Another Word For It Patrick Durusau on Topic Maps and Semantic Diversity

June 16, 2015

A Topic Map Irony

Filed under: Topic Maps — Patrick Durusau @ 12:55 pm

I have been working for weeks to find a “killer” synopsis of topic maps for a presentation later this summer. I have re-read all the old ones, mine, yours, theirs and a number of imaginary ones. None of the really seem to be the uber topic map synopsis.

After my latest exchange with a long time correspondent and my most recent suggestion lamed in the exchange, a topic map irony occurred to me:

For all of the flogging of semantic diversity in promotion of topic maps, it never occurred to me that looking for one (1) uber explanation of topic maps was going in the wrong direction.

What aspect of “topic maps” are important to one audience is very unlikely to be important to another.

What if their requirements are to point to (occurrences) of a subject in a data set and to maintain documentation about those subjects, separate and apart from that data set? The very concept of merging may not be relevant for a documentation use case.

What if their requirements are the modeling of relationships (associations) with multiple inputs of the same role and a pipeline of operations. The focus there being on modeling with associations. That topic maps have other characteristics may be interesting but not terribly important.

What if their requirements are the auditing of the mapping of multiple data sources that are combined together for a data pipeline? There we get into merging and what basis for merging exists, etc. and perhaps not so much into associations.

And there are any number of variations on those use cases, each one of which would require a different explanation and emphasis on topic maps.

To say nothing of having different merging models, some of which might ignore IRIs as a basis for merging.

To approach semantic diversity with an attempt at uniformity seems deeply ironic.

What was I thinking?

PS: To be sure, interchange in a community of use requires the use of standards but those should exist only in domain specific cases. Trying to lasso the universe of subjects in a single representation isn’t a viable enterprise.

1 Comment

  1. The Marx Brothers on merging:

    Capt. Spaulding: I used to know a fellow who looked exactly like you by the name of Emanuel Ravelli. Are you his brother?

    Ravelli: I am Emanuel Ravelli.

    Capt. Spaulding: You’re Emanuel Ravelli?

    Ravelli: I am Emanuel Ravelli.

    Capt. Spaulding: Well, no wonder you look like him. But I still insist there is a resemblance.

    Ravelli: Heh, heh, he thinks I look alike.

    Capt. Spaulding: Well, if you do, it’s a tough break for both of you.

    Comment by shunting — June 21, 2015 @ 12:54 pm

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

Powered by WordPress