There is nothing more important. More important than what? What current preoccupation is that important? Well, it seems that important so therefore it is.
It’s a judgment call, and clearly not all the facts are known.
Largely, the computer industry is driven from very important factors. Much like life, there is plenty of activity both on the development and consumption. The desire for more is always there. The hunger for bigger and better goes on and on.
There is clearly a parallel with how automobiles were first developed. Initially cars were very complicated and costly. The user had to know every control and quirk. The industry was fairly small with not much sympathy for the amateur. Even today, this is still mostly true.
The truth is that computers are not important. The same is true of automobiles. It is what these tools can accomplish for the user that brings value. Functionally, different cars only add value based on what the drivers can derive as a benefit.
What I am getting at is that the computer experience needs to be much more transparent. Sitting at a monitor with keyboard and mouse is not conducive to being easy. The hardware needs to blend in with the environment to the point that the user/consumer does not even have to think about it. This is the mantra of ubiquitous computing. The difficult aspect is thinking outside the range of current thinking.
I was talking with my Dad over the weekend and we started talking about airplanes. In the 60’s to 80’s, he worked as a pilot and navigator. He pointed out that modern control systems on planes make it very difficult for a pilot to override potentially very dangerous automated decisions. We talked about centralizing the power in the computers, it actually takes away from the experience and thinking of pilots. From talking about this, it became clear to me that pilots should see the plane as an extension of themselves. The computers should be used like how the brain controls the body. The brain is the pilot and the body is the plane. Ideally, this would be true except for cases for when the pilot cannot process the information fast enough. Dad mentioned that planes have an automated system for landing. It is well known and a perfect application for planes since there is actually more risk having the pilot do it. He explained that a pilot could easily cause a “Dutch roll” in specific situations. If bad enough, this condition can be catastropic.
This comes back to an interesting idea. Since our computers are still very dumb, it is unwise to put too much faith in them in unknown situations. The lack of intelligence is often forgotten due to the speed of how computers figure out known problems. It is fine for it to focus on something that can be digitized and programmed, but completely helpless with sketchy data and an unknown state. Most faults in computer systems can either be tracked to hardware failures or software that could not recover from an unknown state. Lately I would tend to think that most problems are in software. It is very hard to write software that will work in all cases forever and ever. Things change.
So, this brings us back to what matters. First of all, we should not put too much trust in our existing frameworks to carry us forward to the more casual computing model. The hidden message is that basing computers solely on digital on/off switches limits their abilities considerably. By now, we should be looking more seriously at technologies that better duplicate intelligence. Perhaps the most revolutionary idea is that it is okay not to know things. Much like life, we have to accept that we don’t know everything and that the computer should be allowed to learn. This puts the computer more in the category of having a real brain. Obviously a computer also needs to ability to link multiple locations together (or unlink) so that it can more easily find connections. This should be done more like our neural networks and made to accept unknown states.
The best analogy is this. You are walking down the street and suddenly something amazing happens. For whatever reason, gravity no longer works. You have not been programmed 🙂 to handle this situation but you quickly try to find an idea that will help. If you were a robot, there is a perfect chance that you were never programmed to handle this. So, either your software faults or you just keep on doing what you were doing.
So, what would happen if gravity just stopped? I would guess that you’d be thrown (slowly?) off the earth based on inertia and the earth’s rotation. I don’t know. The point is that if it did happen, at least I would be trying something. At the very best, it is clear that I should not fault. Even if I did fault (pass out), most likely I would reset just fine (perhaps floating off to space).
I like the idea that people usually don’t fault. It implies that we are much too accepting of existing models and hardware.
Please excuse the tangents. It’s hard to think clearly due to my fuzzy programming :).
What really matters is that our faith should be shaken to usher in something new. In fact, given the success of the web, we have largely solved the act of exchanging information. What we have missed is realizing that many our designs are based on 1960’s (or even earlier) thinking.
We’ve come a long way. We still have a long ways to go. A key goal is removing the importance of the computer in how separate it is from everthing else around us. Things should be more specialized, simple, and transparent. It is a process that should be much more evolutionary than it currently is. Competing models should be allowed to win and lose based on what the environment allows.