This one is from a discussion with @CohenCyril, I'm a bit out of my waters so he might be able to better explain what's going wrong. The issue appears here, with the need of a notation/definition to force inference to re-infer the right structure for a term. Afaiu, the problem comes from the fact that the carrier for which I want to infer the structure is not a constant, but some set A where set is a variable (and the argument goes that as long as set has a given structure, then set A is a poset). I'll let Cyril minimize it and/or explain it better, he seemed quite keen :)
This one is from a discussion with @CohenCyril, I'm a bit out of my waters so he might be able to better explain what's going wrong. The issue appears here, with the need of a notation/definition to force inference to re-infer the right structure for a term. Afaiu, the problem comes from the fact that the carrier for which I want to infer the structure is not a constant, but some
set Awheresetis a variable (and the argument goes that as long assethas a given structure, thenset Ais a poset). I'll let Cyril minimize it and/or explain it better, he seemed quite keen :)