What raises red flags to me is that this website seems much more oriented at convincing layfolk, not other scientists. There’s nothing about how it actually helps you do science besides give you a unified theory of everything. They also chase pop-philosophy things like Newcomb’s Paradox.

(As a comparison, something like Lagrangian mechanics can easily be shown to be useful: just provide a problem that Newtonian mechanics makes hard, and then show how easy it is in a Lagrangian formulation.)

What I noticed was that Deutsch has no big name collaborators. They aren’t publishing papers in physics journals (except for the quantum gravity paper that doesn’t mention Constructor Theory by name), and they are funded by organizations like Templeton (which “funds interdisciplinary research on what it means to be human”). They probably need this kind of web site to show to the organizations that are funding them. The work seems very speculative and preliminary. Deutsch has a track record as a deep and original thinker who pioneered the field of quantum computing, so it’s not impossible that the project could bear fruit in the future.

Weird. Interesting, but weird. Thanks for posting. Don’t really “get it”, though, and unsure whether it’s worth further time investment.

I watched a couple of the introductory videos [1,2] they have on the page. From the first one I somehow got a cybernetics-vibe. In both they keep giving examples from quantum physics, biology to ovens and robotics. But then they always refer that this stuff somehow provides a foundation for fundamental physics. The paper, which apparently kicked it off [3], seems to be (based on a very brief browse) about marrying physics and computation together somehow. Is this trying to be a new foundation for physics, or is this a general theory of everything, or the next iteration of physicists implying that all other disciplines are just physics? Can anyone give a brief explanation for an engineer?

My sense is that this is proposed as a new foundation for physics (similar to the situation in mathematics where category theory and homotopy type theory are proposed as alternatives to set theory as a foundation for mathematics). So you would use constructor theory as a framework for defining physical theories. It is suggested that constructor theory is general enough to be used as a framework for theories outside of just physics.

The deep connexion with computation and information theory is interesting. So it ought to have applications in computer science. Maybe it will turn out to be an alternative to category theory: both constructor theory and category theory are built on top of algebra.

The difference is that category is a well established field, and using it as a foundation has gotten us interesting mathematical insights in general. Neither seems to be true here.

Thank you for the clarification. That makes some sense. I feel there’s the typical danger here of proposing a very general model of everything, which in the end doesn’t bring anything new to the table. But I guess if it’s a lens through which you can describe and try to understand some general underlying patterns then it will have served its purpose.

Just throwing some ideas out there, the following is likely complete nonsense:

On the perpetually growing reading list I have Whitehead’s Process and Reality. I wonder if some parallels could be made between the focus here on transformations, or constructions, and Whiteheads’ views of entities as processes.

I wonder whether there’s anything formal here that isn’t just category theory. We’ve been studying compositions of transformations for decades, after all.

What raises red flags to me is that this website seems much more oriented at convincing layfolk, not other scientists. There’s nothing about how it actually helps you do science besides give you a unified theory of everything. They also chase pop-philosophy things like Newcomb’s Paradox.

(As a comparison, something like Lagrangian mechanics can easily be shown to be useful: just provide a problem that Newtonian mechanics makes hard, and then show how easy it is in a Lagrangian formulation.)

What I noticed was that Deutsch has no big name collaborators. They aren’t publishing papers in physics journals (except for the quantum gravity paper that doesn’t mention Constructor Theory by name), and they are funded by organizations like Templeton (which “funds interdisciplinary research on what it means to be human”). They probably need this kind of web site to show to the organizations that are funding them. The work seems very speculative and preliminary. Deutsch has a track record as a deep and original thinker who pioneered the field of quantum computing, so it’s not impossible that the project could bear fruit in the future.

Weird. Interesting, but weird. Thanks for posting. Don’t really “get it”, though, and unsure whether it’s worth further time investment.

I watched a couple of the introductory videos [1,2] they have on the page. From the first one I somehow got a cybernetics-vibe. In both they keep giving examples from quantum physics, biology to ovens and robotics. But then they always refer that this stuff somehow provides a foundation for fundamental physics. The paper, which apparently kicked it off [3], seems to be (based on a very brief browse) about marrying physics and computation together somehow. Is this trying to be a new foundation for physics, or is this a general theory of everything, or the next iteration of physicists implying that all other disciplines are just physics? Can anyone give a brief explanation for an engineer?

[1] https://www.youtube.com/watch?v=8DH2xwIYuT0

[2] https://www.youtube.com/watch?v=0zeT2npYf18

[3] http://constructortheory.org/portfolio/the-philosophy-of-constructor-theory/

My sense is that this is proposed as a new foundation for physics (similar to the situation in mathematics where category theory and homotopy type theory are proposed as alternatives to set theory as a foundation for mathematics). So you would use constructor theory as a framework for defining physical theories. It is suggested that constructor theory is general enough to be used as a framework for theories outside of just physics.

The deep connexion with computation and information theory is interesting. So it ought to have applications in computer science. Maybe it will turn out to be an alternative to category theory: both constructor theory and category theory are built on top of algebra.

The difference is that category is a well established field, and using it as a foundation has gotten us interesting mathematical insights in general. Neither seems to be true here.

Thank you for the clarification. That makes some sense. I feel there’s the typical danger here of proposing a very general model of everything, which in the end doesn’t bring anything new to the table. But I guess if it’s a lens through which you can describe and try to understand some general underlying patterns then it will have served its purpose.

Just throwing some ideas out there, the following is likely complete nonsense:

On the perpetually growing reading list I have Whitehead’s Process and Reality. I wonder if some parallels could be made between the focus here on transformations, or constructions, and Whiteheads’ views of entities as processes.

Another interesting perspective might be to hear how this theory would explain “self-organisation due to entropy” as in this article: https://www.quantamagazine.org/digital-alchemist-sharon-glotzer-seeks-rules-of-emergence-20170308/ . Is entropy a constructor?

I wonder whether there’s anything formal here that isn’t just category theory. We’ve been studying compositions of transformations for decades, after all.