[topicmapmail] TM Question

[Gregory Fox]

Lars, Graham, Alexander,

Thanks for your replies (not sure Graham's made it to the mail list).  My
original question had to do with representing ranking attributes.

[If I understand your request correctly, you're searching for a
ontology, right?]

At the very least I am looking at TM's as a way of navigating and exchanging
factual knowledge.  I have been reading very carefully the discussions,
examples and positions regarding the difference between using TM's as a
primary data store/model VS using TM's as a secondary (output) data
store/model.  I am not savvy enough about TM's to have an opinion, yet, for
our uses.

[Maybe you get better answers than mine if you explain the requirements
a bit more.]

Not to bore you, but our intended use is basically the following:  Our field
is social networking, applied (very differently) to organizations and
children/education. Social networking has become a very broad category and
doesn't apply just to myspace.com and the like. We focus on the integration
software that is needed to obtain data and we are now looking at the many
analyses that are done.  There are two acronyms being used:  SNA (social
networking analysis) when the focus is bottom-up, that is, on the
individuals; and ONA (organizational networking analysis) when the focus is
top-down - on the org.

The point of that last paragraph is this:  There are many outputs from such
analyses; some statistical, some just statements and some that are basically
sub-domain data sets (ie the relationships between members of a project team
based on the details of their communications and focus).  I have not seen
any significant mention of TM's in the SNA/ONA world (if I am missing it
please correct me).  Probably that's because the primary goal of SNA is the
answer to a tested hypothesis; not the navigation of the data which supports
it.  What's different about our plans is that such navigation and
visualization is very important.  TM's appear to be a natural fit 

Another specific requirement is inherent to our field: there will be many
visual maps on the same data set (or processed data set outputs).  So what
is the TM?  The easy answer is that there could be many.  Whether there is
the business or technical case to represent all the data in one big honkin'
TM is not clear to me.  Partly that's because our overall spec is at the
concept stage and this is an investigation.  I will consider the TMRM angle,
also.

Thanks so much for your input.

Best regards,
  Greg
  gfox at veraterm.com