What is the main assumption of science?

There isn't a main idea of science; anyone who knows much about it would tell you that and probably give you different versions of their own ideas. In fact, the very idea of "the main idea of science" is not a very scientific idea; more metaphysical as it is usually dealt with. But such matters fill shelves of books in any suitable library and there is no room, here to deal with them properly.

Here are a few important ideas. You might want to choose a main idea from them. If you do, you probably haven't understood, either me or science.

Assuming that there is a real world about us, or alternatively if it really seems to us that there is such a world...

And that such a world behaves consistently or seems to behave consistently, that is to say, according to patterns (what we commonly call "rules" or "laws") ...

(Here we might add: "And that we are part of such a world or perceived world and behave and cause it to behave accordingly..." but some people would argue with that, though I think it is a good one.)

And that there are patterns by which the patterns of behaviour of the world interact in such ways that we can make sense of them (what we might call "Logically explaining" them.)

Then the main activity of a "scientist" would be firstly to discover such things as we can in the world or apparent world, and such rules as we can about how they behave (what we might call "facts" of one kind) and such things as we can about how they influence each other, their "logical explanations" (what we might call "facts" of another kind)

Secondly another part of the main activity would be to enable us to predict as reliably as possible, events that follow "logically" from what we have discovered so far. So far we have found no limit to such a process, and in fact the more we are in a position to state with confidence, the more new questions we generally discover.

Since most forms of our perceived laws of logic state that a statement must be true, or false, or meaningless (though that gets tricky when we deal with fuzzy or quantitative statements, not to mention para-consistent logic) and that a true statement can only imply a true statement, but that the others can imply anything, more or less, one of our most powerful tests (some people say our only test, which I deny, but never mind me) is whether the predictions we can make from our ideas of the facts and the laws, are correct or not. If not, we say the idea we were testing, our hypothesis, is false. Otherwise we say that it might be true. It has become a stronger hypothesis, meaning that it has survived yet another test.

In practice we never know for certain how many possible, testable, hypotheses there are about any fact or logic, nor, least of all whether our set of hypotheses that we are testing include any correct ones, so really our job comes down to choosing the strongest hypotheses at any given stage of history. If someone discovers a new law tomorrow, it might make much of what we see as obvious today, look ridiculous.

Another main part of science is not to stop and shout insults at each other, but to buckle down and make sense of the new strongest hypotheses and apparent laws.

And all that was pretty vague and superficial, and no one item was any good without the others, so you can guess that anything that anyone says about "the main part of science" should think a bit more carefully, unless he means something like "Ninety percent perspiration and five percent inspiration and five percent error and delusion."