Though not infallible and not claiming to be, science is the only major human collective activity that is internally self-correcting, rather than corrected from without.
To this date, no documented case of scientific misconduct or fraud has ever been uncovered by mystics, alternative cosmologists, creationists, astrologers, homeopaths, psychics, nor by post-modernist philosophers, only by other scientists.
This self-corrective nature is absolutely essential for science to make any progress, as it is but a logical process for testing hypotheses against evidence, to provisionally lend credence to or to falsify them, at least for the time being, as even the best established findings of science are subject to the suspicion that they might need rejection or at least refinement at some point in the future.
Such findings, facts, and the logical constructs, the theories and their component hypotheses that explain and describe them are thus tentative, subject to update with newer and better evidence, newer and better data. Newer and better methodology.
That being said, while a given theoretical framework may not be absolutely conclusive, never beyond all possible doubt, it can if confirmed be proven beyond a rational doubt, though still subject to rejection or fine-tuning by newer data.
This is because one never knows when better observations, experiments, or reasoning will pop up to cast doubt on even the most airtight set of ideas, and require them to be updated, tossed out, or amended.
This does not sit well with those who crave absolute certainty.
That which comes to be recognized after a fair hearing as pseudoscience is not so accommodating, and whose advocates tend to retrofit facts and shoehorn observations post-hoc to conform to theories rather than the converse, and who also tend to work backwards from a predetermined conclusion which is never updated nor corrected, but held to be eternal truth.
Pseudoscience is dogmatic, unwilling to concede to the possibility of its proponents’ error, indeed, perpetuating and often compounding these errors, and it tries to do an end run around the rules of due scientific process. It wants the credibility of science without having to do the hard work or play by the rules of fair conduct to get it.
Science is open to but highly critical of new ideas.
It has to be in order to move forward, to allow us to learn new things, and at the same time weed out what does work from what doesn’t. To weed out the diamonds from the crap.
The work of any scientist, even the most well-regarded, is subject to numerous attempts by other scientists to confirm or refute her ideas, after which the idea is granted conditional assent or rejected by the scientific community at large. Nothing gets a scientist more of a reputation, more prestige, more of a place in history than to overturn an established theory, and if nothing else, at least to show another scientist wrong.
As Benjamin Franklin is said to have put it:
In going on with these experiments, how many pretty systems do we build, which we soon find ourselves obliged to destroy.
Ben, you phrased it perfectly.