Proliferation of Group classes

Question

Proliferation of Group classes

Closed this issue 4 years ago · 2 comments

Groups turn out to be a useful structure in Haskell, and I've seen them in a number of other packages (groups, and reinvented again inside patch).

As people become more aware of them, it will be frustrating if there's a proliferation of similar-but-incompatible Group classes with different instances on different types. Is there a way to depend on groups to get what you need?

Answer 1 · 2020-12-23T11:29:11.000Z

Also group-theory was recently released.

Answer 2 · 2020-12-23T23:01:34.000Z

Looks promising. At the time I wrote this I needed loops and magmas, which are non-monoidal objects weaker than groups <https://en.wikipedia.org/wiki/Magma_(algebra)>. One very unfortunate aspect of the Haskell type class system is the inability to easily introduce upper dependencies to a pre-existing class. Your pet peeve is thus fairly endemic to the ecosystem. I'd suggest voting with your own upstream dependency selections rather than attempting to police the matter.

…

On Wed, Dec 23, 2020 at 3:29 AM endgame ***@***.***> wrote: Also group-theory <https://hackage.haskell.org/package/group-theory> was recently released. — You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub <#1 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABUNG7OE6DNKMV4ZDUNO5ZLSWHIBJANCNFSM4KY2HJSA> .