I agree this is just a unique rationalist use. Same with âagenticâ though that has possibly crossed over into the more mainstream, at least in tech-y discourse.
However I think this is often fine, especially because âepistemicsâ sounds better than âepistemic practicesâ and means something distinct from âepistemologyâ (the study of knowledge).
Always good to be aware you are using jargon though!
Thereâs no accounting for taste, but âepistemicsâ sounds worse to my ear than âepistemic practicesâ because the clunky jargoniness of âepistemicsâ is just so evident. Itâs as if people said âdemocraticsâ instead of âdemocracyâ, or âbiologicsâ instead of âbiologyâ.
I also donât know for sure what âepistemicsâ means. Iâm just inferring that from its use and assuming it means âepistemic practicesâ, or something close to that.
âEpistemologyâ is unfortunately a bit ambiguous and primarily connotes the subfield of philosophy rather than anything you do in practice, but I think it would also be an acceptable and standard use to talk about âepistemologyâ as what one does in practice, e.g., âscientific epistemologyâ or âEA epistemologyâ. Itâs a bit similar to âethicsâ in this regard, which is both an abstract field of study and something one does in practice, although the default interpretation of âepistemologyâ is the field, not the practice, and for âethicsâ itâs the reverse.
Itâs neither here nor there, but I think talking about personal âagencyâ (terminology that goes back decades, long predating the rationalist community) is far more elegant than talking about a person being âagenticâ. (For AI agents, it doesnât matter.)
I find âepistemicsâ neat because it is shorter than âapplied epistemologyâ and reminds me of âathleticsâ and the resulting (implied) focus on being more focused on practice. I donât think anyone ever explained what âepistemicsâ refers to, and I thought it was pretty self-explanatory from the similarity to âathleticsâ.
I also disagree about the general notion that jargon specific to a community is necessarily bad, especially if that jargon has fewer syllables. Most subcultures, engineering disciplines, sciences invent words or abbreviations for more efficient communication, and while some of that may be due to trying to gatekeep, itâs so universal that Iâd be surprised if it doesnât carry value. There can be better and worse coinages of new terms, and three/âfour/âfive-letter abbreviations such as âTAIâ or âPASTAâ or âFLOPâ or âASARAâ are worse than words like âepistemicsâ or âagenticâ.
I guess ethics makes the distinction between normative ethics and applied ethics. My understanding is that epistemology is not about practical techniques, and that one can make a distinction here (just like the distinction between âmethodologyâ and âmethodsâ).
I tried to figure out if thereâs a pair of suffixes that try to express the difference between the theoretic study of some field and the applied version, Claude suggests â-ologyâ/ââ-urgyâ (as in metallurgy, dramaturgy) and â-ologyâ/ââ-iatryâ (as in psychology/âpsychiatry), but notes no general such pattern exists.
Applied ethics is still ethical theory, itâs just that applied ethics is about specific ethical topics, e.g. vegetarianism, whereas normative ethics is about systems of ethics, e.g. utilitarianism. If you wanted to distinguish theory from practice and be absolutely clear, youâd have to say something like ethical practices.
I prefer to say epistemic practices rather than epistemics (which I dislike) or epistemology (which I like, but is more ambiguous).
I donât think the analogy between epistemics and athletics is obvious, and I would be surprised if even 1% of the people who have ever used the term epistemics have made that connection before.
I am very wary of terms that are never defined or explained. It is easy for people to assume they know what they mean, that thereâs a shared meaning everyone agrees on. I really donât know what epistemics means and Iâm only assuming it means epistemic practices.
I fear that thereâs a realistic chance if I started to ask different people to define epistemics, we would quickly uncover that different people have different and incompatible definitions. For example, some people might think of it as epistemic practices and some people might think of it as epistemological theory.
I am more anti-jargon and anti-acronyms than a lot of people. Really common acronyms, like AI or LGBT, or acronyms where the acronym is far better known than the spelled-out version, like NASA or DVD, are, of course, absolutely fine. PASTA and ASARA are egregious.
