Frederik Hanghøj Iversen
|
63b5f5c68d
|
Use long name for product object
|
2018-03-08 10:46:28 +01:00 |
|
Frederik Hanghøj Iversen
|
486238e114
|
Add goals for propositionality of products
|
2018-03-08 10:38:46 +01:00 |
|
Frederik Hanghøj Iversen
|
1ef57a19f4
|
Cosmetics
|
2018-03-08 10:30:35 +01:00 |
|
Frederik Hanghøj Iversen
|
4e7b350188
|
Factor out objects
|
2018-03-08 10:28:05 +01:00 |
|
Frederik Hanghøj Iversen
|
181bd1af53
|
Factor out category
|
2018-03-08 10:24:17 +01:00 |
|
Frederik Hanghøj Iversen
|
faf4c54188
|
Make parameters explicit
|
2018-03-08 10:22:21 +01:00 |
|
Frederik Hanghøj Iversen
|
fae492a1e3
|
Restructure products
|
2018-03-08 10:20:29 +01:00 |
|
Frederik Hanghøj Iversen
|
b61749bb91
|
Fixup some todo-notes
|
2018-03-08 01:10:52 +01:00 |
|
Frederik Hanghøj Iversen
|
fa9a470875
|
Update backlog
|
2018-03-08 00:54:42 +01:00 |
|
Frederik Hanghøj Iversen
|
e43bee6d9f
|
Feels really close
|
2018-03-08 00:36:38 +01:00 |
|
Frederik Hanghøj Iversen
|
c8fef1d2b5
|
Use different name for function composition
|
2018-03-08 00:22:55 +01:00 |
|
Frederik Hanghøj Iversen
|
36cbe711fb
|
Sort of half of the proof of an inverse
|
2018-03-08 00:09:49 +01:00 |
|
Frederik Hanghøj Iversen
|
459718da23
|
Finish proof of equivalence of klesili/monoidal categories!!
|
2018-03-07 17:30:09 +01:00 |
|
Frederik Hanghøj Iversen
|
19103e1678
|
Update cubical
|
2018-03-07 16:24:43 +01:00 |
|
Frederik Hanghøj Iversen
|
93d075a6d3
|
Attempt at proving pureNTEq
|
2018-03-07 15:23:07 +01:00 |
|
Frederik Hanghøj Iversen
|
00e6e1aa66
|
State problem with approach
|
2018-03-07 11:45:11 +01:00 |
|
Frederik Hanghøj Iversen
|
aa64e01084
|
Remove some cruft
|
2018-03-07 11:33:08 +01:00 |
|
Frederik Hanghøj Iversen
|
125123846e
|
Lay out a strategy for showing the equivalence
|
2018-03-07 11:29:58 +01:00 |
|
Frederik Hanghøj Iversen
|
085e6eb3d7
|
Stuff about voe-2-3
|
2018-03-06 23:18:33 +01:00 |
|
Frederik Hanghøj Iversen
|
110e3510c5
|
Use postulates
|
2018-03-06 15:55:03 +01:00 |
|
Frederik Hanghøj Iversen
|
5ae68df582
|
Prove that fmap is mapped correctly
|
2018-03-06 15:53:11 +01:00 |
|
Frederik Hanghøj Iversen
|
4d528a7077
|
Clean-up
|
2018-03-06 11:25:29 +01:00 |
|
Frederik Hanghøj Iversen
|
485703c85e
|
Tidy up
|
2018-03-06 10:16:42 +01:00 |
|
Frederik Hanghøj Iversen
|
0cebe1e866
|
Make private
|
2018-03-06 10:06:45 +01:00 |
|
Frederik Hanghøj Iversen
|
4de27aa06c
|
Naming
|
2018-03-06 10:05:35 +01:00 |
|
Frederik Hanghøj Iversen
|
9173468b03
|
Use omap/fmap
|
2018-03-06 09:56:44 +01:00 |
|
Frederik Hanghøj Iversen
|
bdd67aee53
|
Rename RR to Romap
|
2018-03-06 09:55:18 +01:00 |
|
Frederik Hanghøj Iversen
|
c57cd5c991
