Monday, November 17, 2003

I found Twilight of the PC Era entertaining but it was the d�j� vu thing: "The end is nigh, everything as been invented, now we can go back to sleep …" Or, to quote Steve Levy quoting Nathan Carr: "The big innovations are over, the low-hanging fruit has been picked and 'the IT buildout is much closer to its end than its beginning,'".

While VisiCalc was a great hack I'm actually more proud of home networking. Back in 1970 we "knew" that in order to get executives to use computers we had to put them in wood grained cabinets and get rid of the keyboards. No one figured out that the answer might be to just make them useful and that those executives had to learn to type to get through college. Though maybe the CEO of 1970 grew up in an earlier era when typing was viewed as beneath them just as some now view understanding technology with disdain.

In 1995 we had a similar problem -- there was little reason to buy a PC or, at least, more than one per house. In the corporations we were supposed to stop this PC nonsense and put dumb terminals on peoples' desks. To the advocates of the lamed devices we should be treating employees as children and control them tightly and not give them a device that allows for initiative. The movie "9 to 5" was a parody of this kind of thinking (I'd give a URL but in 1980 movies didn't have web sites).

My goal with home networking was to remove the mystery and allow people to connect the PCs without having an MIS (it was called Management Information Systems before being recast as Information Technology) department in each home or department. You just plug things into the net and it just works. Well, sort of, but at least a lot better than it did.

The transition from MIS to IT is telling. MIS was a service that used computers for you. IT supports people's own use of computers but it only goes so far since computers are treated as dangerous and easily infested with those virus things that scare people. The key to the success of the PC and the Internet was that they made it OK to make disruptive mistakes and thus you could experiment and those experiments that worked were preserved and formed the basis for the next experiments. Even a buggy program like Word could succeed because you lose relatively little work after it restarts but still accomplish far more than with a program that is too timid.

The problem is that we are in a timid phase and are locking down machines and cowering in the corner rather than making it safe to make mistakes (including running malicious programs). We've gone from relying on the MIS department for answers to an IT department which supports tools but is also doing the MIS work for users. We need to go further and give people the ability to create their own solutions and, even better, solutions they can share with others. Part of this is making it safe to experiment which includes getting beyond the naive model of treating trust as all or nothing. Just as we can undelete files we should be able to unrun errant programs. More important is educating people in the concepts they need to take advantage of these devices and beyond that, making the concepts a fundamental part of literacy.

We are simply in a phase of trying to digest what we've got. Some curmudgeons see this as proof that we should go back to the past or at least stop rushing headlong into the future. We forward-looking curmudgeons argue the problem is that we are just at the beginning of the beginning. But then, isn't that always the case? The biggest threat is a failure of imagination and a retreat into comforting somnoblesence.