Along the way I have been reading about the progress of noSQL databases and how they organize data and indexing for both objects and content. And I have also read opinions and articles on BigData both here and in other pupublications The non-relational model with "tables" and "rows" is very interesting. Looking into what would be required to store the images as well as the associated text and metadata for each of the historical docs may lead to an interesting hybrid approach to a datastore. While the ISO document image that EMC may use is probably not searchable or able to be indeded, the associated metadata, keys, images, foreign keys and other elements might be very interesting and could be ground-breaking for this type of data collection.

What a huge educational opportunity this is. To make this collection of history available on one datastore is incredible. To that end I have heard some say over the years that bigger is not always better. But in this case the bigger storage capacity is really going to provide something very valuable.

Thanks for the insight. It is great that current innovations on the technical frontier can better expose the historical path we have taken to get to this point.

M.