I skimmed the paper again, and skimmed his other paper in ref 3, and my total impression is that his research interest lies into understand cosmological problems, such as the acceleration and dark energy problems. And in that context, a possible revision of GR at large scale to help solve the dark energy seems like a clear possibility. So far so good.
But at least from my point of view, the suggested exact implementation of his conceptual idea seem to be a totally wild conjecture. Some plausible reasoning to provide the reader with some confidence in this conjecture would have been nice. If it's supposed to be there, I don't see it.
So disregarding the exact form of this conjecture, perhaps one can try to see what possible logic there is behind the conceptual general part he calls the "Information postulate" (IP), which at least seems to be possibly interesting.
As a background, he takes the example of the elevator analogy, and notes that the balancing needed to maintain the exact right acceleration to be equivalence to a gravity field, requires the rocket operator to be informed about how to tune the engine. Without this information the equivalence comparasion can not be physically realized. And also without this information, any observer is destined to free-fall.
As a way to extend the Equivalence principle, his "Information postulate":
Spaan said:
...every observer is perfectly able to free-fall through space-time under the influence of gravity, which suggests that some form of information is available to tell each observer how to move individually. Therefore, one is lead to the tempting assertion that gravity is not imposed on individual observers as an exterior rule, but that it is imposed on them by the information that is actually necessary to realize specific gravitational dynamics.
So what does this mean? Here is an attempt to reinterpret this.
Spaan said:
...every observer is perfectly able to free-fall through space-time under the influence of gravity, which suggests that some form of information is available to tell each observer how to move individually...
Informationwise, I tend to think of the geodesics as a result of the observers LACK of information about it's own global environment. The geodesic is then somehow the result of the observer doing a kind of random walk, where the feedback he continously gets from it's environment, slowly curving this path. The deformation is as I see it the result of a least speculation principle.
Thus the geodesic would then IMHO, better we described as the result of an evolving inside observer, randomly walking in an environment which is uncertain to him. The curved path is what a third observer would see. But I think what he says that the reason it's curved is not because there is an external constraint to guide him, but rather it's the lack of information that causes the inside observers random walk to become curved.
So the best of my understanding, would suggest that what Spaan refers to as "information is available to tell each observer how to move individually" is rather the feedback of the actions carried out. Which can be though of as infinitesimal corrections to his information, the correction of the information is what slowly causes the curved path. The inertia implied in minimum speculation, implies that that information is changing only slowly, in a "conservative way"
In this way, gravity is the result of the process of environmental feedback to an observer beeing forced to act, (do a random walk) in an environment based on incomplete information. Random walk here, is my metaphor for playing a game, place your bets to the best of your infromation, and let the feedback from the game, suggest howto revise your information. During the game, your information is deforming. Of course it remains to separate from the general abraction of information, what degrees of freedom that constitue spacetime.
This could possibly be what he means with this:
Spaan said:
Hence, under the IP, it is the insufficiency of non-gravitational information to define particle dynamics that imposes gravity.
If that is so, I think he is onto something, but I can't see how his simple attempt to account for this can be satisfactory. As I see it, it is far more involved. I think such a deep idea has far greater potential than his simple treatise admits.
Clearly the physical meaning of "information" here is paramount. Which he also admits in the paragraph where he says
Spaan said:
This is a good point to be more precise about information.
And from that point on I can't stretch myself to see that his attempt at beeing precise is sufficient. I still wonder if I missed something? As I see it, I would combine this idea with an emergent spacetime, and thus hopefully Einsteins Equations (with possible corrections) should follow from a fundamental reconstruciton. Instead he suggests a mystersious patch onto the existing equation, I do not follow that logic. It all looks to me like a completely wild conjecture. A bit too wild, even though his basic motivation seems interesting.
I agree though that the information does matter, and most probably Einsteins equations when we understand how it emergens as per this logic, will get corrections and some of these will involve the third observers own complexity scale. What he suggest might be something possible, but I would expect a serious implementaiton of this idea to be much more involve, and pretty much involved a fundamental derivation of Einsteins equation, which would be emergent.
I suspect that his prediction here are just a toy argument, to show the possible potential of the information postulate, while a proper implementation is left open.
??
/Fredrik