Newsmaker: Mr. Chips takes a final bow

When he appeared at Intel's annual developer forum earlier this week, Craig Barrett took the stage for the last time as the company's chief executive.

But while the craggy-faced Barrett may be approaching official retirement, there's nothing retiring about his demeanor. At 65, Barrett's still full of spice and vinegar and has no intention of taking it easy. During a chat with reporters following his keynote speech, Barrett briefly took stock of his legacy and offered a look ahead at the technologies he believes will keep Intel ahead of its rivals.

What are you most proud about at Intel?
If you look at our market-segment share, you'd be hard-pressed to say that we've lost a lot of ground in the last five years, so I'm proud of that. I'm proud of the ability to have the Intel brand be recognized...as one of the top brands in the world. I'm proud of our international expansion, and I'm proud that we came through the dot-com implosion. I'm proud that we maintain our profitability. I'm proud that we didn't subject our staff to massive layoffs like essentially every one of our competitors. I'm proud that we continue to invest in technology and manufacturing capability. I think you saw the results come out in the second half of '04 and going into '05. That suggested that was the right thing to do.

Could you talk about the next step for you, starting with the shareholder meeting in May?

The fact is that the industry is not slowing down.

I get to move into the chairman's role and make much the same transition as we made in 1998, when Andy (Grove) moved up to be chairman. My role will be to focus on a few things for the company. Education will be one; probably some interface with the government on issues like R&D, funding and infrastructure in the U.S.

I still intend to do international travel and business development for Intel, like we've been doing for the last 10 years, but Paul Otellini will be in charge of the day-to-day aspects of the company. I'll go to all the strategy sessions, as much as Andy has gone to all the strategy sessions in the last five or six years.

So I am very involved with the company, but it is time to pass the day-to-day operational aspect on to the next generation.

Do you plan to appoint someone to take Paul's role?
Paul has structured the organization to have all of his direct reports go directly into him and probably will operate without a COO, as we've had Paul in that job for the last several years.

For an extended period of time?
Well, historically, we've used that COO-type position as part of the orderly transition of management. So I think Paul will be using the next couple of years probably to decide who to move into that role.

What is your advice for Paul, as he takes over your role?
My advice to Paul is the same advice that I give to the rest of the company: Intel is a technology company. Intel is successful, and it continues to invest for the future and continues to function on all eight cylinders. I'll suggest that he pay attention to those...things: Continue to move technology forward and...continue to make sure that we operate in an excellent fashion inside on a day-to-day basis.

How significant is the move to dual-core technology?
A dual core is a little bit more of an evolutionary change. We've gone from single-core to hybrid, threaded to dual-core, and then multicore, so it's kind of a more gradual transition. But I think it's an important

More Newsmakers

Is Moore's law validated or imposed at each generation?

In my opinion Intel has been busy trying to just match moore's exponential prediction although alternative technologies clearly pave way for a growth much more than what is predicted.

"it's the way that the industry is going to continue to follow Moore's Law going forward--to increase the processing power in an exponential fashion over time".

This raises some serious questions? Are'nt Moore's predictions the lowest that industry can take up?

[bluvg]

Interesting point, but...

There are a couple problems with the supposition that much faster technologies are, in essence, being blocked by a willful submission to Moore's Law. For one, Intel mainly deals with consumer-level technologies, or in other words, technologies that receive massive adoption. There is an entire ecosystem that would be disrupted by a wholescale shift to a new technology of the type you are suggesting, such that whatever gains that might be made on the performance side may not be felt on the consumer side until that ecosystem caught up with the change--and that may take quite awhile. OSes, drivers, applications, etc. would likely need to be rewritten.

But another point to consider is that willful submission to Moore's Law may be beneficial and perhaps necessary to sustain the industry. If a company like Intel tempers its technology introductions to the market to ensure a more stable growth rate (rather than releasing every technology advancement as it's ready), it can have the effect of stabilizing their cash flow (as long as they aren't falling behind), and this is good for the long-term health of the company--and arguably the entire industry. It's akin to the US Federal Reserve adjusting interest rates to control the growth of the US economy--boosting it when necessary, but also tempering it when it's too hot.