Changes between Initial Version and Version 1 of Ticket #23204, comment 22
 Timestamp:
 07/20/17 00:26:55 (4 years ago)
Legend:
 Unmodified
 Added
 Removed
 Modified

Ticket #23204, comment 22
initial v1 20 20 > }}} 21 21 > What I am saying is this is really (relatively) expensive for something that is almost certainly going to be implemented. If your category has morphisms such that you cannot effectively simplify a composition `f \circ g` to a single map `h`, then I don't see how you would have enough (generic) information to check that the composition is injective/surjective. So I doubt this would ever get used. 22 I do not understand how this relates to `_composition_`. The point here is simply that in the category of Rings, a morphism that has a domain as its field, knows that it is injective. However, we can not so easily turn the composition into one morphism.22 I do not understand how this relates to `_composition_`. The point here is simply that in the category of Rings, a morphism that has a field as its domain, knows that it is injective. However, we can not so easily turn the composition into one morphism. 23 23 24 24 > > > No bare exceptions.