Syncing optimization

[juanca]

I think there are some possible optimizations to avoid clobbering the store.

1. Fetching event

The fetching flow emits a bunch of events in our store. The logged output looks similar to:

REQUEST_COLLECTION
ADD_MODEL
ADD_MODEL
ADD_MODEL
ADD_MODEL
ADD_MODEL
ADD_MODEL
ADD_MODEL
ADD_MODEL
ADD_MODEL
ADD_MODEL
SYNC_COLLECTION

2. ... Not sure about other code flows but I can see this being a problem throughout the backbone event system.

---

A few solutions off the top of my head:

• Silence events when using the brainstem-redux API? This does not seem ideal since it seems to be side-stepping the issue
• Stateful handling of events (e.g. do not handle events after REQUEST_COLLECTION until SYNC_COLLECTION). This might not work if there is a server error?
• Throttle event handler. This only works if the event handler always does the same thing (regardless of event type) -- always syncs the data between storage manager and the store.