[ExI] Why do the language model and the vision model align?

Wed Feb 25 16:00:38 UTC 2026

On Wed, Feb 25, 2026, 8:10 AM John Clark <johnkclark at gmail.com> wrote:

>
> On Tue, Feb 24, 2026 at 2:12 PM Jason Resch via extropy-chat <
> extropy-chat at lists.extropy.org> wrote:
>
> *> Are you familiar with the block universe view that emerges from
>> Einstein's relativity?*
>>
>
> *Of course, but the standard block universe view doesn't take quantum
> mechanics into consideration, so if you believe in Copenhagen it's just
> wrong. Many Worlds can save it but then it isn't a 4D block, it's a block
> of an astronomical number of dimensions and possibly an infinite number of
> dimensions, but that view is so complicated it is not very useful.  *
>

See this, it might help:
http://www.weidai.com/qm-interpretation.txt

Just because an idea is hard to conceptualize doesn't mean it is false. If
physicists thought that way we wouldn't have QM or GR.

But you are right that things do get harder to understand when QM is
combined with GR. This article lays out some of the difficulties:

https://en.wikipedia.org/wiki/Problem_of_time

What is interesting is that in the updated equation provided by Bryce
DeWitt and John Wheeler, time falls out of the equations of QM completely,
and the notion of time must be recovered as some kind of internal
phenomenon that observers experience.

> *> Change is possible in mathematical objects, or universes, or
>> computational functions, but change is always in respect to something.*
>>
>
> *Yes, and we call that "something" a " Physical Object".*
>

No it is with respect to time, which is only one dimension picked out from
a four dimensional object. And what relativity tells us is that different
observers in different reference frames, consider different directions
through this block (the one orthogonal to what they consider to be their
spatial movement) as *their* time dimension. But note there are infinitely
many different directions one may have as their "time dimension".

This is all standard special relativity. It is why Newton's clockwork
universe with a universal notion of time breaks totally, and is
unsalvageable. See:
https://alwaysasking.com/what-is-time/
Or the video based on that article: https://youtu.be/QC52vRmtQoU for a
quick and easy introduction.

>
>> * > Think of a plot of a graph of y=f(x) on an X-Y coordinate plane. The
>> entire graph is static, and yet, we can say that f(x) changes with respect
>> to x.*
>>
>
> *But "x" could represent everything or it might represent nothing, there
> is no way to tell. And both the symbols x and f(x) never change, they just
> sit there in a book. Neither symbol can add 2+2, and the symbol "cow"
> cannot produce milk. *
>

You deleted my point above bot p(t) could define the position of a particle
with respect to time. This was essential to your understanding.

Next consider a graph containing the plots for many many particles p1(t),
p2(t), ... , pn(t). Again this is steric, however you can imagine very
complex interactions may result as to varies in this graph, if for example,
the particles all represent the particles of your body and brain. Now the
entire graph is a static mathematical structure that does not itself
change. However with respect to t it does change. And the interactions
embodied by all the particles represent counterfactual relations and
computations that could produce a conscious mind (assuming the
computational theory of mind), this this static graph would contain an
observer who experiences a world at different times, and who would
accumulate memories and at each t would have a brain state believing t is
the present time, and that higher t values (the future) remains unknown,
while lower t values (the past) are remembered.

So you don't need the overall mathematical structure to change. You just
need a structure there embodies counterfactual relations (e.g., if two
particles never collide, their trajectories don't change, otherwise they
do) with respect to some dimension within that mathematical structure.

>
>> *> we can say the state of the memory of the Turing machine changes in
>> respect to the number of steps the Turing machine has performed.*
>>
>
> *Yes but a Turing Machine can be a real physical object and not just an
> abstraction described in a book. *
>

That both the laws of physics and the properties of Turing machines can be
described in books is a red herring. A Turing machine may also exist an an
independent structure that contains counterfactual relationships with
respect to one if its dimensions, just as our universe can be seen in that
way, and just as other possible universes may exist in that way.

> *> But to use your own objection "a law can't do anything"*
>>
>
> *They can if the law refers to things that can change, such as position
> speed and entropy.  *
>

Exactly. Now you're getting it!

> *>> Apparently Mueller is as silly as Marchal, like him the man believes
>>> that it is of profound significance that physics cannot give even a
>>> probabilistic answer to the question " if 2 perfect copies of you are made
>>> and one goes to Washington and one goes to Moscow  which city will you find
>>> yourself in?". Well of course physics can't give an answer to that because
>>> not every string of words that happens to have a question mark at the end
>>> is a question, sometimes it's just gibberish. How do I know this thought
>>> experiment is ridiculous? Because even after the experiment has been
>>> completed nobody can say what the correct answer should have been. It's
>>> amazing how good personal pronouns are at hiding nonsense, if instead of
>>> asking which city will you see Mueller and Marchal had asked which city
>>> will John Clark see then that would NOT have been nonsense, it would've had
>>> an answer, and the answer would have been "both". *
>>>
>>
>> *> Add Tegmark to your list of silly people, for he says the exact same
>> thing in Our Mathematical Universe:*
>>
>> *"It gradually hit me that this illusion of randomness business really
>> wasn’t specific to quantum mechanics at all. Suppose that some future
>> technology allows you to be cloned while you’re sleeping, and that your two
>> copies are placed in rooms numbered 0 and 1. When they wake up, they’ll
>> both feel that the room number they read is completely unpredictable and
>> random."-- Max Tegmark in “Our Mathematical Universe” (2014)*
>>
>
> *And I agree with Tegmark's above statement 100%. What I very strongly
> disagree with is the statement "it's impossible to predict what number
> "YOU" will see" is a profundity. It's a silly thing to say because in this
> context the word "you" is undefined. *
>

If you agree with Tegmark, then you agree with Marchal and you agree with
Muller. How profound you consider it to be is a subjective value judgement,
but I find it profound because it shows quantum indeterminacy may be a
direct effect of living in a reality big enough to contain all possible
observer states. In other words, one of the most mysterious and surprising
aspects of QM, can be answered, explained, and predicted as a direct
consequence of observer mind states being computed by something like
Bruno's universal dovetailer, Muller's graph machines, or Schmidhuber's
algorithmic theory of everything.

