"The software revolution" (Cover Story, Dec. 4) was timely and informative--of great use to millions in various stages of trying to understand and/or participate in the Internet evolution.
A notable omission is the potential role of the communications sector. What happens if AT&T buys out Netscape Communications Corp.? What about the cable industry? We all have several potential Internet appliances in our homes now, and they aren't computers. While a telephone with a screen may perform a mini-applet, either that incoming "wire" or the one to the back of the TV could be the dominant conduit for most of our "highway" activity.
The Internet is developing rapidly and will no doubt spawn great changes in computing. But your article glosses over some major obstacles to this "new paradigm." Even if most PCs were equipped with the fastest modems (28.8 bps), downloading even small programs would still be time-consuming. Many of the hardware requirements for a new computing model will not be in place in the foreseeable future.
Your article also fails to address certain questions. For example, will most computer users really prefer scaled-down generic software that must be retrieved from the Internet each time it is used? Will a pay-per-use system requiring Internet access ultimately be cheaper for the consumer over the long run? What sort of fees will the new software providers charge?
There may be a sizable market for this style of computing, just as there is a large market for inexpensive cars with fewer features. But don't read too much into it just yet.
If you think about it, the model that Sun Microsystems and IBM are recommending--using a dumb Internet appliance and depending on a central software server--is not much different from what we had to use in the old days of mainframes. The model might work if we had cheap, high-speed connections available from our homes, but the infrastructure isn't in place and probably won't be anytime soon. Cable may eventually offer high bandwidth, but right now the system is set up to transmit data only one way, and it's going to be very costly to modify it so that it can handle computer traffic.
I don't know about you, but I don't expect to junk my PC anytime soon.
I am not sold on the expectations for Java as they are described in the article in your magazine. This concept does have a future. However, the idea that the need for a Wintel-like suite is eliminated when applets are combined with the information available through the Web relegates our society to just copy-and-paste. I do hope that we can continue to create--and not just creation by a few, but many. Creation will need more features than simply those in the least common denominator.