In order to organize any information you need a guiding principle. That is, information can only be structured when there is some other fact to evaluate it against. For example, “I am green” and “I am five feet tall” could be in the same category, different categories, or related categories, depending on what guiding principle we choose.

The danger here is slipping into mathematics without first a focus on practicality. Because “information” is an abstract subject, it’s too easy to treat it abstractly and not come up with something that would actually be practically applicable.

-Mxa

Yes, General Semantics looks interesting, I’ve just glanced at the wikipedia page, but it was enough to suggest that I might want to dig around there a bit.

In watching people build systems over the years, I’ve come to realize how often they are just guessing about the structure of data. They’ll convince themselves of a specific structure, but if another better one arrives, they’ll convince themselves of that one too. If we know how the data is actually structured, and we know what subset of the data we want, and the users know how they want to view it, then the rest is just connecting the dots.

Paul.

I suppose that the nature of the organization of information is so natural to me that I had never thought about it.

-Max

-Max

I’ve been thinking about this one for a while. I think your right about the missing science, but as in your example of physics to skyscrapers, I figure the problem is ‘information’ to programming. Something like:

http://theprogrammersparadox.blogspot.com/2008/03/science-of-information.html

Paul.

Computer science would be the science of algorithms, and the science of mathematically modeling the machines that those algorithms can run on. It is also sometimes described as the science of information processing. It doesn’t necessarily give us processes for putting together systems, or tell us what designs would be better for us as programmers. It’s mostly focused on the machine, not on the people using or instructing the machine.

You’re right that a “version of software engineering” would be much closer to what I’m talking about. I went over that more in the blog after this one.

-Max

I see the sciences as determining the constraints of an environment. What you do within those constraints isn’t science, that’s engineering (or art, or mucking around in the basement). Maybe what you’re looking for is a version of software engineering that isn’t so tightly bound to specific technologies?

But in your analogy of buildings, there is the science of physics. That is, we know fundamentally what will stay up and what won’t because we can look at gravity and all that. The methodology (or “technology”) is “21st century skyscraper architecture” but the science is physics. That’s the theoretical base that describes laws.

There’s always an art to the application of a science, but I think there still could be a science behind any art. That doesn’t make it any less the creation of the individual, though.

-Max