Aug. 9, 2006

From Metadata-Registry
Revision as of 11:15, 9 August 2006 by Diane (Talk | contribs)

Jump to: navigation, search

Agenda and Notes, Aug. 9, 2006

1. Review of Action items from meeting:

  • Jon: Send Stuart an email about what we need for monitoring our sites. (Done) Stuart has forwarded this to Bruce in Tucson, as they are looking at the same things.
  • Jon: Send the URL of the Ayres model to the list. (Done) Nobody had looked too closely at this but Jon wants some feedback on this as something we might do.
  • Stuart: Report on meetings with Diny. Background is our desire to figure out what Lee Zia is thinking about the future of NSDL and what our place might be in it. She has had a recent conversation on this with Lee and has another coming up--apparently they have no more idea now than they did 6 months ago. She would like to continue to talk to him about this, and keep it informal and backchannel. Stuart thinks that if we get the IMLS grant, that will materially change the nature of the conversation, and use that as leverage. If we don't, we'll continue to need to convince them to disaggregate some of their functions, pulling away the metadata piece and allowing us to do it and feed into Fedora. The saga continues ...
  • Diane: Inquire of Rachel about plans for the Registry WG meeting in Colima (Done; no reply yet)
  • Joe: Will put the policy document on next meeting agenda (was this versioning, or something else?) Some is here [1]. The rest in #6 below.

2. Update on where we are on vacation-necessary IT changes, and mirroring. (Stuart)

3. New things on the beta site

4. Still to do:

  • New front page: Diane will write, Jon will insert and primp
  • Most of remaining Roadmap tasks for May and June (through first three items under number 3) (Jon)
  • Registering Schema and AP properties as vocabularies (Diane)

5. Jack Bradbury's notes to us re: Animal Behavior Ontology and Neuronbank.org vocabularies. Both are using Protege which does not accommodate versions. Jack feels that the vocabularies are intertwined and would like to explore that. What would we need to do to facilitate this? How does this fit in with our work on versioning? What are the timeframes and tasks? (Possibly the first thing is to write up what we're talking about and let Jack and his colleague look it over?)


6. Versioning Policy DRAFT Version Policy – NSDL Registry Project

1. Definitions Resource - Anything addressable by a URI (e.g., documents, images, people, services, institutions, collections, etc.)

Attributes - Characteristics of resources.

Values - Specific names of characteristics of resources.

Assertion - The declaration of an attribute and a value of a resource.

Statement - A machine and human readable assertion. Statements are created from vocabularies.

Vocabulary - Vocabulary is the set of all assertions that can be made about resources. Statements are created from vocabularies. There are two types of vocabularies: schemes and schemas.

Scheme - A set of all values that can be used to describe a resource.

Vocabulary Encoding Scheme - A set of all values that can be used to describe a resource, where the set is governed by a set of rules and a list of given values.

Indexing Language - A set of all values that can be used to describe a resource, where the set is governed by a set of rules and a list of given values, which is used for representing significant characteristics of a resource such as subject matter, methodology, use, etc. for the purposes of retrieval.

ANSI/NISO z39.19-2005 defines indexing language as: - A controlled vocabulary or classification system and the rules for its application. An indexing language is used for the representation of concepts dealt with in documents [content objects] and for the retrieval of such documents [content objects] from an information storage and retrieval system. [ISO 5127/1]


Controlled Vocabulary {def incomplete} - ANSI/NISO z39.19-2005 defines controlled vocabulary as: A list of terms that have been enumerated explicitly. This list is controlled by and is available from a controlled vocabulary registration authority. All terms in a controlled vocabulary must have an unambiguous, non-redundant definition. NOTE: This is a design goal that may not be true in practice; it depends on how strict the controlled vocabulary registration authority is regarding registration of terms into a controlled vocabulary. At a minimum, the following two rules must be enforced: 1. If the same term is commonly used to mean different concepts, then its name is explicitly qualified to resolve this ambiguity. NOTE: This rule does not apply to synonym rings. 2. If multiple terms are used to mean the same thing, one of the terms is identified as the preferred term in the controlled vocabulary and the other terms are listed as synonyms or aliases.


Classification Scheme {def incomplete} - A scheme of classes fitted with terminology and notation. [Ranganathan, 1967 p 72]

Classification Schedule {def incomplete} - A dictionary giving the meaning of each class number or each of its components, in a natural language or jargon. [Ranganathan, 1967 p. 73]

Syntax Encoding Scheme {def incomplete} - A set of rules used to create a set of values that can be used to describe a resource.

Schema - Schema are a set of all attributes that can be used to describe a resource.

Metadata Schema - Set of all attributes that can be used to describe a resource for both humans and machines to read.

Application Profile {def incomplete} Set of all attributes that can be used to describe a resource for both humans and machines to read.

Refereneable Snapshot - Not signaled by documentation, All assertions are current (a vocabulary encoding scheme that is current)

Version (i.e., Edition, Expansion, Abridgement, Enlargement) - Signaled by documentation, Requires a referenceable snapshot plus documentation

Current/Deprecated/Historical Value - Need a vocabulary of values


2. What is versioned {section incomplete} Vocabulary encoding schemes are versioned Each version has a referenceable snapshot This snapshot can change without signaling a new version – in this case values may change – and if they do values are either current, deprecated, or historical… etc.

3. Tracking versions Versions are tracked by a link to documentation and a version identifier (which is embedded in value-record metadata).

4. Structure of versioning Versioning happens at two levels: the assertion record in the registry AND at the documentation level that points to a vocabulary encoding scheme.

5. Levels of conformance with this policy Vocabulary encoding schemes are not required to change versions when they are updated. By default all vocabulary encoding schemes will be Version 1 and be referenceable snapshots of those. It will be an editor’s responsibility to make a decision about versioning and creating documentation for versions. Help and rationale for documentation and versioning can be found here [future link – joe to write rationale and how-to version guide].

Therefore there are two levels of conformance: (1) Minimal (2) Complete, where Minimal conformance where all assertions in the vocabulary encoding scheme in the registry are current, and where Complete includes linked documentation and version numbering in value records).