Here is my long winded and probably naive take on the $500 PC, PClite,
Interpersonal PC, World PC, Internet Appliance or whatever you want to call it.
WHAT IS IT?:
An Internet Appliance (IA) is basically a device that serves as an
interface to the internet. It has a few things in common with a regular PC
in that it has the same basic components such as a display, a microprocessor
(could be an X86 or some other chip), and some memory. The difference is that
the IA is optimized to do one thing and one thing only, and that is to act as
a graphical/audio interface to the net. In other words, to receive/display/playback
image data and audio data coming from the net, and also to transmit user input
(keyboard ,mouse, microphone) out to the net. Because it is designed and optimized
to perform a specific function, it costs less by virtue of the fact that the
function it perfroms requires less ram, less powerful microprocessor etc.
WHAT GOOD IS A LOW PERFORMANCE DEVICE LIKE THAT?
There is a misconception that an IA can't possibly DELIVER the performance of
a $2500 multimedia PC. Wrong. By itself, the IA is nothing to get excited about.
But if the IA is hooked up to a 300 MHz Quad P7 server over the net, it will
probably run Microsoft PowerPoint pretty damn fast. The IA itself does not
run the application. The remote P7 server runs it. All the IA does is
display the graphics that this remote server is sending it. (If you are
familiar with X-terminals, the basic idea is the same.) Obviously, if the network
can't shuttle the data between the IA and the server fast enough, then you will
have lousy perfomance and its the World Wide Wait all over again. So, as George
Gilder and Scott McNealy have been saying for some time now, the network is the key.
Without a fast affordable network connection, the IA idea ain't gonna fly.
ARE THERE HIGH-ENOUGH BANDWIDTH NETWORK CONNECTIONS AVAILABLE FOR THE HOME TO
MAKE THIS IA THING WORK ?
Certainly 14.4 Kb/s won't cut it. ISDN might work at 128 Kb/s but will
probably be marginal especially for multimedia intensive applications. Cable
however may be the way to go for now. Check out @Home's website at home.net.
In 1996 cable operators will begin deploying 4 Mb/s internet access. That is over
200 hundred times faster than a 14.4 modem and 30 times faster than ISDN. The big
question for internet access via cable is cost to the consumer. The cable companies
claim they can match ISDN costs. This remains to be seen. IF they can, then cable
companies will do very well, ISDN might have a short life and this IA thing might
really take off. With Sun, IBM, and Oracle pushing the IA or similar concepts,
there is a decent chance that this will happen.
WHO IS GOING TO PROVIDE THE SERVER AT THE OTHER END OF THE NETWORK?
Maybe AOL, maybe UUNET, maybe AT&T, or maybe IBM. Could be a lot of companies.
What does seem likely though, is that computing power and resources will be looked
at more and more like electric power: as a utility. You plug into the wall to get
at the power. No need to generate your own power. Let the utility company
do it for you and have it piped in to your home.
DOES THIS MEAN THE END OF WINTEL?
The IA idea has nothing directly to do with whether Microsoft or Intel
will fall by the wayside. Certainly, the PC will not just suddenly fade away.
But what the IA idea does mean is that the playing field could change dramatically.
If you had a choice between a $2500 multimedia PC and an IA device that would give
you all the capabilities and performance of that same multimedia PC for HALF
the price, which one would you buy? Furthermore, if that IA device meant you would
no longer need to concern yourself with hard disks crashes, back-ups, memory strips,
hardware/software incompatibilities, and having to upgrade hardware/software, which
one would you buy? This is the big carrot that is being chased with the IA idea,
and why the IA idea seems so compelling to some.
Comments? |
