Cristinel,
Yes a theory that fits all observable data may exist. (It actually does exist if you are considering the global set of all observable data or if you are only considering the true data in a local observable data set). I think we can agree on that. Yes your example was oversimplified (which in essence was all that I was saying). Not a bad beginning guess. Guesses usually need to be refined with additional input to get a final fully valid form. It is really a much more difficult problem than either of us was presenting. I only introduced a couple of minor incremental complexities. I stayed away from more difficult levels such as the possible roll of random structures, etc. That is one of the real questions, can all experimental data be explained by a rule or set of rules or is some of it just random in nature with no discernable rules. I believe your assumption is right, but if quantum physics were right, that concept might not hold. Giovanni is right because the search for anything is based on faith. It is only after you have found what you are searching for, fully understand it, and can absolutely prove its existence to be as you observe it to be that you can say that faith is no longer needed. Man has not found any way to absolutely prove anything, so we really all live in faith in all things. It may be faith based on evidence, but it is faith nevertheless. In reality the ultimate theory of everything will be a very complex and large hierarchical set of interacting rules. What man would like to find is a single base theory or set of theories from which all others expand or follow to explain the complete world. If the world was created by an intelligent being that would be the most likely case. If the world came from random happenings, it would more likely have many bases from which different parts of the world sprang due to many chance happenings over time. In that case a single base theory or set of theories would only expand to describe a part of the world. One would then have to understand all of the sets of base theories and all of their interrelationships (interactions with each other) to get a fully expanded understanding of the world. These are concepts that are well beyond what you presented though, so I will go back to basics so as not to cloud the issue any further. I also believe (have faith) that there is a set of rules in the absolute sense that completely describe the world and I believe that there is just a single base. That is one reason that I said that we are in close agreement in many ways. From man's local perspective, it is likely that man could obtain a close approximation of the absolute set of rules through the level of fifth vector structuring given enough time. The structural disconnect will prohibit man's observation of higher structural levels. If the assumption that the world is completely based on a set of structural rules is true (not random structure), then there is a set of actual rules that describe it and of which it is actually composed. It would therefore likely be possible to make a representation of that set in some other form unless it is infinite, which is not likely although I will not go into why at this time. The world is composed of motions and motions at their simplest level are relatively simple information structures. They mainly possess position, direction, and motion amplitude. These can all be described in man's mathematical terms, so you are right that the world can be explained in man's mathematical terms. It just becomes very difficult to work with in some large-scale interactive structuring descriptions. You do bring up a good point that all of the rules need to be consistent with each other to avoid destructive consequences. This is an argument for the likelihood of a single base structure rather than many randomly generated bases. Within our local perspective, positive observable evidence of something does tend to make it more believable than if no evidence is present, but the belief that the evidence that is observed represents actual reality always requires the same amount of faith in all observations in the long run because we can never prove it absolutely. From what we can see from our local perspective though, the world's construction appears to be based on a very intelligently designed set of rules that starts with some relatively simple structures based on some simple basic rules and then combines them together hierarchically in more and more complex ways to generate the total world that we live in.
Now concerning your perception of three objections from me. In reality I merely throw concepts out to encourage deeper thought.
1. You are right that we both did use set theory in our respective comments. I expressed mine in English and you expressed yours in more of a mathematical form replacing names with letters and placing them in parentheses suggesting a relationship between them according to some rule or rules. I am not saying that it was a real big problem because you did define what each letter meant. As I said the main thing that might have been confusing to some was that you said you would use such symbols as short names, not equations and then later used Kt:= K(E(Dt)) which is in the form of an equation. You may have meant it to be only a name, but a name that looks exactly like an equation could be confusing. Mainly I just wanted to see if you could communicate intelligently without the math abstraction and you have done well. I notice that this comment is completely math free. Very good.
You may be right, but I wouldn't count on it. There are many people in the world that would like to know generally about scientific concepts, but don't want to or don't have the time or money to spend getting all of the tools such as math expertise, etc. to be able to actively participate in science. There wouldn't be so many books or television shows like NOVA, etc. designed for such people if they weren't there. Some are still too young to have attained to a high level of math knowledge and the right motivation of learning about something new and interesting in a form that they can understand could make the difference of career choice for them. Ultimately the amount of money available for science in this world is at least somewhat determined by how many non-scientists believe it is worth the money and will vote for it.
