CNET también está disponible en español.

Ir a español

Don't show this again

Christmas Gift Guide
Tech Industry

Microsoft's point man on Linux

Platform strategies manager Martin Taylor recently got the nod to help pull together a companywide strategy for dealing with open source. What's on his mind?

Around Microsoft's campus, some colleagues jokingly refer to Martin Taylor as their Joe Friday. The story goes that whenever they begin gushing about the company's latest product or technology, he routinely demands the data that would clinch such claims with customers.

In the software business, where companies routinely hype their newest releases to the heavens, Microsoft is no stranger to hyperbole. But the gathering momentum behind Linux provides Taylor, the software maker's newly appointed general manager for platform strategies, with extra incentive to promote his just-the-facts, ma'am, approach.

Three weeks into the job, Taylor has been given the monumental task of coordinating a companywide strategy to support its proprietary technology so that Microsoft can counter the open-source camp. The software giant needs to make a compelling case for cash-strapped IT buyers to buy its story. And skeptical CIOs have heard it all before. Taylor knows this is going to take more than a selling job.

But Taylor, who joined Microsoft in 1992 in Washington, D.C., says the battle will ultimately turn on making sure customer satisfaction is more than lip service. That has sometimes proved more difficult than it might sound at first blush.

"Most of our folks come up through the development ranks and don't have a whole lot of experience managing customer relationships," said Taylor, who worked his way up the ranks the last decade holding a variety of posts involved with customer and partner satisfaction issues--most recently reporting directly to CEO Steve Ballmer as his director of business strategy.

At this point, Microsoft's strategy for Linux remains one of studied ambiguity. Under pressure from governments and customers, the company has borrowed some aspects of the open-source development model with its so-called shared code initiative. For the most part, however, the company's argument comes down to a strict total-cost-of-ownership comparison, with a checklist of Microsoft features and services set against a rival Linux offering. That's an argument Taylor believes he can win.

A self-described closet geek, Taylor used to break apart his Commodore 64 and rewrite his own games. "But I used my hacking skills for good, not evil," Taylor said. He spoke with CNET News.com on the eve of the LinuxWorld Conference and Expo in San Francisco.

Q: Considering Linux's momentum, it doesn't seem Microsoft has yet come up with an answer on how to turn back the tide.
A: This may disappoint you, but I hope we don't say, "This is why Linux is bad and Microsoft is good." That's not the direction that I want to head. I want to say, "Listen, companies are now taking decisions across the stack. They might deploy Linux as a single-purpose server; they might move off Unix and try to deploy Linux in a variety of ways; but at the end of the day enterprises are looking at the stack." They either want to move Linux up the stack or they want to do different things.

Put another way, then, what are going to be the key issues when you have those conversations with customers?
I see my job not so much as to beat my chest and tell the world why we're great, but rather as moving this to a fact-based discussion that will be validated by third parties in a variety of ways. I look at cost and security as the top two things

So then what will be your role as far as figuring out Microsoft's approach to Linux?
Job No. 1 is to orchestrate our cross-company efforts as we talk about our platform and stack versus the Linux stack. And here there are really two things to consider: the complete open-source community free Linux stack, which is Apache, Samba; and the commercial Linux stack, such as WebSphere or Oracle, running on top of this kernel. The conversations in the company now are to make sure we're thinking about this collectively.

With the emergence of "Lintel" (Linux on Intel machines), has that, in turn, increased price pressure on Microsoft?
Software acquisition is 3 percent to 5 percent of the total cost--

I know. But on that one narrow part of the equation, is it having an impact on your pricing strategy?
Do we have competitive factors in the marketplace? We always do. It's not just Linux. We have people who have always put pressure on us in a variety of ways. This is perhaps the broadest phenomenon that we've had to think about across the company, from a pricing point perspective.

Six to nine months ago, we were hearing about shared source and how Microsoft had tried to borrow on that model. These days the

Ten years ago in our wildest dreams, would we have thought that people might not buy Microsoft because of anti-American sentiment? We didn't think about that.
message seems to be that commercial intellectual property is good, that someone owning the technology is good. Is it no longer that important?
No. The shared-source initiative is an example of us learning and moving from a position of "nobody but us shall ever see our source code and that's the end of the discussion" to one where customers and independent software vendors may want to see our source code. This remains a very important part of our strategy.

