Dan Kubb

@snusnu the primitives are in place, but I agree that it might make sense to have convenience methods for the common cases like `#count` and the other [aggregate operations](https://github.com/dkubb/axiom/tree/master/lib/axiom/aggregate). As...

@snusnu does this work like you expected it?

> Are you thinking about refactoring axiom's spec suite? @solnic I was thinking about it. I still really do like the per-method specs, but I realize lots of others don't...

@snusnu there's two ways to do it: 1. Rename all the attributes in the "inner" relation so they are unique and don't conflict with the outer relation, and then transform...

@ktdreyer I assume this is with edge ice_nine? Can you try with 2.0 and 2.1? I just tried quickly to reproduce it here with 2.1 and 1.9.3, and it passed...

That's a good idea. I may add something like `IceNine.no_freeze(*mods)`. I would then exclude a few common namespaces like `Spec`, `RSpec`, `Mocha` and `Test`.

I'm not going to merge this into master yet. We need a period of non-volatility before we start talking about 1.0. @sferik I wouldn't exactly classify that as a crazy...

Yeah, no worries. Those kinds of issues will always be found as you introduce new constraints into the system. For example, someone could use the library with JSON at some...

@mbj I wonder if instead we should memoize a method based on the class _and_ method name, rather than just the method name. This should allow inherited methods to be...

@mbj the question is if we want to support a scenario where a subclass method is _not_ memoized, but the parent class is. We may want to make the call...