2. I saw that you first talked about what appeared to be a local set of theories that were compatible with the local set of experimental data that existed at just a specific time because you said "for each set of data there is a set of possible theories of reality, E(Dt), compatible with that set of data." The implication here is that one set of observable data has a set of theories that describe it and another (say later) set of observable data would have its own set of theories that might differ from those in the first set of observable data because of the difference in the data in the two observable data sets. Later I saw that you appeared to be talking about the ultimate set of all theories that are compatible with the ultimate set of all observable data because you said that E(Dt) should contain all theories, which implies that the global set of all possible theories that could be compatible with the observed data from any possible time would be considered as compatible with all possible observed data sets from all possible times, but it was not clear how you got from the one to the other. I believe part of the problem was that we were applying the word compatible in different ways. I was considering that compatible meant that the compatible theories would be in correspondence with the data so that the compatible theories would explain the complete set of data and no more. In that way of looking at it, the data set generates the theories by informing the theories with the data elements that are then explained by the theories. Such theories would not contain extraneous descriptions of things that were not in the data set at that time because there would be no reason to believe from the then current data set that such extraneous descriptions would have any basis in reality. A theory that contained explanations for things that were not in the current data set would not be compatible with it because those things would not correspond with the data in the data set. You seem to have looked at it with the idea that any possible theory that was compatible with any data set would also be compatible with all other data sets (all sets that were both forward and backwards in time from the compatible set) because you considered theory elements that would be extraneous at one time, but valid at a later time due to added new data as still compatible with the theories from previous times when the data set would have no need for them to completely describe the then current data set. Your approach could be valid in an abstract way considering the abstract global sets and their theories, but it seems that it would not work very well in a real world search for the ultimate theory of everything because even if you achieved the goal of obtaining the complete observable data set, and used it to generate the theory of everything that completely explained the data set, you would also have to consider as still valid any theories that added any additional extraneous descriptions of imagined additions that went beyond reality even if there was no evidential basis for them in the data set. Since reality is likely only a small subset of the set of all things that could be imagined to be parts of reality, the theory of everything would just continue to grow over time with more and more nonsensical additions given to explain things that don't really exist. It sort of sounds somewhat like string theory. Very interesting. Although it can be interesting to think about global and absolute concepts that may or may not exist, but are beyond our ability to observe or prove, such things are only useful to man when they are brought down to man's local level of observation. Of course, that is where man's interactions with them can modify their outcomes and results. This interaction does make for more complex structures to fully comprehend, but if you do, the knowledge is much more useful in the real world, which is the one that we live in after all. In the real world the possibilities exist that data believed by man can be either true or false at any given time and that man can be led by the data to only generate theories that agree with the data and not see possibilities that new data will lead them to later even if the new theories could be said to be an extension of current data. It is evident that there is a possible description of everything in the world because it is already recorded in the world to make it up. It is very likely that it can be recorded (a literal or abstract image of its rules can be made) in some other form also. It is also evident that there are at least a large number of possible false or partly false theories of everything. The problem is that there is only one completely true explanation, but many false ones. The useful concept would be one that would point to the true theory and point away from all the false ones. Since time is a part of the world, removing it from consideration would only go farther away from reality making it less useful.
3. Again, as I said in my last post, you are right from a global perspective, but we don't have a global perspective so it does us very little good in any practical way. We can look back and say that at the time that observed data agreed with the concept of the existence of the atom as the smallest known particle of matter and gave no indication that anything smaller existed, the data did not preclude that smaller particles could exist and make up atoms, but those who looked at that data would have no reason to believe that such smaller particles existed so there would be no reason for them to make up theories that included what to them would be imaginary particles that were not needed to describe the world as it was then known. In reality it is not the possible theories that matter, it is the theories that man actually comes up with or can access based on the observed data at the time that determine man's progress. In the long run I think that the problem is that you are looking at the problem from an overall generalized global perspective of how it would look if we could see all of the possible observational data in reality and all of the possible theories that could attempt to describe the data during all possible times and I am looking at the same problem from the local perspective of what man can actually achieve in his path through all of those times with the observational data that he has access to in each of those times to generate theories that come as close as possible to describing the known data to work toward the time when we could hope that the complete data set would be obtained and a complete theory that would explain completely all of the data and would not require any other data beyond the observed set for a complete understanding of reality would be obtained. Not that I expect man to fully attain to that goal, but much will be gained in the attempt.
Algorithms are sets of instructions or operations that are performed according to predefined rules and/or by rules that are contained within the set of the algorithm. You are right that they can be used to compress or decompress data. In terms of what we are talking about here they are very useful in modeling the data and its rules in a way that simulates reality. Depending on what type of data is involved it is often not possible or at least not practical to use data in its most compressed format. It must often be decompressed to its native format before practical use. The same problem can appear in some theories that contain complex math structures because a math structure such as an equation may contain a symbol that represents a whole series of other math operations on data in other equations, which may also contain symbols that in turn represent more math structures, etc. If the problem were to be worked by a man he would have to unpack the structures do the operations then repack the results to feed the next level of structure. Of course, with computers, the computer can be programmed to do the decompression and recompression along with the data operations at each level, so it only slows things down somewhat, but would likely still be faster than typing the complete uncompressed data structure into the computer.
The if, then, and else, etc. operators are just ways of showing dependencies, relationships, or interactions between entities. Dependencies and interactions are built into and are parts of the world that we live in. They can be expressed in other ways, but they are the same thing no matter how you express them. I can say, if the resistance decreases or the voltage increases then the current in the circuit will increase or I can say I=E/R. In either case I am just showing that there is a dependency relationship between voltage, current, and resistance in an electric circuit, so that a change in one will cause a change in another. The I=E/R is a conditional expression that expresses the same dependency relationships as does the if, then statement. (I only showed one of several if-then statements that would fully show the complete dependency relationship). You could only eliminate all statements in the theory of everything that show that the condition of one thing is dependant on the condition or action of another thing if you eliminated all interactions between all things because an interaction is basically a change in one entity caused by another entity (it could be mutual changes to each other). Another way of saying the same thing is that the condition of one entity is dependant or conditional on an interaction with another entity. The whole world is based on dynamic interactions between entities within it. It's cause and effect in action. If this cause happens then that effect results.
I see that our conversation has been moved to my forum.
To everyone:
If anyone reads Cristinel Stoica's conversation with me here and wants to see the beginning of it you can find it in Giovanni Amelino- Camelia's forum under his paper "The Fairness Principle and the Ultimate Theory of Everything. It starts somewhere around October 25, 2009.