>
>> *> So could it be that all these great thinkers, Marchal, Muller,
>> Tegmark,*
>>
>
> *Tegmark yes, Muller probably not but maybe, I don't know much about him,
> but in no universe would I call Marchal a great thinker.  *
>
>
>
>>> *>>Standish demonstrates a keen grasp of the obvious!  Of course
>>> Schrödinger's equation can be deduced from observation, historically that
>>> is exactly how it was found. But nobody would have proposed such a crazy
>>> thing if the results of experiments hadn't demanded it. Yes it can be
>>> derived from pure mathematics, that is to say it has no mathematical
>>> errors, but an infinite number of equations can be derived from pure
>>> mathematics that contain no mathematical errors however very few of them
>>> have anything to do with physics and many of them have been experimentally
>>> proven to be wrong. *
>>>
>>
>> *> **Standish made some basic assumptions about the nature of
>> observation, and then showed how one can, starting only from those
>> assumptions, derive the Schrödinger equation deductively, (not
>> empirically).*
>>
>
> *If a conclusion is based on an observation, and his is, then it is an
> empirical conclusion because the Dictionary on my iMac says "empirical"
> means "verifiable by observation or experience rather than theory or pure
> logic". *
>

The derivation is deductive, the verification of course is empirical.

Think about it like string theory. No one ever has seen a string, it is a
premise from which string theorists start, and from that assumption they
try to deduce properties of physics which they can connect to observations
of our universe.

This is an analogy for what Standish has done, except unlike string
theorists, Standish has connected his premises (via deduction) to
predictions that match observed features of our universe.

* As I said before, Standish demonstrates a keen grasp of the obvious.
> Standish describes historically how Schrodinger's Equation was discovered,
> the only difference is that Schrodinger found it in 1927, I'm not sure
> exactly when Standish figured it out but I have a hunch it was after that
> date.*
>

I hope you are just playing dumb here.

> *> The more things are clumped together within a given volume, the lower
>> the entropy of that system is.*
>
>
> *That is true for some types of entropy, but the exact opposite of that is
> true for entropy that is produced by gravity, and gravitational entropy is
> BY FAR the dominant form of entropy in the universe.  *
>

Could you explain this more or provide links or references?

> *>>> the early universe could have begun in a maximum entropy state, where
>>>> everything was at thermal equilibrium. However, due to the expansion of the
>>>> universe *[...]
>>>>
>>>
>>> *>>That is impossible. The universe couldn't have been born in a maximum
>>> entropy state because the expansion would cause the entropy to become even
>>> larger. The reason comes down to gravity, for a gas high entropy means that
>>> the gas is spread out evenly, but when gravity comes into the picture high
>>> entropy means that matter is clumped together, like in a Black Hole. In the
>>> early universe matter was spread out very evenly so it had very low
>>> gravitational entropy. If the universe had been born at maximum entropy, it
>>> would have started as a collection of black holes, not a smooth plasma.*
>>>
>>
>> *> That's false. Black holes only have maximum entropy for a given
>> volume. They have less entropy than the same mass spread out over a greater
>> volume. This is clearly evident from Bekenstein's bound calculation.
>> Maximum entropy is proportional to (mass * volume).*
>>
>
> *No. When you take gravity into consideration entropy doesn't behave in
> the same way it does in high school chemistry. The Bekenstein-Hawking
> gravitational entropy of something is proportional to the SQUARE of its
> mass and is proportional to its AREA, not its volume. They aren't opposing
> forces; More Area = More Entropy  and Less Area = Less Entropy.*
>
> *Your confusion may arise because in everyday life if you pack more mass
> into the same volume the density goes up, but for a black hole, as mass (M)
> increases the volume grows so much faster (M^3) that the average density
> drops. The largest known black hole has a mass of 66 billion suns, but its
> density is less than that of water and is only slightly denser than the air
> that we breathe. *
>

I know it increases by the square of the radius. That's not at issue. The
issue is that the same mass in a larger volume of space, still has a larger
bound on entropy than the same mass in a smaller volume of space. Paste the
Bekenstein bound formula here and prove me wrong.

> *> in the early universe (say when it was a quark-gluon plasma) was likely
>> at or near a maximum entropy state (for that epoch of the universe).*
>
>
> *No! If you want to calculate the entropy of the universe during the
> quark-gluon plasma era, or any other error for that matter, the positions
> of quarks and gluons is of trivial importance; the important thing is the
> gravitational entropy, it's about 10 trillion times larger than all other
> sources of entropy combined. Nearly all the entropy in the universe is
> contained within black holes, the super massive variety being the most
> important. *
>

There is one sense in which I can see that as true. But then there is
another sense in which I question it. Consider that the energy density of
our entire observable universe is identical to the density of a black hole
of the same size as the observable universe. Would that not mean that the
information content of the observable universe is already in some sense at
the maximum only could be (according to Bekenstein's bound)? And that all
energy in the universe contributes equally to it?

Jason
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20260225/dba1af75/attachment-0001.htm>