## Stream: maths

### Topic: linear_ordered_semiring

#### Scott Morrison (Sep 28 2020 at 06:53):

linear_ordered_semiring does two things at once, relative to ordered_semiring: it says the ordering is linear (rather than partial), and it asserts 0 < 1.

#### Scott Morrison (Sep 28 2020 at 06:54):

What if I want just 0 < 1, without the linear ordering. (e.g. most of the orders on rings that I know about... :-)

#### Scott Morrison (Sep 28 2020 at 06:54):

What are even some examples of interesting ordered_semirings where we don't have 0 < 1??

#### Patrick Massot (Sep 28 2020 at 07:37):

The order hierarchy is a huge mess, because there are way too many classes including the relation instead of just properties. We simply need someone with a huge amount of time and energy to refactor everything there.

#### Scott Morrison (Sep 28 2020 at 08:49):

Okay. I'll take that as approval to try out any changes that I feel like. :-)

#### Scott Morrison (Sep 28 2020 at 08:50):

But the general trend should be towards splitting out "property only" classes as mixins rather than part of a hierarchy.

#4296

#### Marceline (Oct 08 2020 at 13:39):

I’ll also start cracking on this then, just graduated, got a while load of time now ha.

Last updated: May 11 2021 at 16:22 UTC