Updated SSA documents for discussion

[Doug Tody]

The versions of the SSA documents to be discussed in the DAL WG session
are the following:

    SSA Protocol, V0.95
	http://www.aoc.nrao.edu/~dtody/ssa/ssa-v095.pdf

    Spectrum Data model (V0.96d and V0.97 as noted earlier by Jonathan)
	http://hea-www.harvard.edu/~jcm/vo/docs/index.htm

Links to these should appear later today on the TWiki pages.

The protocol document has had some significant revisions which we will
want to discuss in the WG meeting:

    o	The protocol has been modified to explicitly support version
	negotiation and multiple service operations, e.g. for SSA (or the
	future SIA), we have queryData and getCapabilities, and eventually
	stageData.  This was based on a similar HTTP GET service interface
	defined by the OpenGIS folks, using their Web Map Service (WMS)
	as a template.

    o	The document has been restructured to add a new section on Basic
    	Service Elements which discusses the mechanical aspects of the
	protocol, such as version negotiation, request response, allowed
	numeric formats, etc.

    o	A separate section discusses the operations (methods) defined
    	by the SSA service.

    o	A proposed definition of a general range-list syntax for list
    	and range-stgructured query parameters has been added.  Note
	there have been some changes from our discussions on the mailing
	list of a couple months ago.  In particular, the range delimiter
	was changed from ":" to "/", following the convention already
	defined by OpenGIS/WMS.  The reason for this is that ":" is a
	legal character in the ISO 8601 time format which we may want to
	use for time range expressions.

    o	The query parameters for the queryData operation have been revised
    	as per our discussions in Kyoto and Madrid.

    o	The classification model for data collections and services (terms
	like atlas, pointed, cutout, etc.) has been revised per our earlier
	discussions.

The query response section has not been updated since Madrid since we are
still actively discussing related Spectrum, Characterization, and STC data
model issues with the DM group.

	- Doug