Name: Siver Andrey Date: 05/13/04-09:40:40 PM Z
Hi All.
I would like to make several suggestions and get reactions on they.
I want use FeynCalc as a framework to build somelike expert system (ES) for the calculation in QFT. It implies that programmer solves a task by designing new “derivation rules” rather then new [Computer Algebra System’s] functions. Each “derivation rules” is described by two functions:{Condition_to_apply, Rule_to_apply}. This way of the calculation would become quantitative new in comparison with the procedural-programming because of the following reasoning. Since a chain of the applyings of the rules can become infinitely long and for each piece of the chain it can be assigned the function in correspondence, so only infinite number of programmers :) of procedural-programming paradigm could made such number of the functions.
With this email I have attached a very simple example of the
calculation in “derivation rules” paradigm taken from [1]:
LorentzIndex simplification calculations in the symmetric
Energy-Momentum-Tensor (of free electro-magnetic field) building.
Only 3 “derivation rules” were created:
1) Simplification by renaming free indexs (`Simplify001’ function);
2) Simplification by contracting indexes (Uses `Contract’ function);
3) Simplification by ordering partial derivations (`SortingPartials’
function).
During my calculation I had had a question and found a misprint in
help:
1) Why FunctionalD[ f,
QuantumField[PartialD[L.i.[m]], A,
{L.i.[n]}] ] does not make calculation but for
QuantumField[PartialD[L.i.[m]],A,{L.i.[n]}]->QuantumField[B,{L.i.[m],
L.i.[n]}] does?
2) In the help for `Symmetrize’ function word
“antisymmetrizes”->“symmetrizes”.
Ref. [1]: M. Peskin, D. Schroeder, “An Introduction to Quantum Field Theory” / Trans. from English edited by A. A. Belavin, A. V. Berkov // M.-Izhevsk, NIC “R\&C Dynamics”, 2001; Task 2.1(a,b)
Best regards,
Andrey Siver,
Russia, Protvino