/postscript"/> <title>Aggregated Resource http://arxiv.org/ps/astro-ph/0601007</title> <dcterms:hasFormat>http://arxiv.org/ps/astro-ph/0601007v1</dcterms:hasFormat> <updated>2006-05-31T12:52:00Z</updated> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:pdf</id> <link href="http://arxiv.org/pdf/astro-ph/0601007" rel="alternate" type="application/pdf"/> <title>Aggregated Resource http://arxiv.org/pdf/astro-ph/0601007</title> <dcterms:hasFormat>http://arxiv.org/ps/astro-ph/0601007v1</dcterms:hasFormat> <updated>2006-05-31T12:52:00Z</updated> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:e-print</id> <link href="http://arxiv.org/e-print/astro-ph/0601007" rel="alternate"/> <title>Aggregated Resource http://arxiv.org/e-print/astro-ph/0601007</title> <dc:description>LaTeX Source Files</dc:description> <updated>2006-05-31T12:52:00Z</updated> </entry> </feed>
The arXiv example we have been working with is actually version
2 of the e-print. Up until now, we have not distinguished
the second (current) version vs. the first (older version).
To do so, we import the hasVersion
element from
the dcterms
namespace and place this new metadata
element at the Aggregation level. The value of this element is
the URI of the previous version. If we wanted to inform the
agent that the older version should be considered deprecated,
we would use the replaces element from the same namespace.
This example also has a DOI identifying a related (i.e.,
published) version of the e-print. We express this DOI with
the feed/link[@rel="related"]
element.
<?xml version="1.0" encoding="utf-8"?> <feed xmlns="http://www.w3.org/2005/Atom" xmlns:grddl="http://www.w3.org/2003/g/data-view#" grddl:transformation="http://www.openarchives.org/ore/atom-grddl.xsl" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:dcterms="http://purl.org/dc/terms/"> <id>tag:arxiv.org,2007:astro-ph/0601007v2</id> <link href="http://arxiv.org/rem/astro-ph/0601007" rel="self" type="application/atom+xml"/> <category scheme="http://www.openarchives.org/ore/terms/" term="http://www.openarchives.org/ore/terms/ResourceMap" label="Resource Map" /> <link rel="describes" href="http://arxiv.org/rem/astro-ph/0601007#aggregation" /> <title>Resource Map http://arxiv.org/rem/astro-ph/0601007</title> <author> <name>arXiv.org e-Print Repository</name> <uri>http://arxiv.org/</uri> <email>www-admin@arxiv.org</email> </author> <updated>2007-10-10T18:30:02Z</updated> <link href="info:doi/10.1142/S0217732306019475" rel="related"/> <link href="info:arxiv/astro-ph/0601007v2" rel="related"/> <link href="http://jp.arxiv.org/abs/astro-ph/0601007" rel="related"/> <dc:title>Parametrization of K-essence and Its Kinetic Term</dc:title> <dc:creator>Hui Li</dc:creator> <dc:creator>Zong-Kuan Guo</dc:creator> <dc:creator>Yuan-Zhong Zhang</dc:creator> <dcterms:hasVersion>http://arxiv.org/abs/astro-ph/0601007v1</dcterms:hasVersion> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:oai-opmh</id> <link href="http://export.arxiv.org/oai2?verb=GetRecord&metadataPrefix=oai_dc&identifier=astro-ph/0601007" rel="alternate"/> <title>Aggregated Resource Dublin Core Metadata</title> <rdf:type>info:eu-repo/semantics/DescriptiveMetadata</rdf:type> <updated>2006-09-27T00:00:00Z</updated> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:ps</id> <link href="http://arxiv.org/ps/astro-ph/0601007" rel="alternate" type="application/postscript"/> <title>Aggregated Resource http://arxiv.org/ps/astro-ph/0601007</title> <dcterms:hasFormat>http://arxiv.org/pdf/astro-ph/0601007v1</dcterms:hasFormat> <updated>2006-05-31T12:52:00Z</updated> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:pdf</id> <link href="http://arxiv.org/pdf/astro-ph/0601007" rel="alternate" type="application/pdf"/> <title>Aggregated Resource http://arxiv.org/pdf/astro-ph/0601007</title> <dcterms:hasFormat>http://arxiv.org/ps/astro-ph/0601007v1</dcterms:hasFormat> <updated>2006-05-31T12:52:00Z</updated> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:e-print</id> <link href="http://arxiv.org/e-print/astro-ph/0601007" rel="alternate"/> <title>Aggregated Resource http://arxiv.org/e-print/astro-ph/0601007</title> <dc:description>LaTeX Source Files</dc:description> <updated>2006-05-31T12:52:00Z</updated> </entry> </feed>
So-called "splash" pages that conventionally serve as a
human-readable surrogate for an Aggregation other resources; they
are the page that human agents "start" at. From the point of view
of the ReM, they are just another Aggregated Resource, albeit
with an element from the RDF namespace indicating to automated
agents that one of the Aggregated Resources is distinguished
and can serve as the entry point for humans navigating the
Aggregation. This is done in the same manner as marking the
bibliographic metadata above, using the values from the proposed
info:eu-repo/
namespace. In our example, the arXiv
"splash" page is added as an Aggregated Resource.
