CNET también está disponible en español.

Ir a español

Don't show this again

Tech Industry

Ballmer: No sleep lost over Linux

Microsoft's CEO talks about the software maker's bid to gain a stronger foothold in corporate data centers, and he discounts the threat presented by the Penguinistas.

Will IT buy his pitch?

Steve Ballmer had the stage to himself Thursday in San Francisco as he introduced Windows Server 2003, a new version of the company's server operating system that Microsoft's CEO described as "the right product" to help companies stretch their IT budgets.

Just how long fence-sitting CIOs will take to heed that message and upgrade their existing servers is the big wild card.

Clearly, Windows Server 2003 is the most ambitious version of the company's server operating system. Microsoft, keen on gaining a stronger foothold in corporate data centers, plans to spend as much as $250 million to promote the product. Though the software took four different names and four release dates to get here, Ballmer says that Windows Server 2003 offers a heap of new features designed to appeal to security-conscious corporate buyers.

But in raising its sights, Microsoft also must grapple with a new constellation of challenges--not the least of which is the increasing popularity of Linux among IT administrators who manage corporate servers.

Ballmer, playing up to his legendary image as a hard-driving salesman, did not mince words in discounting the gathering threat to Microsoft's ambitions posed by the Penguinistas. Although Microsoft has distanced itself from Ballmer's earlier criticisms that the license governing Linux makes the software a "cancer," the CEO still believes that Linux is qualitatively different and that its license poses insurmountable problems.

Sitting down for an in-depth interview with CNET, Ballmer discussed the Linux phenomenon as well as a range of other issues facing Microsoft.

Q: Microsoft has beaten companies touting free software before. But as one of your former executives pointed out, Linux is a completely different kind of free. There is no single company promoting it. There are people voluntarily coding for it. Do these aspects change how you compete against them.
A: We have competed with things that had no price attached with them before. There is a clear set of guideposts for adding value to customers to differentiate you from the guy who has no price or a lower price. It is a different model in the sense that there is no commercial company behind it, but I think that winds up being an advantage for us, rather than a disadvantage.

In what respect?
Innovation is not something that is easy to do in the kind of distributed environment that the open-source/Linux world works in. I would argue that our customers have seen a lot more innovation from us than they have seen from that community.

Linux itself is a clone of an operating system that is 20-plus years old. That's what it is. That is what you can get today, a clone of a 20-year-old system. I'm not saying that it doesn't have some place for some customers, but that is not an innovative proposition.

Some people say it is an advantage that Linux gets built in all of these little pieces. The fact is that if you want to do some kind of integrated innovation that touches the kernel, that touches the user interface--there is no way. Maybe Linus (Torvalds) can control the innovation in the piece called the kernel, but there are many pieces.

The Linux world in some sense is a lot like the Unix world. There is not much communality. There is this distribution; there is that distribution. There is this user interface, there is that. Some people might see some advantages to that. On the other hand, in terms of putting a clear, simple proposition in front of the customer, I think we have a leading edge proposition.

Linux itself is a clone of an operating system that is twenty-plus years old. That's what it is.

So when it comes to development models, you're claiming the edge?
If you want a fix now, we may need to perform better, but you know where to go. There is nobody to turn to if you as a (Linux) customer says, 'I need this.' You can't turn to IBM. They don't write the thing. It's not like IBM can support Linux the way they support the mainframe operating system. They don't write the code for it. All they can say is, 'You can call us and ask us a question, but if you actually want something done we can't do it.'

But why do you think people are adopting Linux? Is it because they can look at the code? Is it because they don't have to go through an 80-page licensing agreement?
No. There are some scenarios where people consider it. People don't really consider Linux much on the client--that's my market observation. On the server side, you have people who have skill sets and applications on Sun that they want to move now to Intel hardware to save costs. I think we have a pretty good story, but I tell you, game's on. We've got to prove ourselves, and some people are choosing Linux. I don't think that is going to continue to be the case.

In the past you've used strong words like "cancer" to describe Linux. There is the un-American comment from a colleague about GPL (General Public License) and open source. Are you backing off from that position and taking more of a technological or business view?
I think there are many parts of the discussion. I do think there are things that people don't understand very well about the new alternative, where it is important for us to help customers understand the issues.

Steve Ballmer in an interview with CNET The way things are structured today, from a licensing perspective, in the Linux world nobody will ever commercialize Linux the way the Sun commercialized FreeBSD. For some customers, that can be viewed as advantageous. But customers will never really know who stands behind this product. If the lead developer for this component chooses to do something else with his life, who will carry on the mantle for that? The fact that it will never be commercialized is assured by the GPL. The GPL licensing form does that, as opposed to the open-source license for FreeBSD, where you could say Sun took it and commercialized it and can say that they own it. Nobody can ever do that (with GPL).

There are advantages (to Windows) that are more subtle. We may not have always been sharp in the way we have communicated about those (laughs), but there are some things that are important for customers to understand. We think that in some sense our commercial form is a major asset for us.

