I just watched this talk by Rick Hickey and it was very eye opening. Now I can't simply go back and use postgresql and mongodb without thinking how much more interesting it could be. There is just one thing that I don't quite understand.
If I understand correctly, the basic premise here is that your database has to be a persistent data structure and how you work with it is by using pure functions that receive it as an argument and return a new one, pretty standard functional programming.
But what if I have a huge amount of data? All of that is just supposed to be constantly going back and fourth between my code and the datomic instance? How can that be efficient? Or am I misunderstanding something here?
Thanks