|
Define stuff in monoidal record
|
2018-03-06 09:52:37 +01:00 |
|
Frederik Hanghøj Iversen
|
cfb7925cb5
|
Renaming
|
2018-03-06 09:45:04 +01:00 |
|
Frederik Hanghøj Iversen
|
b6457a0b14
|
Add comment
|
2018-03-06 09:41:29 +01:00 |
|
Frederik Hanghøj Iversen
|
7647a452cd
|
Tidy up proof a bit
|
2018-03-06 09:39:48 +01:00 |
|
Frederik Hanghøj Iversen
|
35419ad86e
|
Rename eta and mu
|
2018-03-06 09:35:50 +01:00 |
|
Frederik Hanghøj Iversen
|
f8e08288a0
|
Cosmetics
|
2018-03-05 17:31:13 +01:00 |
|
Frederik Hanghøj Iversen
|
9ec6ce9eba
|
Use other equality principle
|
2018-03-05 17:10:41 +01:00 |
|
Frederik Hanghøj Iversen
|
3151fb3e46
|
Prove propositionality for naturality
|
2018-03-05 16:35:47 +01:00 |
|
Frederik Hanghøj Iversen
|
7f4a8a65b8
|
More stuff about opposite being an involution
|
2018-03-05 16:10:27 +01:00 |
|
Frederik Hanghøj Iversen
|
b26ea18257
|
Cleanup in nattrans
|
2018-03-05 15:04:16 +01:00 |
|
Frederik Hanghøj Iversen
|
ddd5f17c05
|
Move propositionality stuff about natural transformations to that module
|
2018-03-05 15:02:36 +01:00 |
|
Frederik Hanghøj Iversen
|
2b92cee254
|
Prettier names in Fun
|
2018-03-05 14:55:45 +01:00 |
|
Frederik Hanghøj Iversen
|
bb379fa196
|
Implement category of presheaves
|
2018-03-05 14:50:53 +01:00 |
|
Frederik Hanghøj Iversen
|
ce4dd83969
|
Prove that the yoneda embedding is distributive
|
2018-03-05 14:42:12 +01:00 |
|
Frederik Hanghøj Iversen
|
7fbca1aeeb
|
Clean-up yoneda embedding
|
2018-03-05 14:04:04 +01:00 |
|
Frederik Hanghøj Iversen
|
1bf565b87a
|
Have yoneda without having a category of categories
I did break some things in Cat.Categories.Cat but since this is
unprovable anyways it's not that big a deal.
|
2018-03-05 13:52:59 +01:00 |
|
Frederik Hanghøj Iversen
|
5c3616bca5
|
Make argument to presheaf explicit
|
2018-03-05 11:17:31 +01:00 |
|
Frederik Hanghøj Iversen
|
059c74b687
|
Use already defined category
|
2018-03-05 11:15:45 +01:00 |
|
Frederik Hanghøj Iversen
|
a4890a42cf
|
Define Monoidal categories without depending on category of categories
|
2018-03-05 11:13:58 +01:00 |
|
Frederik Hanghøj Iversen
|
5902c6121b
|
Further reduce dependency on impossible facts.
Provide the data for the product in the category of categories without
requiring such a category to actually exist
|
2018-03-05 11:07:42 +01:00 |
|
Frederik Hanghøj Iversen
|
77006011d3
|
Minimize dependency on category of categories
|
2018-03-05 10:35:33 +01:00 |
|
Frederik Hanghøj Iversen
|
8f8800cb67
|
More stuff about kleisli \equiv monoidal
|
2018-03-05 10:28:16 +01:00 |
|
Frederik Hanghøj Iversen
|
b079f5e426
|
Prove propositionality for IsMonad
|
2018-03-02 13:31:46 +01:00 |
|