The Media Types application/mods+xml, application/mads+xml, application/mets+xml, application/marcxml+xml, application/sru+xml

The Library of Congress, on behalf of and in collaboration with various components of the metadata and information retrieval community, has issued specifications which define formats for representation of various forms of information including metadata and search results. This memo provides information about the Media Types associated with several of these formats, all of which are XML schemas. An XML schema for a bibliographic element set that may be used for a variety of purposes, and particularly for library applications. An XML schema for an authority element set used to provide metadata about agents (people, organizations), events, and terms (topics, geographics, genres, etc.). It is a companion to the MODS Schema. An XML schema for encoding descriptive, administrative, and structural metadata regarding objects within a digital library. An XML schema for the direct XML representation of the MARC format (for which there already exists a media type, application/marc; see ). By "direct XML representation" we mean that it encodes the actual MARC data within XML. (This is in contrast to MODS: MARC uses codes for its element names; MODS represents the same information but uses semantically meaningful names while MARCXML uses the MARC codes.) An XML schema for the SRU response. SRU is a protocol, and the media type sru+xml pertains specifically to the default SRU response. The SRU response may be supplied in any of a number of suitable schemas, RSS, ATOM, for example, and the client identifies the desired format in the request, hence the need for a media type. This mechanism will be introduced in SRU 2.0; in previous versions (that is, all versions to date; 2.0 is in development) all responses are supplied in the existing default format, so no media type was necessary. SRU 2.0 is being developed within the Organization for the Advancement of Structured Information Standards (OASIS).

This parameter has semantics identical to the charset parameter of the "application/xml" media type as specified in . Identical to those of "application/xml" as described in . See of RFC &rfc.number;. [[NOTE TO RFC EDITOR: Please replace "XXXX" with the number issued to this specification.]] There are no known interoperability issues. , Various MODS conformant toolkits use this media type. Ray Denenberg <rden@loc.gov> The MODS specification was developed by the Library of Congress and is maintained by the Library of Congress in conjunction with the MODS Editorial Committee which has change control over the specification.

This parameter has semantics identical to the charset parameter of the "application/xml" media type as specified in . Identical to those of "application/xml" as described in . See of RFC &rfc.number;. [[NOTE TO RFC EDITOR: Please replace "XXXX" with the number issued to this specification.]] There are no known interoperability issues. , Various MADS conformant toolkits use this media type. Ray Denenberg <rden@loc.gov> The MADS specification was developed by the Library of Congress and is maintained by the Library of Congress in conjunction with the MODS Editorial Committee which has change control over the specification.

This parameter has semantics identical to the charset parameter of the "application/xml" media type as specified in . Identical to those of "application/xml" as described in . See of RFC &rfc.number;. [[NOTE TO RFC EDITOR: Please replace "XXXX" with the number issued to this specification.]] There are no known interoperability issues. , Various METS conformant toolkits use this media type. Ray Denenberg <rden@loc.gov> The METS specification was developed by the Library of Congress and is maintained by the Library of Congress in conjunction with the METS Editorial Board as an initiative of the Digital Library Federation.

This parameter has semantics identical to the charset parameter of the "application/xml" media type as specified in . Identical to those of "application/xml" as described in . See of RFC &rfc.number;. [[NOTE TO RFC EDITOR: Please replace "XXXX" with the number issued to this specification.]] There are no known interoperability issues. , Various MARCXML conformant toolkits use this media type. Ray Denenberg <rden@loc.gov> The MARCXML standard is driven by the MARC standards, maintained at the Library of Congress.

This parameter has semantics identical to the charset parameter of the "application/xml" media type as specified in . Identical to those of "application/xml" as described in . See of RFC &rfc.number;. [[NOTE TO RFC EDITOR: Please replace "XXXX" with the number issued to this specification.]] There are no known interoperability issues. , Various SRU conformant toolkits use this media type. Ray Denenberg <rden@loc.gov> The SRU specification for versions earlier than 2.0 is maintained at the Library of Congress, in conjunction with the SRU Editorial Board. Version 2.0 (for which the sru+xml media type pertains) is being developed within OASIS.

IANA is requested to register five media types described in Sections 2, 3, 4, 5, and 6 in the MIME media type registry (in the Standards Tree).

An XML resource does not in itself compromise data security. Applications that retrieve XML files over a network by means of dereferencing a Uniform Resource Identifier are advised to properly interpret the data so as to prevent unintended access. Hence the security issues described in Section 7 of apply. Because the media types described in this document use the "+xml" convention, they share the security considerations described in Section 10 of . In general, security issues related to the use of XML in IETF protocols are discussed in Section 7 of also apply. In the following sections we review but review some aspects that are important for document-centric XML as applied to text encoding.

Any application that retrieves the XML media types described in this specification needs to be aware of risks connected with injection of harmful scripts and executable XML (i.e., "active content" as described in ). Although XML inclusion mechanisms and the use of external entities can introduce vulnerabilities to various forms of spoofing and also reveal aspects of a service in a way that may compromise its security, such vulnerabilities are application-specific. In any case, MODS documents do not contain "active content".

Historical and bibliographical information can be often encoded in MODS documents, and such information might even have legal force in some jurisdictions. Digitization and encoding of such information might require technologies for assuring authenticity, such as cryptographic check sums and electronic signatures. Similarly, historical documents might in part or in their entirety be confidential. Such confidentiality might be required by law or by the terms and conditions such as in the case of donated or deposited text from private sources. A text archive might need content filtering or cryptographic technologies to meet such requirements.