There's a conceptual/cultural phenomenon I want to record here: people sometimes waste time arguing because of different implementations of concepts, right? While thinking they're sharing an implementation. And some implementations are better than others by actually, well, implementing different sayable stuff. And here's a common example I want to highlight: (in/ex)clusion-based implementations versus quality-based implementations.
Let's take "science". "Remember that science can be wrong! Phrenology and race science and astrology were science, after all!"
Someone who agrees with this would likely acknowledge that even though these were science by cultural definitions of the time (and by bearing the sign 'science'), they weren't good science. They didn't have all of the properties required of good science, and indeed, not all current science is good in that sense either. This is a quality-based implementation of 'science'. There's good science and bad science and better and worse science, but it's all science, and all of the things done by any form of science are things that science can do—such as be wrong, etc.
Someone who doesn't agree with this only because they're using a clusionary implementation might say: well, those things aren't science. They don't satisfy all the properties required of science, and indeed, some of the stuff we call 'science' currently isn't really science. 'Being wrong' is not something attributable to science, under this implementation—that would disqualify it from sciencehood.
(Sometimes people are quick to dismiss the latter as "no true scotsman"-ing. It's not! That's when you keep changing the boundaries of such an implementation in an unreasonable and inconsistent way. This isn't that, because the latter person has a relatively reasonable and consistent good-faith notion of what constitutes 'science'.)
Now, these two function a bit differently. They've got a different...API. Sure. Let's say that.
If you subscribe to a clusionary sense of science, you've got some restrictions. For one, it's easier to talk about something being science or not being science, and harder to conceptualize that something can partially be science—you've got to kind of use this separate equipment of 'partiality' to qualify your statements, and the underlying statements themselves are naturally less rich.
On the other, "discursive" side (where you're receiving stuff other people say), you've also got to remember that what other people call 'science' may or may not be 'science' according to you. You have to potentially re-sort things; you can't use other people's clusion judgments uncritically! You've got to re-evaluate whether something is in- or excluded from 'science' yourself, or else rely on a trusted source.
This is quite a bit of management, and expressiveness is more difficult.
On the other hand, using a quality-based implementation of sciencehood relieves these two difficulties to some extent:
(1) because an instance of 'science' is nonspecific without its qualifying quality, you don't decide or imply anything about it until its got its qualifier. Sure, it's science, but so what? Science can be good or bad. You don't know useful statements until you know about its goodness or badness.
(2) everything that people call science is still science. You don't need to call into question the sciencehood of it; you don't need to manage these categories in your internal worldview. Instead, the work is shifted to the quality of the science, which you may or may not have heard about. Until then, you don't have preconceptions (ideally).
But moreover, you can say more about an instance of science! Is it better than some other science? Does it have some flaws, but not as many?
Now, a more nuanced understanding of sciencehood definitely includes more dimensions than "how good of a science is it". But as a basic definition, this seems a lot more expressive and more convenient-to-be-correct-with than the clusionary implementation of sciencehood.
You can see these two specific implementations in many different ___hood characterizations by different people, and even within the same person. Likely within yourself! I know that my brain supports both—or maybe a mix—of these two implementations. These two types of implementations are both simplified models to get some handle on certain patterns in conceptual practice—neither is absolutely reflective of how people think about things, and there are many other dissimilar ways people think about things besides. But maybe this is helpful nonetheless in recognizing conceptual-management-level differences, and in helping us cut deeper into concepts, to the levels we actually want to spend our time talking about.