<?xml version="1.0" encoding="utf-8"?> <feed xmlns="http://www.w3.org/2005/Atom" xmlns:grddl="http://www.w3.org/2003/g/data-view#" grddl:transformation="http://www.openarchives.org/ore/atom-grddl.xsl" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:dcterms="http://purl.org/dc/terms/"> <id>tag:arxiv.org,2007:astro-ph/0601007v2</id> <link href="http://arxiv.org/rem/astro-ph/0601007" rel="self" type="application/atom+xml"/> <category scheme="http://www.openarchives.org/ore/terms/" term="http://www.openarchives.org/ore/terms/ResourceMap" label="Resource Map" /> <link rel="describes" href="http://arxiv.org/rem/astro-ph/0601007#aggregation" /> <title>Resource Map http://arxiv.org/rem/astro-ph/0601007</title> <author> <name>arXiv.org e-Print Repository</name> <uri>http://arxiv.org/</uri> <email>www-admin@arxiv.org</email> </author> <updated>2007-10-10T18:30:02Z</updated> <link href="info:doi/10.1142/S0217732306019475" rel="related"/> <link href="info:arxiv/astro-ph/0601007v2" rel="related"/> <link href="http://jp.arxiv.org/abs/astro-ph/0601007" rel="related"/> <dc:title>Parametrization of K-essence and Its Kinetic Term</dc:title> <dc:creator>Hui Li</dc:creator> <dc:creator>Zong-Kuan Guo</dc:creator> <dc:creator>Yuan-Zhong Zhang</dc:creator> <dcterms:hasVersion>http://arxiv.org/abs/astro-ph/0601007v1</dcterms:hasVersion> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:abs</id> <link href="http://arxiv.org/abs/astro-ph/0601007" rel="alternate" type="text/html"/> <title>Aggregated Resource http://arxiv.org/abs/astro-ph/0601007</title> <rdf:type>info:eu-repo/semantics/humanStartPage</rdf:type> <updated>2006-05-31T12:52:00Z</updated> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:oai-opmh</id> <link href="http://export.arxiv.org/oai2?verb=GetRecord&metadataPrefix=oai_dc&identifier=astro-ph/0601007" rel="alternate"/> <title>Aggregated Resource Dublin Core Metadata</title> <rdf:type>info:eu-repo/semantics/DescriptiveMetadata</rdf:type> <updated>2006-09-27T00:00:00Z</updated> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:ps</id> <link href="http://arxiv.org/ps/astro-ph/0601007" rel="alternate" type="application/postscript"/> <title>Aggregated Resource http://arxiv.org/ps/astro-ph/0601007</title> <dcterms:hasFormat>http://arxiv.org/pdf/astro-ph/0601007v1</dcterms:hasFormat> <updated>2006-05-31T12:52:00Z</updated> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:pdf</id> <link href="http://arxiv.org/pdf/astro-ph/0601007" rel="alternate" type="application/pdf"/> <title>Aggregated Resource http://arxiv.org/pdf/astro-ph/0601007</title> <dcterms:hasFormat>http://arxiv.org/ps/astro-ph/0601007v1</dcterms:hasFormat> <updated>2006-05-31T12:52:00Z</updated> </entry> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:e-print</id> <link href="http://arxiv.org/e-print/astro-ph/0601007" rel="alternate"/> <title>Aggregated Resource http://arxiv.org/e-print/astro-ph/0601007</title> <dc:description>LaTeX Source Files</dc:description> <updated>2006-05-31T12:52:00Z</updated> </entry> </feed>
The Data Model Document [ORE
Model] defines the notion of Aggregations
linking to other Aggregations with two relationships:
ore:aggregates and ore:isAggregatedBy. The latter
allows a ReM to reference the origin of its Aggregated
Resources in another ReM, providing a verifiable lineage of
discovery and provenance. Atom provides two mechanisms for
establishing the provenance of /feed/entry
elements:
/feed/entry/source
(used to link to the initial
occurence) and feed/entry/link[@rel="via"]/@href
(used to link to the most recent occurence). The
feed/entry/link[@rel="via"]/@href
approach is used
to implement the ore:isAggregatedBy relationship (for details,
see Table 3 of the [ReM Profile
of Atom]). Although /feed/entry/source
has no direct correspondence in the ORE Abstract Data Model
[ORE Model], it is covered
here because it is part of the Atom idiom.
