While reading one of the surveys on Big Data it occurred to me that the W3C was correct about one thing. Data without semantics isn’t going to be very useful.
Attempting to impose semantics world wide reminds me of an article by Mehar Omar Khan, Don’t Try to Arrest the Sea: An Alternative Approach for Afghanistan. I comment it to you for reading but in summary it advocates what is known in some circles as the oil-spot strategy.
That is to create safe havens that offer benefits to the local populace and use those to attract others to the same benefits.
Topic maps, unlike some semantic strategies, have the potential to be semantic oil-spots. Their semantics are driven by the group or department where they are deployed and do not require consent or agreement beyond that range.
Which means that the group or department can begin to derive benefits from their use of topic maps, resulting in benefits that are not accruing to others. This allows topic maps and their use to sell themselves, rather than being imposed from the top down. (The FBI Virtual Case File project being a well known example of top down IT planning.)
Mehar Omar Khan summarizes his strategy as:
Don’t try to arrest the sea. Create islands. Having gone well past the phase of breaking the back of Al-Qaeda and dispersing the Taliban, concentrate on ‘creating and building’ examples. Set the beacon and you’ll see that all the lost ships and boats will come ashore.
Where are you setting your next topic map beacon?