Iâm such an anti-acronym fanatic I even spell out artificial general intelligence (AGI) and large language model (LLM) whenever I use them for the first time in a post.
My biggest problem with jargon is that nobody knows what it means. The in-group who is supposed to know what it means also doesnât know what it means. They think they do, but theyâre just fooling themselves. Ask them probing questions, and theyâll start to disagree and fight about the definition. This isnât always true, but itâs true often enough to make me suspicious of jargon.
Jargon can be useful, but it should be defined, and you should give examples of it. If a common word or phrase exists that is equally good or better, then you should use that instead. For example, James Herbert recently made the brilliant comment that instead of âtruthseekingâ â an inscrutable term that, for all I know, would turn out to have no definite meaning if I took the effort to try to get multiple people to try to define it â an older term used on effectivealtruism.org was âa scientific mindsetâ, which is nearly self-explanatory. Science is a well-known and well-defined concept. Truthseeking â whatever that means â is not.
This isnât just true for a subculture like the effective altruist community, itâs also true for a field like academic philosophy (maybe philosophy is unique in this regard among academic fields). You wouldnât believe the number of times people disagree about the basic meaning of terms. (For example, do sentience and consciousness mean the same thing, or two different things? What about autonomy and freedom?) This has made me so suspicious that shared jargon actually isnât understood in the same way by the people who are using it.
Just avoiding jargon isnât the whole trick (for one, itâs often impossible or undesirable), itâs got to be a multi-pronged approach.
Youâve really got to give examples of things. Examples are probably more important than definitions. Think about when youâre trying to learn a card game, a board game, or a parlour game (like charades). The instructions can be very precise and accurate, but reading the instructions out loud often makes half the table go googly-eyed and start shaking their heads. If the instructions contain even one example, or if you can watch one round of play, thatâs so much more useful than a precise âdefinitionâ of the game. Examples, examples, examples.
Also, just say it simpler. Speak plainly. Instead of ASANA, why not AI doing AI? Instead of PASTA, why not AI scientists and engineers? Itâs so much cleaner, and simpler, and to the point.
I agree this is just a unique rationalist use. Same with âagenticâ though that has possibly crossed over into the more mainstream, at least in tech-y discourse.
However I think this is often fine, especially because âepistemicsâ sounds better than âepistemic practicesâ and means something distinct from âepistemologyâ (the study of knowledge).
Always good to be aware you are using jargon though!
Thereâs no accounting for taste, but âepistemicsâ sounds worse to my ear than âepistemic practicesâ because the clunky jargoniness of âepistemicsâ is just so evident. Itâs as if people said âdemocraticsâ instead of âdemocracyâ, or âbiologicsâ instead of âbiologyâ.
I also donât know for sure what âepistemicsâ means. Iâm just inferring that from its use and assuming it means âepistemic practicesâ, or something close to that.
âEpistemologyâ is unfortunately a bit ambiguous and primarily connotes the subfield of philosophy rather than anything you do in practice, but I think it would also be an acceptable and standard use to talk about âepistemologyâ as what one does in practice, e.g., âscientific epistemologyâ or âEA epistemologyâ. Itâs a bit similar to âethicsâ in this regard, which is both an abstract field of study and something one does in practice, although the default interpretation of âepistemologyâ is the field, not the practice, and for âethicsâ itâs the reverse.
Itâs neither here nor there, but I think talking about personal âagencyâ (terminology that goes back decades, long predating the rationalist community) is far more elegant than talking about a person being âagenticâ. (For AI agents, it doesnât matter.)
I find âepistemicsâ neat because it is shorter than âapplied epistemologyâ and reminds me of âathleticsâ and the resulting (implied) focus on being more focused on practice. I donât think anyone ever explained what âepistemicsâ refers to, and I thought it was pretty self-explanatory from the similarity to âathleticsâ.
I also disagree about the general notion that jargon specific to a community is necessarily bad, especially if that jargon has fewer syllables. Most subcultures, engineering disciplines, sciences invent words or abbreviations for more efficient communication, and while some of that may be due to trying to gatekeep, itâs so universal that Iâd be surprised if it doesnât carry value. There can be better and worse coinages of new terms, and three/âfour/âfive-letter abbreviations such as âTAIâ or âPASTAâ or âFLOPâ or âASARAâ are worse than words like âepistemicsâ or âagenticâ.
