[tech_ken]

I don't think paradigm shifts have to be 'better' in some march-toward-progress sense, they can be lateral or even regressive in that way and still lead to longer-horizon improvements.

I think also what's practically applicable changes constantly. Perhaps we're truly at the End of Science, but empirically we've been wrong every other time we've said that. My money is that there's more race to run.

[ainch]

On that note, Terence Tao gave a good interview to Dwarkesh Patel talking about Kepler. He pointed out that the previous geocentric models were actually more accurate than Kepler's at the time, in part because they'd had so much complexity piled on to solve minor errors. Kepler's theory was more elegant, but at the time it wasn't necessarily a better model.

I think important paradigm shifts can often look like this - there's not necessarily a reason to expect them to be instantly optimal. Deep Learning vs 'good old-fashioned AI' is another example of this dichotomy; it took a long time for deep learning to establish itself.

[Nevermark]

I like this a lot. The Innovators Dilemma for science.

The new simpler tool always competes with highly adapted complex tools to get to a region of value generation.

Starting where it’s greater simplicity, despite less complementary adaptations, is of great advantage.

Then slowly accumulates its own version of practical complements that let it excel overall.

[cogman10]

> I don't think paradigm shifts have to be 'better'

But they do. Paradigm shifts happen because the new paradigm explains the unexplained and importantly also covers the old model. If prior data is unexplained with a paradigm shift, the shift will never be adopted.

> Perhaps we're truly at the End of Science

Who said that? Just because the core of our current models seem pretty rock steady doesn't mean there's not more science. It simply means that we can mostly just expect refining rather than radical discovery.

There will be sub-paradigm shifts, but there's likely not going to be major "relativity" moments from here on out.

[tech_ken]

> Paradigm shifts happen because the new paradigm explains the unexplained and importantly also covers the old model

Empirically it seems that paradigm shifts are more driven by deaths and retirement rather than improved fit to the data. Moreover the way that you reconcile old data with the new model can be contestable; it's not like everyone all at once says "oh this new model is clearly a strict superset of the previous one, time to adopt it". With all that said I think one could argue that this stuff is basically noise and that the process still 'trends toward progress' (and I'd agree). But I would say that the scale of noise can also be quite large relative to things a human might experience in their life. I was sort of imagining social-disruption (like a dark-age type regression) as the 'backwards paradigm shift'.

> but there's likely not going to be major "relativity" moments from here on out

I cannot understand how anyone treat this as something that can be objectively concluded; by definition these kinds of radical paradigm shifts are basically unforeseeable up until they happen. I called it the "End of Science" to draw a parallel to "End of History"-type thinking because both (IMO) take this view of "there will be no more revolutions, only incremental adjustments on an unshakeable core into infinity", which I feel is personally a 'vibes based' assessment of things. It's not even that I disagree with it so much as I feel like the statement is basically (and will always be) a pure guess, one which many people have made and been wrong about in the past.

[cogman10]

> I cannot understand how anyone treat this as something that can be objectively concluded

Mostly because the room for the unexplained in physics is really small. It's possible that we end up finding some sort of big revelation about quantum physics that completely changes how we view relativity. But even in that case we are more likely to find that relativity is just a simplification of a more complex model with better explanatory power. Very much like how Newtonian physics still works really well from quite small things to anything most humans will deal with on earth. It's only when you start talking about uncommon experiences in extreme environments where relativity starts being a requirements to make the math work.

> there will be no more revolutions, only incremental adjustments on an unshakeable core into infinity

I guess I'm just more comfortable with that position. A lot of the revolutions in science circled around detecting and measure things previously immeasurable and unsee-able. The study of EMF exploded when it did because that's also when our ability to generate and measure electricity in more than just a party trick happened.

We are at a point where things are more of an unknown unknowns with no theoretical way to observe. The physics models at the fringes are mostly centered around things we can't measure.

There just aren't interactions we can't currently predict. The only one I know about is radioactive decay.

And a lot of this shows in modern society. In physics, the last major paradigm shift was relativity. That's a nearly 100 year old model at this point. Everything we have currently is just incremental improvements on the physics model.

I don't think this is because we just aren't as smart today as we once were. Quiet the opposite. There are far more people on the planet. There are almost certainly a lot more "Einsteins" trying to find a new paradigm and they've simply failed over the decades because it's seemingly increasingly unlikely that there is something to find.

[throwaway27448]

> Empirically it seems that paradigm shifts are more driven by deaths and retirement rather than improved fit to the data.

Indeed, Kuhn's own work acknowledged this.

[paganel]

> It simply means that we can mostly just expect refining rather

The practical issue is if there will enough funds for just "refining", instead of "paradigm shifts", which I understand as new and "exciting" discoveries. I'm not a scientist, of course, this is just my layman's understanding.