Nicola Mometto

what cases would something like ``` clojure (defn definline-hof? [{:keys [op args]}] (and (= :invoke op) (some #(= #'definline (:var %)) args))) ``` not be able to catch?

I have investigated http://dev.clojure.org/jira/browse/CLJ-1544 today and I can confirm this is the root cause for this ticket.

Here's hoping we'll be able to close this 7yo ticket http://dev.clojure.org/jira/browse/CLJ-19

I can't imagine how this could be an issue with tools.decompiler as there's no compilation/evaluation involved

> A big design question is whether to pick: > > * never empty: > ```ocaml > type 'a t > > val create : init:(unit -> 'a) -> unit...

note: briefly discussed with @ewenmaclean and @c-cube, the current use of `duration` to represent FIX's tenor is not good, we should have a custom tenor type implemented as a sum...

> _Coercion strategy_. @bronsa, would not holding `coercion_strategy` (and `effective_type`) in `metabase_field_user_settings` result in no update to coercion strategy (nil) when base type of a field changes during sync? No,...

> _Database overrides_. @thebiglabasky, (1) with this change we lose way of shutting down unfolding on database level (connection page). If field is overridden to unfold, changing the setting on...

> _Coercion_. My observation: > [..] Ah, apologies for misunderstanding you. I see your point now. That is indeed wrong, thanks for spotting and testing. I will follow up with...

> One more thing that I'm unsure of is the nature of the change to the behavior of `t2/update! :model/Field`. > > With this change merged we'd get to the...