I agree with
No single scale fully captures technological progress
and still like Jason Crawford's idea of a dashboard of metrics for civ progress instead of any single scale, expanded beyond consumption to include productivity and "amount of stuff we can get done", including more future-proof metrics than the near-term ones they all listed, etc. Also Ivan Vendrov's suggestion to counteract the tendency to use Goodhart-resistant-but-terrible metrics with richer ideologies.
I agree with the conclusions. Extraction and use of energy at a bigger scale is certainly part of the picture, but mastering details at small scale is also important. I have another example in mind. Before fire, for millions of years, human/hominin technology was almost only represented by wood sticks and stone choppers. All this time, progress was about mastering very subtle movements of hands to craft slightly better choppers and sticks, and about noticing subtle nuances in stone and wood to pick better raw material. It is true that the point of all this was in the end to extract more energy from the environment, but the focus was on subtle details, small scale rather than big scale.
This post is making a point that would appear to be obvious, however given how the Kardashev scale and direct energy usage comes up often in popular discussion, it seems it needs to be made. The Kardashev scale is a poor measure of civilizations progress and power.
An essential feature a scale must have to be useful is that the levels on it must be somewhat evenly spaced. The Kardashev scale does not have that, and I will stretch the point here by arguing that K3, harnessing the power of a galaxy is in fact below K2 (star) from one point of view.
A common way to achieve K2 is the Dyson Swarm. To progress from <K2 to K2, a civilization would need to both design, understand and build such a swarm. However to progress to K3, you would just need to use a Dyson swarm with some pretty minor modifications. You could have lost the ability to make it and understand it fully but still use it to progress to K3.
Is it possible to make a Dyson Swarm and not be able to make an interstellar starship? Making a starship when you have such a massive amount of energy and resources at your disposal would not be hard. Given that a swarm must be grown from a much simpler replicator that makes many copies of itself, simply sending such a replicator on a starship to the next star is all it takes to reach K3 as the process repeats exponentially.
As an analogy, consider measuring the complexity and sophistication of life based on how much of a homogeneous unclaimed environment it can colonize. For an example of bacteria in a pool, you could make K1 = 0.1% of the pool, K2=1%, K3=10% etc. However the same bacteria that could take 0.1% would obviously quickly grow to take the whole pool, just like our replicators would take the whole galaxy.
This is also known as Barrow's microdimensional mastering. While I am not sure about the details of this scale, especially if levels below Type IV minus are meaningfully possible, the general principle of mastering ever smaller sizes sure seems to hold up.
If we must have a scale, bulk control of ever smaller scales seems to be the best measure. After all, that is pretty much literally what Moore's law is. Bulk control is important, just moving a single atom around doesn't give civilizational power. Note that energy usage is not so indicative - in the last 50 years, electricity usage per capita in the developed world has barely changed if compared to the vast improvements on the Barrow scale. When you fully master the small, you get AGI/ASI.
With that, you then master the large with vast quantities of replicators without much more time or effort.
A possible objection to this scale is that nature is already type IV minus so what exactly is humanity achieving. This isn't such an objection as microchips are very different to DNA, and the same objection applies to human vs nature regarding energy usage if we are using that as a scale. For energy, if you take the sunlight hitting plants, then the direct energy usage of humanity would put us below trees. (With a fudge where you don't count sunlight hitting our crops)
How to compare humanity to nature?
As a very rough estimate lets take the number of transistors humanity produces as 10^23 per year. Now for the nature side of humanity, a person has ~30 trillion cells, with 6 billion DNA letters per cell. That gives ~10^23 DNA letters for just a single human. Does that mean that your DNA is equal to the entire transistor production in some sense? I think its clear that isn't a fair comparison, a transistor does a lot more than a DNA letter that may not even be used. If we instead compare a transistor to a cell, then we get a similar 10^23 for the number of human cells (including all humans) as transistors. So humanities natural (cells) anti-K level is about the same as its technological (transistor) one, at about the same time that its artificial intelligence passes its natural one.
No single scale fully captures technological progress. Very early on, mastery of fire and beasts of burden were important and perhaps had a more energy focused nature, however in the last 50-100 years bulk control of ever smaller scales seems to be a better scale.