I may be the Pnictogen Wing's "science officer" so to speak but I am really just an amateur, and an experimentalist at heart. "Science" to me is hands-on laboratory or field work, and I am somewhat weak on theory. I have had some graduate-level coursework in such subjects as quantum chemistry and fluid dynamics and other heavy-duty topics, but I am not great with that stuff. I am badly in need of refresher courses.
Hence I feel like I am on uncertain ground here when I say: doesn't the 2018 redefinition of the kilogram by the General Conference on Weights and Measures... smell funny? I don't like what they did, and I can't say exactly why. I am not @qualia or anyone else good with physical fundamentals. My objection is chiefly on experimental grounds: faced with an inexplicable physical phenomenon, the international boffins of scientific definition simply gave up and redefined the kilogram in terms of a quantity which is simply asserted by fiatāthe value of Planck's constant h, which is now merely defined by international law. It's not an experimentally determined value any more. h is fixed by arbitrary human decision.
Doesn't that seem strange to anyone else? The second, the fundamental SI unit of time, is now, so far as I know, the only SI unit that's defined in terms of experimental observation, in terms of the frequency of photons emitted by the "hyperfine" transition between the two ground states of the valence electron of cesium-133, whose energy level is split by the nuclear spin of the 133Cs atom. Everything else is dependent on some constant that's been fixed by an arbitrary decision of an international standards body. h (Planck's constant), c (the speed of light), e (the elementary charge), k (Boltzmann's constant), and A (Avogadro's number) are all simply...defined. They're not based in experiment; they're defined solely by human authority.
I don't like this. I don't like this at all, especially because of why this happened, specifically to the kilogram.
The kilo used to be defined in terms of a physical object, a standard kilogram fashioned out of iridium-platinum alloy in 1889. The "International Prototype of the Kilogram was (still is, I guess) kept in Paris, and a number of standard duplicates made and kept in different placesāyou can see the whole list in that Wikipedia article. Elaborate cleaning and weighing procedures were defined when comparing these standard kilograms against each other and for standardizing other masses. Iridium-platinum alloy is one of the most chemically resistant metals known, scarcely subject to surface oxidation, and therefore it was thought these Ir-Pt cylinders would make excellent mass standards.
Until...they didn't. A very strange thing happened, which you can see in the chart I've excerpted from the Wikipedia article: the masses of the standard replicas began diverging from each other by many micrograms over time. And nobody could explain why.
If you search around you'll find papers about this matter, conjecturing all sorts of things. Was the Ir-Pt alloy actually subject to more oxidation than once thought? Were they acquiring surface contaminants? There were many hypotheses and tests, all of which came to nothing. The international standards boffins simply gave up, blamed the standard itself, and then fudged up their Planck's-constant definition to replace it, which was basically equivalent to saying: OK we'll just define the kilogram in terms of an arbitrary number. Yes, there's an extremely elaborate device called a "watt balance" or Kibble balance that produces a force proportional to the product of a known voltage and current (hence "watt balance"), and this is now used to standardize masses. But because all the quantities involved in deriving mass from this information are arbitrarily defined anyway (except for the second) I'm not sure how this is an adequate replacement for the standard kilogram. It seems, to my naĆÆve horse sense, like a cheat.
There was a competing project to replace the Ir-Pt standard kilogram with another physical object, a perfect sphere of single-crystal silicon. One could then calculate exactly how many atoms of silicon ought to be present in a sphere of a given size, correct for the superficial oxidation of the silicon, and obtain a mass standard that way. I liked this idea a lot; it was in keeping with the nature of the original standard, and suited my temperament as an experimental scientist. But the international boffins didn't like it, I guess. In my opinion they were wowed by the idea of having everything nice and "theoretical" (i.e. defined solely in terms of arbitrary constants, and cut loose from experimental verification.)
The experimentalist in me is hollering in protest at this whole business. An experimentally verified phenomenon was taking placeāthe divergence of the standard kilograms from various placesāand nobody could explain it, so what did the scientific authorities do? NOTHING! They shrugged it off, as if to say, "lol I guess that's what happens when you rely on physical objects, you know how wacky and unpredictable they are!" I tell you, it's bullshit. Something really stinks about this whole business. It's like they wanted to sweep an awkward physical finding under the rug because it got in the way of an image they wanted to projectāan illusion, not a reality, that the international scientific authorities had everything sewn up and neatly defined and fixed.
Yeah, it's fixed all right.
~Alyx Woodward
I'd really appreciate it, @qualia, if you could look over what Alyx has just written here, and maybe correct her misconceptions (if any). ~Chara
To further add onto this - hi, I'm not a phyusicist but I am a marine biologist and thus familiar with science work.
Literally...all the physical constants for SI units are proven physically. You can't fuck with c, because c is just the speed of light in a vacuum. Which is observed. Like Cassette said, the Planck constant is also experimentally tested.
The old kilo was thrown out because...objects deteriorate over time, even under perfectly ideal conditions. While the exact reason why isn't known, the reason it was thrown out was because if it changes...then it won't work. The constants don't change, and aren't "arbitrarily decided".
Trust me, nothing done in science - even seemingly "arbitrary" numbers - is done without some observation behind it, and you can't just say that something's a random number without some evidence. Every constant is observed and calculated via observations. Saying otherwise means you have a fundamentally incorrect view on what constants are.
Though yeah if you can figure out how to change c go right ahead, because that'd rule to be able to arbitrarily decide fundamental laws of the universe like that so we can go ever faster.

