Closed jonrkarr closed 3 years ago
But the update policy is a critical part of the logic modelling approach. Synchronous and asynchronous update will give different results (different state graphs).
Now, perhaps they should be parameters instead of algorithms.
Original comment by: gambardella
Indeed, see the other tickets I created. I should have mentioned it here: The idea is to remove the terms from the Algorithm branch and add the feature to the Characteristics branch (rahter than Parameters branch, actually)
Original comment by: dagwa
Adding them to the chararcteristics branch would not solve anything. You would still need two algorithms that bear those characteristics.
Original comment by: gambardella
Ok, so you would rather "double" the concepts and leave them in the algorithm branch as well as in the characteristics? This would be fine with me.
Original comment by: dagwa
We do not double the concepts. If synchronicity is a characteristics, synchronous and asynchronous are two different simulation algorithms. That would be different if there was a parameter called synchronicity that would take two values, synchronous and asynchronous.
In any case, those terms have been requested by the colomoto community, for use in SED-ML. I think they should be the one deciding at the end. Maybe this should be moved to sed-ml-discuss?
Original comment by: gambardella
Original comment by: annazhukova
Let's keep the terms for now then. I have added the sudgected characteristic terms.
Original comment by: annazhukova
We suggest to depricate these terms, because the rather describe the update policy of the model than the actual simulation algorithm.
Reported by: dagwa