utype questions

[Doug Tody]

On Tue, 30 Jun 2009, Frederic Hessman wrote:

> No, a utype is a label for a member of a data structure, which sounds like 
> data-structure vocabulary to me.  Isn't Norman's original comment that what 
> utypes mean appears to depend upon who you ask?  That sounds like utypes are 
> closed systems unless we find a standardized means of translating them into 
> other forms of knowledge.  The whole point of the vocabulary effort was to 
> define a common platform for systematising and translating such labels.

There is no need to translate or infer anything from UTYPEs, on the
contrary this is precisely what we are trying to avoid.  A major point
of UTYPEs is to provide a simple, direct, and *unique* way to specify
the field of a specific data model.  Otherwise we might have been able
to use UCDs, but these do have some meaning in a more global sense, as
separate entities, causing issues and ambiguities when used as UTYPEs.

To get back to an actual vocabulary, consider our use case of
Target.Class:

On Mon, 29 Jun 2009, Matthew Graham wrote:
> To refer to an AGN use "http://eurovotech.org/objects-structure.owl#AGN"
>
> This has subclasses:
>
> "http://eurovotech.org/objects-structure#LINER"
> "http://eurovotech.org/objects-structure#RadioLoudAGN"
> "http://eurovotech.org/objects-structure#RadioQuietAGN"

This is exactly what we need; the only issue is how we use this
facility.

If we do a data query and get back a query response which we want to
display to the user, it is a requirement that we be able to display
the object classification as something like "AGN" or "RadioQuietAGN".
Displaying a URL in this case is not acceptable, nor is displaying
the content of the URL, unless the user asks for a more complete
description.

The query response of a data query is often displayed directly, or
with simple tools.  Any number of such tools might manipulate and
display such a query response - a generic table display tool is a
typical example.  Hence if the value of Target.Class is something like

     http://eurovotech.org/objects-structure#RadioQuietAGN

then this is the string which will be displayed to the user in
many (probably most) cases.  This is not acceptable.

The way I would expect this to work is more like this:

     o	The data model specifies the vocabulary to be used for the
 	acceptable values of Target.Class.  Somewhere this vocabulary
 	is defined, e.g., by a document such as

 	  http://www.ivoa.net/internal/IVOA/IvoaSemantics/WD_2007-02-19.pdf

 	or via some online service.  These data provider will use these
 	tools to determine the appropriate target classification when the
 	metadata is created.

     o	Target.Class has a value such as "RadioQuietAGN" in our query
     	response table.  When we use simple tools to view the query
 	response this is what the user will see, e.g., in a table listing
 	the objects found in some data query.

     o	If more information on the object classification is desired
 	a client application might consult the registry to find a
 	service capable of doing useful things with the vocabulary
 	in question.  There could well be several such vocabulary
 	services available, hence we do not want to wire a string
 	such as "http://eurovotech.org/objects-structure" into the
 	metadata in our DBMS.

We could then display more detailed information describing the given
class of object, or do things such as identify related classes of
objects, and possibly repeat the search using this information.
In general there are any number of such things one might want to
do, which is one reason that using a specific URL instead of the
vocabulary-specific term is inadvisable.  In most cases just displaying
"RadioQuietAGN" is all that is needed.

One comparision of vocabularies with UTYPEs that is valid is that
both define a controlled namespace.  In the case of a vocabulary
these namespaces can be very large.  In such a case I would think it
is even more important to separate out the identity of the vocabulary
from the use of instances such as AGN etc.

 	- Doug