My Photo

Your email address:

Powered by FeedBlitz

January 2015

Sun Mon Tue Wed Thu Fri Sat
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31
Blog powered by Typepad

Become a Fan

« 110th Congress Debates Data Mining | Main | The Next Generation of Network-Centric Warfare: "Process at Posting" or "Post at Processing" (Same thing) »

January 25, 2007


Feed You can follow this conversation by subscribing to the comment feed for this post.


I think you have some very solid ideas, but your concept of federated search failing due to lack of computational cycles could be avoided with proper architecture. All searches don't have to happen instantly (Google model). If an asynchronus approach is taken, allowing search terms to be executed during low demand periods, federation can begin to scale on a far larger basis. The issue then is building the repository that stores and forwards query terms based on business rules, but that is very solvable. Having this infrastructure would also allow analytics based upon proximity of query terms from researchers (identifying collisions of interest thereby potentially reducing cross-organizational effort and decreasing decision cycle time). The only drawback is some latency. You must recognize that some latency is far superior than what we have today or waiting for perfect. As we used to say at DARPA, perfect is the enemy of good...

The comments to this entry are closed.