Gauge the piracy problem. Are you finally getting a handle on it?
Some countries yes, some countries no. It has improved certainly in Europe, particularly in Eastern Europe and Southern Europe. Piracy was high at one time. There are still challenges in parts of Asia. We have seen improvements in Latin America.

What is driving it, legal enforcement or a growing PC industry that wants to make money?
Both. It really takes both. Neither one nor the other by itself is generally enough.

On the desktop, in developing countries, computer dealers will tell you that they sell Linux-based PCs, but in a lot of these places, Windows only costs $1 in the street.
A Linux PC in most countries is a PC in which somebody is being encouraged to pirate Windows. We conducted some surveys about this in one large Asian country, where we found that, of all PCs that didn't have Windows installed on them when they were sold, 99 percent wound up with Windows on them within 30 days.

Who is the target audience for Windows 2003 server? Is it Windows 4.0 users? The Unix crowd?
The initial market will probably evolve into three categories. Category No. 1 will be NT 4.0 users. There are a lot out there, and I think we offer an important step forward. It's not like a client, where everybody might want to upgrade at the same time, but I think we've set up a real wave there.

No. 2, there are people moving applications from expensive gear onto cheap gear. I think Windows Server 2003 is going to look good to a lot of that community. Those are high-performance computing applications, or applications running on Solaris or AIX.

And No. 3, people who want to put in new applications or people who want to support new working information-type scenarios--and people are always putting in new applications. They really are.

There are millions of new servers sold a year, and that is a market that continues to grow. It grows faster than the PC market does. By hook or by crook, so to speak, there will be 5-plus million servers, roughly, sold in the next 12 months.

Over time, will servers become 50 percent of your business?
It's not a line of dialogue I choose to engage in because then all you are doing is comparing the growth prospects of various businesses. I think we have great growth opportunities in our server business. I think we have great growth opportunities in small business and medium business, as well as the enterprise. I think the percentage growth rate in that (the server) business will exceed that which we see in the Windows client business or the Office business. But those are huge businesses. This is a very large business, but those are huge businesses.

Shifting gears here, when you look at your balance sheet, Office does well, Windows does well; but there are these other projects, such as MSNBC, Xbox, that don't have that same sort of trajectory. Why is that?
They absolutely have a trajectory we are used to.

I don't think the price of software and the price of hardware have some inextricable link.
Are you happy with the growth of Xbox?
Yes. We're a clear No. 2 in the market. We are coming on strong. It is probably going to take us another turn of the crank, from a product cycle perspective, before we make money. But most of the things we do as a company successfully today we worked at for years before they made money. Remember, we brought Windows 1 out in 1983 and we didn't have any real volume until 1991. It took us eight years to get volume. I don't know when we got profit, but it took us eight years to get volume.

Take Windows server. We started on it in 1988, but it was probably 1998 before we had real volume, and I don't know when we would have said we had profitability on that product. But most of the good businesses require long-term patience, commitment, tenacity...and you can't be impatient. I feel very good that we have great teams to take MSN and Xbox in exactly those same directions.

Are you looking at search?
We are in the search business today. In fact we have a lot of search users. If you take a look at search usage today--everybody likes to talk about Google, which is fine. They are doing a good job as a company. But for traffic, Yahoo is doing quite well and we are doing quite well. We've got some of the best people in the world in the area of search thinking about information retrieval.

If you put Office on a PC, it can be one-third of the material cost of the system. Is that sustainable? Hard drives are going down in price and processors are going down in price.
I think that is a bad way to look at it. I don't think the price of software and the price of hardware have some inextricable link. I think what we need to make sure of is customer perception of value versus competitive offerings. I think we've got the right mix of capability, functionality, simplicity, price, etc. I don't think looking at it relative to hardware prices takes you any place.

And I also don't think hardware prices have come down, at least at the client. Hardware prices have not come down significantly in a number of years...The capability goes up, as the capability goes up in our software.

Average selling prices are pretty far down.
No, no, no. Not in the home. It hasn't come down in the last several years at all. Remember when sub-$1,000 PCs were all the rage. The percentage of sub-$1,000 or $500 PCs is not significantly different today than it was several years ago. There is more capability every year for the price, but the same could be said for Microsoft Office 2003.

The Licensing 6 program was announced nearly two years ago, and there was a big stink about it. Customers weren't happy. Has that subsided? And do you foresee any further changes?
I think we've learned a lot from the experience, and I think the most important thing is the lesson of consistency. Most of the issue was that the new thing was different than the old thing. What we learned is that customers don't want us to change things very often.

We've worked very hard with our customers to get them through the transition, to educate them on the new opportunities, and make sure they are getting good value for their money. Certainly, we've seen that even customers who thought they had issues with licensing 6.0 have been able to move to the new licensing programs more successfully. We always make small tweaks. You'll see that we've tuned up some of the licensing conditions to be more clear and more favorable to the customer. But we are going to show consistency.

Licensing 6-- the thing that was significant about it was that it was a very different way to think about the licensing of software to business. No, I don't anticipate making a change of that ilk in the foreseeable future.