...surface is barely getting scratched.
Palmisano has touted this idea of process improvement, where your business consultants can analyse a business process by breaking it down into its component parts. Then you can tell them how best to streamline their processes. Can you explain?
That piece of the business tends to be more in consulting practices and in what we call business performance transformational services. This is really one of the single biggest opportunities for IBM.
Think about it this way: Companies have some inherent inefficiencies in the way they work, and they do certain processes themselves which they could potentially get from the outside. Or they don't integrate external processes and services particularly well with their internal stuff, and all that leads to inefficiency.
Just think of the gross domestic product of the world with all these companies operating in some inefficient manner. If you could squeeze a little inefficiency out by detailed analysis and modelling of how a company operates — there are huge opportunities there. By the way, you can be very innovative and create a new iPod and you're talking about tens or hundreds of millions of dollars. And here we're talking about hundreds of billions of dollars, so the opportunities are really big.
What are some of the things you're excited about in hardware?
BladeCentre. That's a huge, huge opportunity. One of the big research projects [here] is exactly in that space: to make it simple for people to just plug in accelerator boards, and your application will run faster. No software, no nothing — hands off by the customer except one plug, and all of a sudden the whole system runs faster. That's harder than you might think.
Last year, IBM bought DataPower and the idea of the founder, Eugene Kuznetsov, was to put middleware functions into hardware, such as XML acceleration. Do you think we'll see more things that might have been done in software being built onto specialised chips and stuck into these blade racks?
Absolutely. Actually, a lot of DataPower's strength is its clever use of software. So it's not really that it's unique hardware as much as it is unique software. It is applied in a box or in an appliance that you can just plug in, in the front of your servers, to speed up your XML processing. You're going to see more and more of those sorts of things.
There are a lot of underlying reasons why that's going to happen. Traditional silicon technology is running into power limitations and other problems. There are a set of things that are going to make it more attractive than ever before for having special purpose [appliances]. There have always been accelerators. The problem is that by the time you program the damn things, the general-purpose processors got better. So you had all applications running on general-purpose processors and except for niche markets, accelerators never made it. What's going to happen now is that the general purpose ones aren't going to accelerate so fast, and the technologies and tools and software for integrating the accelerators have gotten better. So you're going to see a lot more special purpose stuff that you can just plug in simply.
One of the high-profile projects that came out of IBM Research in the last couple of years was WebFountain, which was called "Google for the enterprise". Where are you going with that?
It continues to be a big thing for IBM and for IBM Research, but it's not just WebFountain. The basic issues are, really, natural language understanding in general. What WebFountain was able to do, which made it powerful, was it would go in and would scan text documents on the Web and it would understand enough about what people were saying that you could query it about what people were saying. You could imagine that there are a lot of countries... that would care a lot about scanning documents and even open documents and crawling through them to see what people were saying. A lot of the early work on WebFountain was done in three languages — English, Arabic and Chinese — and you can guess who might sponsor that work.
WebFountain is an example of a natural language technology that allows you to essentially analyse from an intelligence point of view what people are saying, but the important point is that this is just a small piece of many, many problems that companies have and where you want to take advantage of natural language understanding, such as translating spoken English to Russian and back again.
We talked about call centres. Natural language understanding can be incredibly powerful, even if you've got a call centre operator, just by monitoring the calls and trying to understand what the issues are. There are enormous amounts of natural language and analytic issues in how companies interact with their customers. WebFountain was a specific application of natural language and search technology, but it's just one.
It sounds like you're very involved with customers, and IBM Research Centre makes a lot of money from licensing. Do you deliberately try to research commercially oriented products attached to specific product groups? How do you manage that process?
It's tricky, let me tell you. We try to keep a balance. We don't want to be like the old Bell Labs, which was exploratory research but no connections to the marketplace. We want to make sure that we build channels for the flow of our innovation and our ideas into the marketplace. We spend a lot of time thinking about getting the right balance — that is, you can't just be doing short-term market work and short-term development or you become a development lab. You can't just do corporate funded basic research or you don't have the channels for the flow of your intellectual property into the marketplace. So you need a balance, and everything we do in the division is focused on that balance. In the end, to me that's the big trick of running a research division: getting that balance right.
