I am doing some research for my marketing course and more specifically I'm looking at forecasting. This reminded me of a conversation I had last week with Juval Lowy of IDesign. He posed the idea that COM was circa 1993 and was replaced in 2001-an 8 year technology cycle. Given that .net is now 6 years old or so it could be replaced in 2 years, but with what?

Well Juval proposed service orientation. At this point most people go to sleep (heard it all before). But wait, what if services were everywhere - I mean what if there were so embedded into the framework everything could be service oriented. A log class could be accessed over a named pipe whilst a BizTalk service could be accessed via named TCP. Both could be aggregated, transacted, queued and instrumented without any manual work. 

All I can say is - its going to make for some large config files! That, or more attributes? And what would be called Next Generation Net Services (NGNT?).

Forecasting the "science of predicting future outcomes". Is it really possible. Not says Martin Banks here

My gut feel says you need to know an industry and simply buying the tool doesn't interpret the mining for you. Internal forecasting maybe is what I need to research? At Microsoft BI has been built into the product for some time, Oracle is now acquiring to position.