Gvildenu | Mintys / Tobulumas

完美

....

Deriniai yra tinkamiausi sprendimai.

In working on database systems (and for John Harland), I noticed that every solution opens up new problems in what seemed to be a well defined way. For example, if we create a key for our front door, then we have the problem: what happens if we lose the key? and so on. This means that we might be able to generate a complete structuring of all problems and their solutions. Then the point is to have optimal solutions (as there would generally be no perfect solutions). An optimal solution might be one where we typically spend half of our energy on the general solution and half on the exceptions.
Consider also my thoughts on databases, finding the optimum between overengineering and underengineering. I have this set of ideas that every solution opens up new problems. So the optimal solution is when we neither overengineer nor underengineer, but spend roughly half our energy on the general cases and half our energy on the exceptions.
Optimality constraints are key to the languages. We minimalize the general constructions - do not have more spaces than needed. We maximalize the specific constructions - have the conversation carry as much transferable information as possible.