The Wall Street Journal Interactive Edition -- December 6, 1999
Special Report: Convergence
Sun's Chief on 'Web Tone'
An Interview with Scott McNealy
Talking tech with Scott McNealy can be like discussing theology with a TV evangelist. He's got a
point of view, and lets you have it.
For the past few years, the CEO and chairman of Sun Microsystems Inc. has been one of the most
forceful preachers for the religion of "Web tone" -- the idea that the entire computer industry will
reorganize itself to provide services as easy and cheap as the phone industry does with its simple
dial tone. No more fiddly, crash-prone personal computers loaded with expensive software;
instead, lots of cheap, easy-to-use appliances that get their jobs done by connecting online to
bigger software "servers" and switches.
In an interview with Convergence's Richard L. Hudson at the Telecom 99 trade show in Geneva
six weeks ago, Mr. McNealy explained how it will all work. He started off with an interactive
sermon on the phone industry's virtues -- as exemplified by the reliable switches of Lucent
Technologies Inc. and the easy-to-use cell phones of Nokia Corp.
Mr. McNealy (waving his cell phone): What is the OS on this? Do you know?
Convergence: The operating system? No, I don't, actually.
Neither do I. Do you know the Lucent OS? Do you even know the name of it?
No.
Do you care?
No.
Right. When was the last time you added an application (to your phone?) What was the last time
you bought some software for your telephone switch?
Never.
What was the last time you backed it up?
Never.
What was the last time you hauled a systems integrator, like IBM Global Services, in? Or brought
armies of people to do call-forwarding software for you?
Never.
Lucent sells you a complete package on the switch, that has all of the services bundled in,
seamlessly, reliably, availably and scalably. So you get dial tone, call forwarding, call waiting,
unified messaging, voice mail, voice recognition, blah blah blah blah blah. They have got thousands
of features they offer on a switch. And there is no third-party software industry like you see in the
shrink-wrap computer world. Same is true with my cell phone (waving it again). There's lots of
(software) code here, that I get for free. I just buy the hardware. So the model of selling software is
gone. This is the functional equivalent of the dial-tone switch: This is the Web-tone switch.
So, the whole idea is: This is a service. We are going to have the mail
directory, calendar, encryption, news; but we're also going to have
word processors, spreadsheets, presentation graphics, collaborative
software, all of the rest of that, on the server, as services. And we sell
these products to any of the telephone companies, to any of the service
providers. So the service provider is going to take this switch, this
Web-tone switch, and offer those services that we expose to his or her
customer. They are going to charge for it based on what they think the
market will bear.
So how do we make money? We sell servers. By the bazillions. We run
all these different services, and every time a consumer signs up for the
server, the disk drive gets a little bigger, and they get a couple more
CPUs. And after they add up more subscribers, then they have to buy
another server. And then another server. And then another server, and
then another server, and then another server. And we provide services,
support and consulting and all the rest of it.
But this kind of telephone-industry model is always criticized as rather rigid.
But it works. This building is on fire. Are you going to log in and e-mail the fire station, or are you
going to pick this (cell phone) up and dial 112, or whatever the number is here? That is awfully
rigid, but you are going to pick up the phone and say: 'The building is on fire.' You are not going to
e-mail, because this (computer) environment has too many components: The company buys the
hardware, finds some version of the operating system, but will have to get a different file system.
Then go out and buy some middleware, and buy some application from yet another company,
some storage from EMC, and some service and support and integration from yet another company
-- and then they can't figure out why it doesn't work!
How will the computer industry reorganize to deliver that model?
It will look like Sun, where we have the microprocessor under our control, the operating system,
the user interface, the networking, the middleware with all of the Netscape stuff, and all of the
features, like mail, calendar, news, electronic commerce, all the rest of it.
Doesn't that make you like Alcatel or an Ericsson or ... ?
... Or Nortel or Lucent? But instead of doing dial tone, we do Web tone.
But why didn't this work first time around? This was the "network computer" (that didn't
sell well when introduced three years ago).
Three reasons: One is we hadn't matured the Web protocols to be open. Two: We didn't have a
Java browser. And three: We didn't have a language with object-oriented environments, so you
could have downloadable software. We didn't have the network speed and power and reliability
yet.
Will the diffusion of this technology differ in Europe vs. the U.S.?
Oh, you guys will probably drive on the wrong side of the street or some darn thing like that. |
