1)"In english please".
I have proved that absolutely anything may be explained via the simple constraint that there exists a function F such that only entities which satisfy the rule F=0 exist. The only question which then remains is "what entities exist and what entities don't". Just tell me what line of the proof you didn't understand!
2)"Bas Van Fraassen in his book 'Laws and Symmetry' has made a big point on this issue."
Apparently Yanniru is ignorant of it anyway!
3)"You are completely correct on this issue."
If our ignorance is absolutely guaranteed then the associated symmetries are also guaranteed!
4)"In other words, scientific anti-realism is correct?"
As I understand it, scientific anti-realism says that nothing is real; and everything is conjured up. This would imply that there is no "knowable" data. I certainly have not made that assumption!
5)"Define what you mean by 'knowable', 'true', 'prove'. I've never gotten any kind of definition from you on those terms. "
At the moment, my first two paragraphs constitute the best of my ability to clarify what I mean by the classifications "knowable" and "unknowable". If you cannot understand that, I am at a loss. Regarding the other terms, define them anyway you choose so long as your definition is at least halfway consistent with common usage.
The fact that you ask this question together with 1) above imply that what I am doing is completely over your head. The essence of my attack evolves from the realization that trying to figure out what is and is not knowable is a complete waste of time. As I have said a number of times, what I have done in analogous to Alexander's cutting of the Gordian knot. The issues you bring up are nothing more than strings sticking out of the knot. Nothing more than a distraction from the central issue.
If you understood what I am doing, you would understand that the exact definition of these terms is completely immaterial. My deductions are based merely on the existence of the two categories (which are to be handled differently as "knowable" data is the only thing which we know cannot change). That deduction is valid even if the content of either category vanishes. However, I doubt very much that either extreme could be considered a reasonable expectation.
Have fun -- Dick