GE1/1 trigger fibers are swapped in the backend
This problem was already discussed in the past and somewhat got forgotten.
It appears that the trigger fiber numbering convention is swapped between the OptoHybrid and the backend. As far as I can tell, the OptoHybrid and the OTMB use the same convention while the GEM backend uses the opposite one. The definition of fiber 0 and fiber 1 should be swapped.
This is not a critical issue for now since all clusters are considered for physics, regardless of their order. It is simply confusing. In the future, it may become an issue, depending on the protocol, if clusters form the previous BX are sent in the next BX.