Imagine our example arXiv e-print has been selected for inclusion in an overlay journal, and this overlay journal also uses ReMs to describe Aggregations (in this case, journal issues) at their site. The journal wishes to use only the PDF version of the e-print and not the entire arXiv Aggregation.
As described in section 4.2.11 of [ReM Profile of Atom], the
overlay journal MUST copy and paste the entire entry
element that conveys the PDF file as an Aggrgated Resource
into the ReM of the overlay journal. In addition, they MUST
populate the Atom /feed/entry/source
element
with metadata from the original (arXiv) ReM, specifically
the /feed/id
, /feed/link
,
/feed/title
, /feed/rights
,
/feed/category
, and /feed/updated
elements MUST be copied. In the example below, we are looking
at the ReM of the overlay journal, with only the entry element
from the arXiv ReM shown.
<?xml version="1.0" encoding="utf-8"?> <feed xmlns="http://www.w3.org/2005/Atom"> <id>tag:overlay-journal.org,2007:vol=12,number=5</id> <link href="http://www.overlay-journal.org/12/05/rem.atom" rel="self" type="application/atom+xml"/> <category scheme="http://www.openarchives.org/ore/terms/" term="http://www.openarchives.org/ore/terms/ResourceMap" label="Resource Map" /> <link rel="describes" href="http://www.overlay-journal.org/12/05/rem.atom#aggregation" /> <title>Resource Map http://www.overlay-journal.org/12/05/rem.atom</title> <author> <name>Overlay Journal for Frog & Toad Studies</name> <uri>http://www.overlay-journal.org/</uri> <email>help@overlay-journal.org</email> </author> <updated>2007-12-12T15:30:02Z</updated> <!--other entries of the overlay journal--> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:pdf</id> <link href="http://arxiv.org/ps/astro-ph/0601007" rel="alternate" type="application/postscript"/> <title>Aggregated Resource http://arxiv.org/ps/astro-ph/0601007</title> <updated>2006-05-31T12:52:00Z</updated> <source> <id>tag:arxiv.org,2007:astro-ph/0601007v2</id> <link href="http://arxiv.org/rem/astro-ph/0601007" rel="self" type="application/atom+xml"/> <category scheme="http://www.openarchives.org/ore/terms/" term="http://www.openarchives.org/ore/terms/ResourceMap" label="Resource Map" /> <link rel="describes" href="http://arxiv.org/rem/astro-ph/0601007#aggregation" /> <title>Resource Map http://arxiv.org/rem/astro-ph/0601007</title> <author> <name>arXiv.org e-Print Repository</name> <uri>http://arxiv.org/</uri> <email>www-admin@arxiv.org</email> </author> <updated>2007-10-10T18:30:02Z</updated> <link href="info:doi/10.1142/S0217732306019475" rel="related"/> <link href="info:arxiv/astro-ph/0601007v2" rel="related"/> <link href="http://jp.arxiv.org/abs/astro-ph/0601007" rel="related"/> </source> </entry> <!--other entries of the overlay journal--> </feed>
Now imagine this issue of the overlay journal becomes popular in the scientific blog community, and this particular arXiv PDF is linked to many times. If the arXiv ReM is the first occurrence and the overlay journal is the second occurrence, then the first blog to comment on it is the third occurrence, and if a second blog discovers the PDF from the first blog, that is then the fourth occurrence and so forth.
Now because the /feed/entry/source
element
is already present in the ReM of the overlay journal, each
subsequent blog reference will be copied verbatim with
the original arXiv information retained in each blog's
/feed/entry/source
element. That means the 98th,
99th and 100th blog will each have the a reference back to the
original arXiv ReM. But we would also like to capture that blog
100 discovered the PDF from blog 99. This is accomplished using
the /feed/entry/link[@rel="via"]/@href
element.
If /feed/entry/source
always maintains a reference to
the original ReM, /feed/entry/link[@rel="via"]/@href
always maintains a reference to the previous (i.e., "n-1") ReM.
In the example below, the ReM for blog 100 is shown with
a /feed/entry/link[@rel="via"]/@href
element
establishing the lineage to the ReM of blog 99 and the arXiv ReM.