Microsoft prefers to concentrate on the total-cost-of-ownership issue when you're making the comparison with Linux. But sometimes Microsoft has been willing to do whatever's necessary on price--recently you did that, but the city of Munich still went with Linux.
You're right in that there was a decision not based on technical direction or price but more on relationships and political sorts of things. It was a validation that we have to build better relationships with governments at a variety of levels and not assume they will take decisions based on traditional decision-making processes, such as value, cost, technical direction, stack, etc.

It sounds as if you believe Microsoft needs a foreign policy.
It means we need to build better relationships with governments at all levels. Honestly, we're a company of geeks. We're a company of folks who build really good technology and think that people will make decisions based on the strength of that.

But it's more than that. Microsoft also is a symbol of America's high-tech prowess and that may not play well in some countries because of political sensitivities.
It's a new set of challenges. Ten years ago in our wildest dreams, would we have thought that people might not buy Microsoft because of anti-American sentiment? We didn't think about that. Did we think that they might not buy Microsoft because they have this illusion they can build a local software economy using noncommercial software? All this has presented a new set of challenges, but in some way also a new set of opportunities for us to get smarter and better and understand the global scene better than in the past.

It's also forced us in a good way to become sensitive to global issues. For example, with companies that have very low gross domestic products, we have to think about the way they use and acquire software differently than a G7 country might.

Is Microsoft going to offer bigger discounts to those categories of nations compared with what you might charge an industrialized state?
I was responsible for the Dominican Republic, and the country's president and I flew to Redmond, (Wash.), to meet with Steve (Ballmer) and Bill (Gates); and it was interesting because that was a real life case study for me. We started the discussion with him saying, "I can get all this stuff for free and here's this big American company wanting to charge a bunch of money." And as we began to educate him a little more--you take 5 percent off the budget but you've increased by an X factor all these other scenarios around training, development and administration--and at that time it was even more difficult to administrate those environments.

But just to clarify, are you saying that Microsoft will not have an A price and a B price for First World and Third World countries?
No, we have to look at it. There are some challenges there...this challenge of how do you properly value and price software is a tough, tough job. We know we have to be relevant in all these markets and we know we have to be receptive and understanding of market conditions in a variety of ways.

Do you think you'll have decided this by the end of the year?

It will be our stack versus (Linux's) stack and security and cost and then the value of integration.
No. It might not be one big bang thing. But I think that over time we'll begin to do a series of things to work with governments and underdeveloped countries.

Have you run across countries where the feeling is that they are just not going to consider Microsoft because of the larger political context?
I don't know of any government that's said, across the board, they are going to use open source and nothing else. We're spending a very good amount of time with a variety of these governments. At the end of the day, we will be ready to do what they need us to do.

How is Longhorn going to improve your position against Linux?
Actually, I don't look at it that way. You talk to customers and partners and deliver value. It will be our stack versus (Linux's) stack and security and cost and then the value of integration--Longhorn or not--that's there.

Is this a reflection of a different culture at Microsoft? You used to be enemy-centric.
Longhorn really should help us outpace other folks from an innovation standpoint, and that's where we're putting a lot of muscle. But we're doing it because this is what customers want to see from us. They want the scenario that we're building.

But every software company says it's providing value to customers. With Linux, you also face the reality that the product is good enough and what's more, it costs nothing.
Customers have accepted the fact that this is not about the acquisition costs. They're more concerned with the 95 percent to 97 percent that goes into things like management and training. That's why I'm not worried at all whether Linux costs less.

If not cost, then what does worry you about Linux?
I'm thinking about how we can tell our story to tell the difference between our stack and their stack. And you know what? It's not about religion anymore. It's not a new thing that you can throw in a Microsoft guy's face and see if he jumps. It's a commercial discussion. People are looking at the total cost of ownership. They are looking at the security model. They're looking at reliability. They are looking at interoperability. They're taking decisions the way they take decisions on mainframes, Unix, Novell and Oracle--that's where we know how to operate. That's how we really know how to show the value of integrated innovation.