In the old days with meteor, I did not think too much about how to manage data. just define collections, define pubs, and do subs in client as needed.
But these days, there are some discussions about data management. such as:
- in infinite scroll case, will all the old data be resent after the limit increased?
- in items-authors case, will all duplicate authors data been sent?
with redux, we can manage app state in a single place. but it seems meteor has its own idea about data management. with all the problems, how should we do data management? should we still pub/sub as need, or do some centralized data manage, or apply some lib or framework, or apply some special patterns?