> Cold didn't refer to the temperature at the point of fusion, but rather the ambient temperature.

Perhaps that was true then, but nowadays it seems like a rather ill-defined and not particularly useful distinction. Sure, the ambient temperature might be relatively cold, but why should that matter if the fusion itself is still taking place at high temperatures/energies? Should fusion in a tokamak be considered "cold fusion" because the apparatus itself is sitting in a room with a relatively cold ambient temperature? What about if someone develops a significantly miniaturized tokamak; say, a hand-sized one? Should fusion in that case be considered "cold"? This can be continued to smaller and smaller scales; in other words, at what point/scale does "high-temperature/energy fusion surrounded by low-temperature matter" move from "hot" fusion to "cold"?

I think it'd be fair to say that "cold fusion" (heavily) implies new physics these days; otherwise, it's "just" a miniaturized fusion reactor.

> Sure, the ambient temperature might be relatively cold, but why should that matter if the fusion itself is still taking place at high temperatures/energies? Should fusion in a tokamak be considered "cold fusion" because the apparatus itself is sitting in a room with a relatively cold ambient temperature?

A tokamak isn't "cold" as there's a clear and deliberate containment mechanism and a physical boundary delineating a macro environment of extreme energies within which fusion occurs. By contrast, circa 1989 the NASA setup might have been characterized as cold. But I don't think there's any answer that will satisfy people who disagree with me because the question rather quickly devolves to something like a sorites paradox (i.e. paradox of the heap of sand), and there's no chance mainstream science would ever attempt to qualify and quantify what counts as "cold" because the phrase "cold fusion" is already taboo, as is anything that even alludes to the phrase, making the exercise pointless.

But previous comments seemed to make the assumption that "cold" meant a complete absence of extreme temperatures and thus no plausible mechanism for achieving kinetic energies sufficient for consistent fusion, presumably because they assumed that Fleischmann and Pon's claim was similar to a stereotypical free-energy device.

> I think it'd be fair to say that "cold fusion" (heavily) implies new physics these days; otherwise, it's "just" a miniaturized fusion reactor.

I agree, today cold fusion implies new physics. And the way we got here wasn't because Fleischmann and Pons claimed to invent a free-energy device, or a way to trigger spontaneous fusion without any appreciable input of energy. They were inputting energy, and doing so to a device that, if we squint really hard and make allowance for then contemporary knowledge, wasn't on its face qualitatively different from the present NASA claim. On its face nothing in their claim violated known physics, though that's because their claim was so thin.

The whole incident came to a head the way it did because people got super excited, many teams tried to replicate, and at least one did replicate--at least, they replicated false signals. The claims triggered a kind of geek hysteria--people started seeings things that weren't there and then rationalizing them, because that's what happens when a ton of people in a frenzy jump into the fray simultaneously. See, e.g., COVID-19 research.

How does a community prevent that from happening again? Notably, Fleischmann and Pon's claim didn't have a theory for what, precisely, was happening. They were just exploring paths which at that time were perfectly legitimate (and might still be; IANAP), found a signal, and published. You couldn't critically judge the credibility of the result without first investing yourself in it by attempting replication. There was nothing of substance to filter beyond their credentials. The fix was to require claims to come with a theoretical model that could be judged on its own merits.[1] The absence of a model doesn't necessarily mean a claimed result can't possibly be legitimate; it's just a time and risk management strategy used to front-end the community process. Before the debacle an attached model was a nice-to-have; afterward it was basically a categorical requirement, reflecting the reality that, a priori, absence of evidence is evidence of absence. (And absence of evidence is even stronger evidence of absence after you institute a rule requiring new claims to include a model.)

So that's why I say that today "it's not cold fusion" implies that a claimed result comes packaged with a plausible model, where in turn plausibility is partly a function of the need for new, fundamental physics. If a claim doesn't come packaged with a plausible model (or models, as in the NASA clam) it's analogized to "cold fusion" because that debacle also lacked a plausible model. (In actuality, it lacked any kind of model, but that nuance is lost because this isn't the kind of context where normal people apply predicate logic. But the nuance matters when people start to argue that cold fusion == bad science because Fleischmann and Pon's claim contradicted known physics. It didn't and it couldn't, because they didn't make any claim about mechanism beyond the necessary implication that it somehow involved, in at least one manifestation, palladium, deuterium, and an applied electrical charge.)

[1] Scientists still use credentials to filter claims. The amount of time someone is willing to invest critically analyzing a paper is still a function of the author's perceived credibility.

> A tokamak isn't "cold" as there's a clear and deliberate containment mechanism and a physical boundary delineating a macro environment of extreme energies within which fusion occurs.

Sure, but as you noted this definition becomes fuzzier at smaller and smaller scales, which IMHO reduces its usefulness quite a bit as it doesn't do a very good job at differentiating between hot and cold fusion beyond "how small of an apparatus can we build with current technology".

> and there's no chance mainstream science would ever attempt to qualify and quantify what counts as "cold" because the phrase "cold fusion" is already taboo

At least to me, it seems that there are at least two potential definitions of "cold fusion" already floating about (though to be fair, I don't know whether these are "mainstream" definitions):

1. Fusion which doesn't require high-energy/temperature nuclei

2. Fusion (maybe "regular" fusion, maybe not) which takes place at small-enough scales that large containment apparatuses are not necessary

The first definition might work for a (relatively) precise definition of cold fusion, while coming up with a similarly precise one for the second definition looks like it'd be harder.

The physicists in the article appear to be using the first definition, although I don't know whether that is the most widespread definition used among "mainstream" physicists. Fleischmann and Pon's setup seems to fall under the second category.

Granted, it looks like you're much more familiar with the subject matter than I am, so there's a good chance I messed up something or another.

> So that's why I say that today "it's not cold fusion" implies that a claimed result comes packaged with a plausible model, where in turn plausibility is partly a function of the need for new, fundamental physics.

That was an interesting dive into the historical aspect of things. I really appreciate you taking the time to type out all that explanation!