There was a pretty good question this morning when I checked comments on posts.
“What is the difference between projections and say rxjs”. We could as easily include any of the functional reactive libraries out there (bacon.js, flapjax, elm, rx, etc). We can also start going way back and looking at predecessors like CLM
In particular the tools are handling CEP
When we talk about projections, the streams are persistent (and distributed). We host v8 internally and use this API as our query language over persistent streams of events. A projection is not just for running from this point forward but for running against previous history as well. This is a very powerful idea (and not a new one).
Projections are the query language of the Event Store. When you write a projection you can specify that you want it to run for history and stop or you want it to run for the history and continue. The Event Store will also handle you doing this in a durable, consistent, and distributed way (over terrabytes of indexed data). A projection will survive a power outage as example if configured to do so.
I hope this helps to clear up some of the similarities and differences between the options.