[twerquie]

> Refactoring across the call stack is orders of magnitude easier than refactoring across a socket.

That's assuming that things haven't become deeply coupled within the "call stack". Separating things across a socket often forces a separation of concerns, which does tend to make refactoring easier.

[bad_user]

In my experience communications across a socket does not force a separation of concerns. It doesn't even encourage it.

Heck, because asynchronous communication in a non-deterministic environment is so hard to deal with, my experience is that such communications encourage shortcuts to be taken, so yes, I think it encourages tight coupling.

The only thing that somewhat encourages a separation is having different people responsible for different modules and because people are selfish, they'll fight for their components to have less responsibilities, not more. So it becomes a territorial thing. But this happens only if you have seniors that know what they are doing, otherwise rookies or less competent folks end up cooperating to "get things done".

[AnonJ]

Weird reasoning, that. I hardly see how a selfish, "territorial" and noncooperative approach could benefit anybody. Sure people might cooperate, but that doesn't mean they will take shortcuts and mess things up.

[crdoconnor]

>Separating things across a socket often forces a separation of concerns

It doesn't really force it. It just makes life really difficult for you if you don't.

That doesn't mean it happens, though.

[ecoffey]

> That's assuming that things haven't become deeply coupled within the "call stack".

Yep; Monolith or SOA you still have to write SOLID code :)

> Separating things across a socket often forces a separation of concerns, which does tend to make refactoring easier.

It /should/, and if it does now your changing more than the one variable of "invoked across stacks -> invoked across socket", and are more likely to introduce a regression :)

[nickbauman]

This is why AppEngine modules are so awesome. Have a route that takes on the majority of your traffic? Slice it off the main app declaritavely to its own cluster of machines.