Science is intended more to adapt an actual "theory" over time to better suit the evidence that it is presented with until it increasingly encompasses all edge cases that relate to the topic in question. That "adaption" can be considered disproving with an immediate re-creation of an alternate theory moments later to encompass the changing circumstances.
Or, well... it "could be considered" exactly what it is: retaining an existing theory and all of its basic assumptions, while tacking on modifications or qualifiers to make it better fit the data. That's not "disproving" anything. That's improving an existing theory, and that's what the vast majority of everyday science is about. Most active scientists are working within existing paradigms and working out the details of theories by starting with all the assumed knowledge of their fields.
No actual scientist is walking around questioning every scientific "fact" on a daily basis. "Oh, you know what, I really don't believe that whole 'water is made up of hydrogen and oxygen' thing. That whole 'atomic elements build compounds' thing sounds potentially bogus and 'unproven,' so let's have my lab spend the next six months retesting that and finding out what water really is made of!"
No sane scientist thinks like that, and scientific progress would be practically impossible if we went out actually worrying about potentially falsifying everything.
Instead -- we accept much of scientific knowledge as "proven" (not in a formal logical/mathematical sense, but the normal everyday sense of "well-tested"), and we go on with our lives filling out the "edge cases" as you put it. Only if some major discrepancies arise repeatedly do we begin to wonder whether underlying assumptions may be at fault... and even then it would take a heck of a lot to overturn the idea that water is composed of oxygen and hydrogen, for example.
On the flip side, actually "proving" something is exceptionally hard work. It is saying that at no point, ever, under any circumstances in this or any conceivable universe, with any natural or unnatural influence could this situation *EVER* take place for *ANY* reason. These are the rules, these are how things behave, and this is how things will always, and forever behave; EXACTLY like this and there's not a damn thing that anyone including the hand of God himself could do to change that.
That is a completely and utterly BS definition of "proof" that no one EVER uses except in discussions like this. Seriously. It's not what the math or formal logic people mean by "proof," because actual math and formal logic people generally recognize that their claims are not directly relatable to the real world, let alone "any conceivable universe." Math and formal logic are abstract symbolic systems. While they may at times be very good models for talking about the real world, they do NOT have any exact correspondence with the real world.
Find me an exact "triangle," for example, in the real world. Not something that looks vaguely like a triangle -- something that fits the mathematical definition of one, with three exact points, precisely straight line segments, etc. Even if you came close, to create something that on a microscopic level still seemed to be a triangle, on the individual atomic level there would be irregularities -- and even if that were somehow "perfect," we could keep moving down until we got to the "quantum foam" level... a real EXACT triangle in the mathematical sense doesn't exist in the real world.
Does that mean all of Euclid and geometry and all the formal "proofs" are wrong or irrelevant? Of course not. But they are working with certain kinds of abstract assumptions that have no exact correspondence with the real world. They are a MODEL. So, when you try to apply that standard of "proof" to "this or any conceivable universe," you're doing something completely illogical. No one ever actually does that -- and if they claim they are doing that, they are being disingenuous. The logic of abstract models is simply that -- the logic of abstract models. Standards of proof there are simply not applicable to the real world.
Saying that something is "proved" means that there is nothing more that could ever be known about that topic, and that nothing could ever impact that field, be it further advances anywhere else, supernatural influence, extra dimensional characteristics, weird things that we haven't even considered possible
Sorry, but that's just NOT TRUE. In normal, everyday English, according to Merriam-Webster, "to prove" is "to test the truth, validity, or genuineness of" or "to test the worth or quality of." Note the emphasis on the word "to TEST." In normal language (i.e., >99% of the time "prove" is actually used in English), it means to TEST something to determine how good it is. If something passes a test, it has met some sort of standard -- that doesn't necessarily mean it can never be questioned again, or we should stop thinking about the subject forever, or that it's true in "any conceivable universe." Something that is "proved" has passed some standard test for quality, or, as another definition in Merriam-Webster puts it: "to establish the existence, truth, or validity of" or "to demonstrate as having a particular quality or worth." The normal way of doing these things in the real world (as opposed to some sort of abstract formal logic exercise) is through empirical testing.
Gravity is one such theory. We know that it exists, we know how it works, we know how to calculate it, we know how to utilize it's traits for all kinds of things. But "proving" that water goes downhill ... It's something that we take for granted and require to base civilization as a whole on, through irrigation and plumbing. Something doesn't need to be "proved" to be immeasurably useful in the daily lives of incalculable people over countless generations.
Once again -- this is NOT what the word means in real life. Say Person A claims, "I claim water will flow uphill and gravity doesn't exist." Person B says: "Prove it." Person A: "Uh... I can't. But can you PROVE that it flows downhill?" Person B: "Sure, watch. [Pours water on a slope -- it flows downhill.] See!" Person A: "Yeah, I guess you're right."
That's what "proof" actually means in English when referring to the real world, including empirical science. You subject a claim to a test, and you take the results of the test as some sort of "proof." Depending on the quality of the test, you may draw various conclusions about the quality of "proof" obtained -- in cases where the truth is really important, you may insist on more tests or more rigorous ones, or, as we say in English, "a higher standard of proof." See -- proof has degrees depending on the kind of tests you use and the standards that are upheld by those tests. It's why in a courtroom we can have a standard that says "proof beyond a reasonable doubt." If "proof" only meant what you think it means, then such a definition would be redundant and misleading -- if "proof" means something is true in "any conceivable universe" for all time, then what would it possibly mean to be "proving beyond a reasonable doubt"? What would that qualifier add? Well, obviously it sets the standard for the test -- a "reasonable person" [legal term] would have no doubt of the guilt in that situation.
You may think that this is getting pedantic, and it is, but at the same point, it is the difference in Science between "Proving" a theory and not.
There's a difference between being "pedantic" and being WRONG. Let's subject your hypothesis about the use of the word "proof" to some empirical testing, shall we?
You claim that actual expert empirical scientists don't use the word "proof" in this loose way -- that they always mean this very abstract version of formal logic "proof" where something would be true in "any conceivable universe."
Well, let's look at the abstract of TFA, linked to in TFS. What do we see?
Although solar neutrinos from secondary processes have been observed, proving the nuclear origin of the Sunâ(TM)s energy and contributing to the discovery of neutrino oscillations, those from protonâ"proton fusion have hitherto eluded direct detection. Here we report spectral observations of pp neutrinos, demonstrating that about 99 per cent of the power of the Sun, 3.84 Ã-- 1033 ergs per second, is generated by the protonâ"proton fusion process.
Not only does this scientific abstract include the word "proving" regarding previous research, but they characterize their own research in TFA as "demonstrating" that something "IS generated" -- that's language of formal logic or math proofs (e.g., "quod erat demonstratum"). They didn't say they were "theorizing" what something "might be" or that something merely "provided some evidence" of what "could be." They said that previous scientists had evidence "proving" something, now they had "demonstrated" what something truly "is."
So, let's see, I could either believe some random guy's definition of "proof" on Slashdot that never seems to apply to the real world. OR, I could look at the actual empirical usage of the term "proof" by actual leading scientists, which has been published in one of the top journals of the entire scientific field, with language that has been peer-reviewed and checked by editors of the highest scientific caliber.
Well, my empirical evidence suggests your definition of "proof" is wrong. QED. :)