I guess ethics makes the distinction between normative ethics and applied ethics. My understanding is that epistemology is not about practical techniques, and that one can make a distinction here (just like the distinction between âmethodologyâ and âmethodsâ).
I tried to figure out if thereâs a pair of suffixes that try to express the difference between the theoretic study of some field and the applied version, Claude suggests â-ologyâ/ââ-urgyâ (as in metallurgy, dramaturgy) and â-ologyâ/ââ-iatryâ (as in psychology/âpsychiatry), but notes no general such pattern exists.
Applied ethics is still ethical theory, itâs just that applied ethics is about specific ethical topics, e.g. vegetarianism, whereas normative ethics is about systems of ethics, e.g. utilitarianism. If you wanted to distinguish theory from practice and be absolutely clear, youâd have to say something like ethical practices.
I prefer to say epistemic practices rather than epistemics (which I dislike) or epistemology (which I like, but is more ambiguous).
I donât think the analogy between epistemics and athletics is obvious, and I would be surprised if even 1% of the people who have ever used the term epistemics have made that connection before.
I am very wary of terms that are never defined or explained. It is easy for people to assume they know what they mean, that thereâs a shared meaning everyone agrees on. I really donât know what epistemics means and Iâm only assuming it means epistemic practices.
I fear that thereâs a realistic chance if I started to ask different people to define epistemics, we would quickly uncover that different people have different and incompatible definitions. For example, some people might think of it as epistemic practices and some people might think of it as epistemological theory.
I am more anti-jargon and anti-acronyms than a lot of people. Really common acronyms, like AI or LGBT, or acronyms where the acronym is far better known than the spelled-out version, like NASA or DVD, are, of course, absolutely fine. PASTA and ASARA are egregious.
Iâm such an anti-acronym fanatic I even spell out artificial general intelligence (AGI) and large language model (LLM) whenever I use them for the first time in a post.
My biggest problem with jargon is that nobody knows what it means. The in-group who is supposed to know what it means also doesnât know what it means. They think they do, but theyâre just fooling themselves. Ask them probing questions, and theyâll start to disagree and fight about the definition. This isnât always true, but itâs true often enough to make me suspicious of jargon.
Jargon can be useful, but it should be defined, and you should give examples of it. If a common word or phrase exists that is equally good or better, then you should use that instead. For example, James Herbert recently made the brilliant comment that instead of âtruthseekingâ â an inscrutable term that, for all I know, would turn out to have no definite meaning if I took the effort to try to get multiple people to try to define it â an older term used on effectivealtruism.org was âa scientific mindsetâ, which is nearly self-explanatory. Science is a well-known and well-defined concept. Truthseeking â whatever that means â is not.
This isnât just true for a subculture like the effective altruist community, itâs also true for a field like academic philosophy (maybe philosophy is unique in this regard among academic fields). You wouldnât believe the number of times people disagree about the basic meaning of terms. (For example, do sentience and consciousness mean the same thing, or two different things? What about autonomy and freedom?) This has made me so suspicious that shared jargon actually isnât understood in the same way by the people who are using it.
Just avoiding jargon isnât the whole trick (for one, itâs often impossible or undesirable), itâs got to be a multi-pronged approach.
Youâve really got to give examples of things. Examples are probably more important than definitions. Think about when youâre trying to learn a card game, a board game, or a parlour game (like charades). The instructions can be very precise and accurate, but reading the instructions out loud often makes half the table go googly-eyed and start shaking their heads. If the instructions contain even one example, or if you can watch one round of play, thatâs so much more useful than a precise âdefinitionâ of the game. Examples, examples, examples.
Also, just say it simpler. Speak plainly. Instead of ASANA, why not AI doing AI? Instead of PASTA, why not AI scientists and engineers? Itâs so much cleaner, and simpler, and to the point.