innovation in metadata design, implementation & best practices

Recommendations Concerning Capitalization in DCMI Namespace and Term Declarations by the DCMI Directorate

Dublin Core Metadata Initiative logo
About the Initiative Documents Working Groups Resources
Dublin Core Metadata Initiative News Tools and Software Meetings Projects
Dublin Core Metadata Initiative
Home Documents > 


Recommendations Concerning Capitalization in DCMI Namespace and Term Declarations by the DCMI Directorate

Stuart Weibel
Date Issued:
Is Replaced By:
Not applicable
Latest Version:
Status of Document:
This is a DCMI Proposed Recommendation.
Description of Document: This document describes an abstract model for Dublin Core metadata records.

The Problem

The Dublin Core Metadata Initiative maintains a set of metadata terms agreed by a large cross-section of communities represented on the Worldwide Web. Deploying these semantic agreements necessitates conventions on syntax and encoding of metadata terms as well. Parallel protocol developments within various segments of the Web and the DC Community have resulted in naming conventions that are inconsistent with regard to upper and lower case in some DCMI legacy documentation.

This document elucidates DCMI policy concerning case in DCMI term declarations and explains the historical context for this policy.

The Policy: Case Sensitivity in DCMI Term Declarations

Names of DCMI elements and element refinements start with lowercase characters, but may contain uppercase characters where a term name is comprised of multiple concatenated words. In such cases, the leading character of additional words will be capitalized to improve human-readable clarity.
Examples include: creator, audience, isReplacedBy

Names of encoding schemes identified within the DCMI namespace are represented in all upper case characters when the name represents an acronym.
Examples include: DDC, W3CDTF, ISO639-2

Names of encoding schemes identified within the DCMI namespace that are not acronyms are represented with a leading upper case character followed by lowercase characters.
Examples include: Period, Box

Names of values within a controlled vocabulary such as the DCMI Type Vocabulary begin with an uppercase character followed by lowercase, but with subsequent concatenated words in the name capitalized as well
Examples include: InteractiveResource, Collection, Dataset

No DCMI term names will be assigned that differ from other names only in regard to case.


All DCMI metadata terms are identified with a formal name string that uniquely distinguishes the term within the DCMI namespace. This string is invariant across encodings and languages; it is the case specification of such name strings that is the subject of this policy.
In addition to the invariant name string, DCMI metadata terms can be localized with a label expressed in a character set and language appropriate to a particular application or language encoding. Thus, while a given term has only a single formal name, it may have many labels.
DCMI metadata terms include all elements, element refinements, controlled-vocabulary terms, and encoding schemes.

Advice to Implementers

Confusion over case sensitivity is commonplace in Web protocols. Metadata application developers are likely to find many permutations of case in instance data and term declarations. Thus, to paraphrase a dictum from early Web protocol development, applications should: rigorous in what you export, and tolerant in what you import from other applications.

In practice, this suggests that applications are well advised to normalize case when parsing terms for identity comparisons. Prudence mitigates against the use of case to distinguish between alternative identities of related terms in any namespace, and it is DCMI policy that such distinctions not be made, so it is unlikely that errors would be introduced by normalizing case.

In situations where applications specify URI links (rather than parse identifiers for comparison purposes), it will be the policy of DCMI to maintain legacy links even when they conflict with this stated policy, in order that existing applications not be broken.

The historical context for case confusion within the DC community

DC metadata is deployed in a variety of encodings (HTML, XML, and RDF/XML), and naming conventions have historically differed among these communities. These differences have led to confusion and resulted in conflicting specifications, documentation and legacy practices.

The present policy brings DCMI in line with the conventions in use for the declaration of namespaces in the XML and RDF/XML communities, and in wide (though not universal) use in Dublin Core applications. The decision to do so was agreed by the DCMI Advisory Committee at a meeting held in conjunction with the DC-8 workshop in Ottawa in October of 2000, but this decision was not propagated throughout the documentation process at that time.

Remediation of inconsistencies in DCMI specifications and documentation

The DCMI Directorate will undertake to amend existing specifications and documentation to alert readers to these issues and bring such documentation into consistency, as permitted by the maintenance cycles of such documentation.

Feedback on this document

The DCMI Directorate welcomes comments and suggestions on this policy:

Valid XHTML 1.0! Valid CSS!

Metadata associated with this resource:

Copyright © 1995-2002 DCMI All Rights Reserved. DCMI liability, trademark/service mark, document use and software licensing rules apply. Your interactions with this site are in accordance with our privacy statements. Please feel free to contact us for any questions, comments or media inquiries.