SO ...
I've neglected my data "Harvesting" over the last few months in favor of developing a shared web-based front end for the data (See developments around TCC).
But the problem with this data, is that there is always more to harvest. Turn your back for a week and there are 10 new products. Each with their own concaphony of inserts, subsets, parallel sets. 1/1, etc.
For someone working on a database for them all, its a challenge.
Working on an interface that the public and manufacturers can add missing ones is the way to go ... but who will participate, who will verify ?
should it be wikipedia style .. or IMDB style ?
more to come on these adventures ....
-Jim