<?xml version="1.0" encoding="utf-8"?> <feed xmlns="http://www.w3.org/2005/Atom"> <id>tag:blog100.blogsrus.org,2007:entry1322</id> <link href="http://blog100.blogsrus.org/entry/1322.atom" rel="self" type="application/atom+xml"/> <category scheme="http://www.openarchives.org/ore/terms/" term="http://www.openarchives.org/ore/terms/ResourceMap" label="Resource Map" /> <link rel="describes" href="http://blog100.blogsrus.org/entry/1322.atom#aggregation" /> <title>Resource Map http://blog100.blogsrus.org/entry/1322.atom</title> <author> <name>Blog100</name> <uri>http://blog100.blogsrus.org//</uri> <email>blog100@blogsrus.org</email> </author> <updated>2007-12-25T12:30:42Z</updated> <!--other entries of the blog100--> <entry> <id>tag:arxiv.org,2007:astro-ph/0601007v2:pdf</id> <link href="http://arxiv.org/ps/astro-ph/0601007" rel="alternate" type="application/postscript"/> <title>Aggregated Resource http://arxiv.org/ps/astro-ph/0601007</title> <updated>2006-05-31T12:52:00Z</updated> <link href="http://blog99.toadscience.com/entry742.atom" rel="via" type="application/atom+xml"/> <source> <id>tag:arxiv.org,2007:astro-ph/0601007v2</id> <link href="http://arxiv.org/rem/astro-ph/0601007" rel="self" type="application/atom+xml"/> <category scheme="http://www.openarchives.org/ore/terms/" term="http://www.openarchives.org/ore/terms/ResourceMap" label="Resource Map" /> <link rel="describes" href="http://arxiv.org/rem/astro-ph/0601007#aggregation" /> <title>Resource Map http://arxiv.org/rem/astro-ph/0601007</title> <author> <name>arXiv.org e-Print Repository</name> <uri>http://arxiv.org/</uri> <email>www-admin@arxiv.org</email> </author> <updated>2007-10-10T18:30:02Z</updated> <link href="info:doi/10.1142/S0217732306019475" rel="related"/> <link href="info:arxiv/astro-ph/0601007v2" rel="related"/> <link href="http://jp.arxiv.org/abs/astro-ph/0601007" rel="related"/> </source> </entry> <!--other entries of blog100--> </feed>
Note that in section 5.1 above, the overlay journal could have
used /feed/entry/link[@rel="via"]/@href element
in addition to /feed/entry/source
to reference the
original arXiv ReM. However, this was omitted from section 5.1
for clarity because that is a special case where the original
ReM and the previous ReM were the same (i.e., the arXiv ReM).
This document is the work of the Open Archives Initiative. Funding for Open Archives Initiative Object Reuse and Exchange is provided by the Andrew W. Mellon Foundation, Microsoft, and the National Science Foundation. Additional support is provided by the Coalition for Networked Information.
This document is based on meetings of the OAI-ORE Technical Committee (ORE-TC), with participation from the OAI-ORE Liaison Group (ORE-LG). Members of the ORE-TC are: Chris Bizer (Freie Universität Berlin), Les Carr (University of Southampton), Tim DiLauro (Johns Hopkins University), Leigh Dodds (Ingenta), David Fulker (UCAR), Tony Hammond (Nature Publishing Group), Pete Johnston (Eduserv Foundation), Richard Jones (Imperial College), Peter Murray (OhioLINK), Michael Nelson (Old Dominion University), Ray Plante (NCSA and National Virtual Observatory), Rob Sanderson (University of Liverpool), Simeon Warner (Cornell University), and Jeff Young (OCLC). Members of ORE-LG are: Leonardo Candela (DRIVER), Tim Cole (DLF Aquifer and UIUC Library), Julie Allinson (JISC), Jane Hunter (DEST), Savas Parastatidis (Microsoft), Sandy Payette (Fedora Commons), Thomas Place (DARE and University of Tilburg), Andy Powell (DCMI), and Robert Tansley (Google, Inc. and DSpace)
We also acknowledge comments from the OAI-ORE Advisory Committee (ORE-AC).
Many thanks to the Digital Library Research & Prototyping Team of the Los Alamos National Laboratory for their inspiring explorations into Atom and ORE space: Lyudmilla Balakireva, Ryan Chute, Stephan Drescher, Alberto Pepe, Zhiwu Xie.
This work is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported License.
Use of this page is tracked to collect anonymous traffic data. See OAI privacy policy.