What is an Astro MOAT?

The following definition is copied from the MOAT home page: MOAT (Meaning Of A Tag) provides a Semantic Web framework to publish semantically-annotated content from free-tagging. It is a set of mappings between tags and URIs that define the meaning of the tag.

In other words, for each tag we store a URI that represents the meaning of the tag and an identifier - in the form of the location of the FOAF file - for the person who stated this association. A tag can have multiple URIs asociated with it, perhaps because it has multiple meanings, or because there are multiple URIs that refer to the same "thing".

To populate this "Astro MOAT", I have extracted tags from two SKOS vocabularies produced by the IVOA Semantics Working Group, as well as related projects, such as the Astronomical Ontology project by the UMD group, and some more general semantic resources, such as DBpedia, OpenCyc, and UMBEL. For further information on these see the Tags page.

The real use of this information is that "agents" (i.e. some bit of software somewhere) can query the MOAT server to identify possible meanings for tags, and so allow the author to quickly add semantic knowledge to their work (i.e. whatever it is that they are adding tags to). Or at least that is the theory; at present there aren't really any agents that can be used to aid the Astronomer.

Tag Syntax

At present tags are assumed to consist of the characters a-z and 0-9, and are taken to be case insensitive. Upper-case characters are converted to lower case, and any characters outside the supported range are ignored. These restrictions may be modified or removed in the future.

Additions to the MOAT server

1. The HTML version of each tag includes information extracted from its associated URI, rather than just list the associated URI and author. This means that tags that are associated with a term in a SKOS vocabulary will include references to other tags that are also associated with that vocabulary, using the "broader than", "narrower than", and "related" relationships given in the vocabulary. As an example: the HTML output for the tag aestar contains at least two entries, one from the Astronomical Ontology of the UMD group, and one from the SKOS version of the IVOA thesaurus. The SKOS entry is annotated with extra information such as a list of more-general tags - such as astar - which has been derived from the SKOS vocabulary.

   This information is not present for all tags since I have not included relevant information for tags related to the Astronomical Ontology or from more-general references such as UMBEL.

2. The data for a tag can be requested in either of the Notation 3 (N3) or turtle formats. If using content negotiation, the MIME media types of text/rdf+n3 or application/x-turtle should be used. For direct access, append ".n3" or ".ttl" to the URL for a tag. An example of this is provided in the curl example on the Data Access page.

   Since the structure of the data returned by this server is simple, the contents of the N3 and Turtle formats are essentially identical.

3. A list of all the defined tags can be retrieved by:

   1. Accessing the /taglist URL, using content negotiation, or one of taglist.html, taglist.rdf, taglist.n3. and taglist.ttl.

   At present this list can not be retrieved in either the "full" or "light" versions of the JSON format since it is unclear (to me) what exactly should be returned.

4. Basic search capabilities used to be provided on the home page, but this feature is currently unavailable as of February 24 2009.

Missing features of the MOAT server

New (February 24 2009): The addition of "/<format>" to a tag to return data in the specified format is currently not supported.

At present there is no way to add new meanings to a tag, or add new tags. This may be added, depending on how useful this service turns out to be.