Sun evangelist yawns about new Windows server
- By Will Kilburn
The release of Windows Server 2003, Visual Studio .NET and SQL Server was a big deal for Microsoft, but it drew barely a yawn from one spectator: Simon Phipps. Phipps, chief technology evangelist at Sun Microsystems, shared his thoughts about the releases and the differences between the two companies, during a phone conversation with ADT.
How does Sun's approach to developers differ from that of Microsoft?
The number one difference, I think, is that Sun believes in choice for developers, not phony choices, but the genuine choice of vendor, platform and deployment environment.
You'll hear Microsoft talking a lot about choice, but their choices are all choices that only allow you to run on Windows, only allow you to run on x86, and only allow you to use their products. So it's kind of [like saying], 'You can go anywhere you want within this cell, Mr. Prisoner.'
The approach that Sun in particular and the J2EE community in general offers to its customers, and to the developers who provide them with solutions, is a choice of any platform, including Windows. If you are a CIO and you've already locked yourself into Windows, that doesn't mean you are left without a choice -- you actually have a choice of using J2EE-based software, [and] you then have a wide choice of vendors who can provide you with both the development tool and the deployment environment.
I think the second way that we're different is that we're going with the industry trend of reducing customer's costs, whereas Microsoft is going against the industry trend and trying to get people to pay in advance, and to pay inflated prices and upgrade unnecessarily. The industry is trying to lower the costs, but Microsoft is heading in the opposite direction.
And just by offering more choices, you contend, you lower costs.
The fact that there actually is a free market rather than a monopoly for J2EE software means that it is possible for there to be a range from commodity through to specialization. And there is a range of capability available in development tools, and there is a range of deployment capability in application servers. And that range and choice and freedom means that the customer is able to optimize on cost if they want to.
So, for example, Sun is able to equip customers who want to deploy on Linux -- we can provide them with the hardware, the operating system and the development tools. Similarly, they can go to a vast range of other companies to get that software and hardware if they choose to.
The third way that we're different is that we are actually enterprise-ready for networked software. Our motto for many years has been 'The network is the computer,' and we're ready to deliver on enterprise networking; fundamentally, Microsoft is just not ready for it yet.
How far behind are they, do you think?
A: I think that they are still at least two releases of Windows behind [this]. They still believe that the sort of clustering, multiprocessor, distributed-load activities that are commonplace for people in the Unix world are considered to be new features for them.
What about the server rollout?
A: They have a run for their money this time because they actually have to fight against Unix. There is Unix on x86 now, and there is both Linux and Solaris x86, and both are gaining rapidly in popularity because of the way they open up choice and reduce costs.
Microsoft thinks that it's OK to take their customers' money in advance for products that they haven't yet delivered, and may not deliver. I would encourage them to continue doing it, because by doing it they've been driving customers to us. Our StarOffice product has been so successful, I think, in part because Microsoft has been driving customers to us. And once they come to us, they discover the product is actually pretty good, despite what they'd heard. What you buy when you buy StarOffice is a supported capability from us. So if you didn't want support, you would just go to the open-source Web site and download OpenOffice.org. And indeed, there are millions of people who just use OpenOffice.org, and don't come to Sun and buy the supported version of StarOffice. We don't mind that at all, because what we have found is that there are many customers who do want a supported product.
It's just the same with our development tools -- if you just want a Java development environment, NetBeans is a great Java development environment. But we do have a product called SunONE Studio, which developers who are targeting an enterprise J2EE environment will find they're getting good value for their money from.
We're quite committed to this approach of having a dual path: An open-source path, and then a supported product based on the open source. And you can expect to see more from us on that, particularly on the enterprise desktop.
So you don't look at the growth of Linux with dread.
Oh, not at all. Ultimately, the people who are writing the software have got to eat, and there has to be some model by which they're compensated. And all that happens in the open-source model is that the path for compensation is not as direct and visible as it used to be. Take StarOffice: We have found, to the surprise of some of our executives, that people are willing to pay a few bucks per box for StarOffice when they could get the same software free of charge for a download. And the reason they want that is that they want our name on the box, and they want our 800 number inside the box. So we make money from that.
In no way do I think Linux is a big problem. What I do see happening here, with Linux, is that there hasn't been a viable competitor on the x86 architecture up until now. Microsoft has had much too free a hand on the x86 architecture, and what Linux is doing is correcting that problem. We're going through a period of correction at the moment, and when there has been a huge imbalance, the correction may look like it's a tidal wave; but once the correction has finished, I think we'll see an even more vibrant Unix market, with a range of different flavors of highly intercompatible Unix. And we'll see market share gradually but inexorably drifting away from Microsoft.
I compliment them. They've done great work, doing what they think they need to do in Windows Server 2003. But unfortunately, they're very busily and very expertly solving the wrong problem.
To read an ADT Briefing Book about the often-heated competition between Sun and Microsoft, click here, or for an assessment of Windows Server 2003 from Summit Strategies' Dwight Davis, click here.