Robert Fletcher

Also tried adding `execute("ALTER TABLE students ALTER COLUMN data SET DEFAULT hstore(array[]::varchar[])")` to my migration, but it's still `nil`. From here: https://github.com/engageis/activerecord-postgres-hstore/issues/22

Adding `serialize :data, ActiveRecord::Coders::Hstore` to my model fixed this.

Yeah, it doesn't add it to the database. The serializer assigns an empty hash to it on the fly.

This issue [is purportedly fixed in Thin 1.8.0](https://github.com/macournoyer/thin/issues/372), so if the version constraint is adjusted maybe the issue would go away?

I generally just ignore those messages. It's looking for method definitions within your file, and when it doesn't find them for superclass methods it just moves on and doesn't bother...

It's a little hard to understand what you're saying. What I seem to be hearing is that your validations aren't running during mutations, otherwise they would cause the specs to...

@mbj Ha, well technically, it's at `v0.0.2`, but yeah, there's not much to it. Right now I've just been using the readme to document how I use mutant. Unfortunately, I...

@tjchambers I actually find that a lot of times mutant causes me to want to alter my code because it shows me that I have unnecessary code or I've made...

@tjchambers Yeah, it feels strange at first. Especially as your code improves, it starts to feel pretty pedantic to write a spec for a one-line method. But it's worth it....

I also find this recommendation to be problematic. As mentioned above, it's common for records to have the same `created_at` timestamp, as we often batch insert records. The only reliably...