Watson in Your Hip Pocket: Win TV Game Shows Anywhere, Anytime
August 29, 2012
Autumn is fast approaching, and it is time for IBM to wheel out the Big Berthas of its marketing campaigns. I spotted the story “IBM Creating Pocket Sized Watson in $16 Billion Sales Push.” The headline snagged my attention. First, it pointed out that the $100 billion services giant wants to generate $16 billion more. A goal can be useful. Second, one of the products which will help blast through this sales target is Watson.
Will IBM Watson pull a rabbit from a hat or wear a hat which looks like a rabbit?
I found this passage particularly interesting:
The next version, dubbed “Watson 2.0,” would be energy- efficient enough to work on smartphones and tablets. The challenge for IBM is overcoming the technical obstacles to making Watson a handheld product, and figuring out how to price and deliver it. Watson’s nerve center is 10 racks of IBM Power750 servers running in Yorktown Heights, New York, that have the same processing power as 6,000 desktop computers. Even though most of the computations occur at the data center, a Watson smartphone application would still consume too much server power for it to be practical today.
Okay. No problem.
The fact that a noted technology expert like Ray Kurzweil is passing along public relations output which flowed from IBM to Bloomberg to Mr. Kurzweil is instructive. The Bloomberg story is “IBM Envisions Watson as a Super-Siri for Businesses.”
Several observations:
First, Watson won a game show. I questioned the credibility of a machine victory particularly when game shows are produced. The good old days of College Bowl on which I appeared in the 1960s were live. Today’s game shows are subject to lots of work by men and women in edit booths. I was okay with the stunt, but did it sell Watson? I don’t know.
Second, I called attention to IBM’s assertions that Watson would rework health care. You can find those write ups—“IBM Watson in Health Care,” “IBM Watson Still Chasing Health Care,” and “IBM Public Relations Chugs Away on Watson and Health Care”—in Beyond Search’s archive. I have lost track of Watson’s revolutionizing medicine.
Third, I have been skeptical about IBM’s claims that Watson slices, dices, and performs various works of retrieval magic in an affordable manner. The most recent analysis of Watson appears in my discussion of Watson in a monograph about open source search which is in press at this time. The full megillah will be available from IDC, the giant consulting firm, in the fall. You can get a taste of what we do at this link: LucidWorks Profile. Let me say that Fancy Dan systems are difficult to make into profitable businesses of the magnitude of IBM’s $16 billion. Autonomy was about a $1.0 billion when it was sold to Hewlett Packard. I mention this as a point of reference only.
Third, the voice search thing is viewed as one of the next big Star Trek things for big companies to do. The challenges, however, are intriguing. There is the issue of accents. There is the issue of ambient noise. There is the issue of a lousy phone due to the owner’s dropping it or spilling a smoothie on it. There is the issue of computational horsepower. There is network latency. There is confusing strings of speech which, if you consider your conversation with a colleague, is full of fits and starts and direction changes. There is the issue of lousy hardware in an automobile’s “smart” dashboard. You get the idea.
Will Watson slam dunk voice search? In demos, Watson will be impressive. In the real world, it may perform as the system did on Jeopardy.
Stephen E Arnold, August 29, 2012
Sponsored by Augmentext
Comments
-
- Subscribe to Beyond Search
Feature archive
News archive
-
