Sébastien HOUZÉ

@pascalandy yes, I will stuck with docker stuff only till rancher-compose support docker-compose v2 format https://github.com/rancher/rancher/issues/3973 (because libcompose itself still not ready for v2 format https://github.com/docker/libcompose/issues/147).

@clescot yes that's great but volumes & networks have some issues opened https://github.com/rancher/rancher/issues/5714 & https://github.com/rancher/rancher/issues/5715. BTW, close to the goal ;)

Ok, this is not a BC break feature... any chance to get it backported to `1.x` or does it need a big internal refactoring?

@rtheunissen ok so the main issue is equality so it's not a big issue about ``. And yes why not with `Hashable`, we're ok that Set, Vector, ... don't implement...

First of all, to me identity comparison (`===`) is something else and not part of the scope of this issue. > For example, if we implement equals on Collection, and...

@rtheunissen yes ok for `$set->equals($another)` if it can help to maintain the polyfill at the same level of features alongside.

Yes, can be a scheduled deprecation but you're right it's probably not the time to do that as it would break adoption.

@Hywan I think that Ruler should stay a [Rete rule algorithm](http://en.wikipedia.org/wiki/Rete_algorithm) like lib. I'm not very confortable with the fact you can assign a value to any variable in this...

ok, so I'm fully ok with that :+1:

@Hywan just in case of BC break array notation support, the best thing should be indeed to adopt something similar to [matlab notation](http://fr.mathworks.com/videos/working-with-arrays-in-matlab-101637.html) no? So yes `[` and `]`, so...