eh, I disagree.
hypothesizing that atoms exist allows and encourages you to design experiments which can prove or disprove their existence.
assuming that they exist, and never experimentally verifying that is not research and it is not science.
assuming that some feature of a product that you do not understand exists for any reason is at most an hypothesis, and dismissing the hypothesis without experimental evidence disproving it is simply bad science.
Roman concrete is not something far away that we can not interact with. we can verify by experiment. for years we chose not to validate the assumption that the white specs were accidental, and that delayed our understanding of Roman concrete.
I just want scientists and historians that take the time to recognize their assumptions and to prove them correct or incorrect, if possible, instead of just relying on them as immutable facts; assumptions are often wrong.
I write software for a living, and the number of times that my assumptions about the cause of any particular problem have been correct approximately 0% of the time. whether it's a logic bug or a performance problem or a usability problem. despite decades of experience in this field, I am still almost always wrong about the cause of any particular problem I encounter, unless I have seen that exact same issue previously, which is rare.
historians, some of them at least, would probably benefit quite a bit from making fewer assumptions and by experimentally proving the assumptions they do make. if it isn't possible to prove or disprove your assumption, then you must never forget that it is an assumption and not proven fact.
I've seen so many scientists and historians say they know things which they simply do not know. things like "we know dark matter exists." no, we don't. we know that the currently accepted theories of our observations require dark matter to exist in order for our observations to make sense. that is not proof of dark matter.
dark matter may exist! I don't know, but for a scientist to say that it exists is incorrect because we've never observed it directly, and have devised no tests which can prove that it does exist.
our theories which require it in order for our observations to make sense could be wrong. our observations may not be as accurate as we assume. we may be observing things that we aren't aware of which look like something that we are aware of. etc.
I'm saying that professing that you know something without proof is folly. just like the white specs in the concrete were assumed to be accidental impurity without that hypothesis being proven is folly. these things set us back as a scientific community that wants to understand its environment.