I think many of us watched in fascination as IBM's "Watson" machine competed with some of the best Jeopardy players of all time. Even Jay Leno was talking about it but in case you missed it here is the link:
To see the deep analytics, understanding of English and massive amount of unstructured data all come together to answer imprecisely worded questions in under 3 seconds was truly shocking.
In my eyes it redefined the "Art of the possible"
But I think the real question is what is the new normal for analytics? What does this mean? With this proof point what should be our expectations for business analytics?
And I think it is important to keep in mind that in many ways the most impressive feature of the machine is the fact it interacts with us and gives us an answer.
We do not have to alte
r our behavior to interact with it. It is not google -- we don't stress over the right group of keywords and then hope the right answer is near the top of the 10,000 results google returns
(and I think it is time to stop thinking of the number of results returned as an accomplishment -- it is time to start thinking of that stat google returns telling us 10,000 pages returned is really just a failure to understand my question and therefore you are throwing stuff at me and hoping I can sort it out)
When you make this shift and realize we can now develop machines that interact with us, in our langauge, to give us precise answers to questions -- not a list of things a computer thinks might be close -- then you are at a whole different level of computing. A different level of analytics.
I'll explore this more and provide some details about the Watson machine at the Orlando SAP SCM Insider event. This will be in IBM's innovation room - I hope to see you there or tweet me for more details. @moosechris is a great way to reach me on twitter...