-
Notifications
You must be signed in to change notification settings - Fork 100
Add lgseisenlem4 to iset.mm #5010
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
This is gsumzsplit , gsumsplit , gsumsplit2 , gsummptfidmsplit , gsummptfidmsplitres , gsummptfzsplit , and gsummptfzsplitl
This is gsumconst from set.mm where the sum is indexed by consecutive integers. The proof is a new one by induction.
This is gsummhm from set.mm but where the sum is indexed by consecutive integers. The proof is from gsumfzval , seqhomog , and some monoid theorems.
This is gsummhm2 from set.mm where the sum is indexed by consecutive integers. The proof is similar to the set.mm proof.
This is gsumconstf from set.mm but indexed by consecutive integers. It is also gsumfzconst from iset.mm but with a freeness hypothesis in place of distinct variables. The proof is similar to the set.mm proof of gsumconstf .
This is gsumsnfd from set.mm except that M has to be an integer. Naming this similarly to the other gsumfz* theorems seems more appealing than trying to use a different convention. The proof is similar to the set.mm proof.
Stated as in set.mm. This is provided for compatibility with set.mm but at least so far makes no effort to reduce axiom usage. The proof is based on the set.mm proof of cnfldadd .
This is gsumfsum from set.mm but where the sums are indexed by consecutive integers. As with the set.mm proof the proof separates the empty case (lemma gsumfzfsumlem0 ) from the inhabited case (lemma gsumfzfsumlemm , here proved by induction which is perhaps simpler than trying to adapt theorems like fsum3 ).
Stated as in set.mm. The proof is the set.mm proof with one small change in extensible structure theorems.
Stated as in set.mm. At least for now, this uses the same complex number axioms as cnfldmul . The proof is from ax-mulf and cnfldmul .
This is similar to drngui from set.mm but for CCfld in particular, not division rings in general. The proof is a new one from crngunit , dvdsrd , and recapb .
This is expghm from set.mm with not equal changed to apart. The proof is basically the set.mm proof but needs changes of not equal to apart and other intuitionizations on most steps.
Stated as in set.mm. The proof is basically the set.mm proof but requires intuitionizing in many places for a variety of reasons. Remove one sentence from the comment in set.mm (which seemed to be an incorrect copy-paste from lgseisenlem2 ).
New usage of "ax-mulcl" is discouraged (1 uses). | ||
New usage of "ax-mulcom" is discouraged (1 uses). | ||
New usage of "ax-mulf" is discouraged (2 uses). | ||
New usage of "ax-mulf" is discouraged (3 uses). |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why is it nessesary to use discouraged theorems here? Aren't there corresponding not discouraged theorems available?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The short answer is that the two choices are to use https://us.metamath.org/ileuni/cnfldmul.html (already in iset.mm) or add mpocnfldmul
(as it is stated in set.mm). The most plausible way to prove mpocnfldmul
in the short run is via ax-mulf
. The cnfldmul
approach wouldn't show up as an additional use of a discouraged theorem but it would actually entrench ax-mulf
further.
I don't know how familiar you are with the ax-mulf
background but in a nutshell:
- Reduce ax-mulf usage 7 #4795 and previous pull requests in that series make a bunch of changes including the addition of
mpocnfldmul
- various comments on Clarify ax-mulf comment #4796
There are also some dissenting voices:
I'm not really trying to get involved in something which is still being worked out in set.mm - I am just trying to follow current set.mm practice as readily as can be done without a large amount of effort.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
OK, if this is still in discussion in set.mm, we can take its current state over to iset.mm.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yeah, that's pretty much the situation. Even if/when things settle down in set.mm, it is still work to bring it over to iset.mm (adjusting as needed for the differences between the two) - my philosophy for how to do that is "one step at a time".
There are a variety of prerequisites, including gsum theorems and a few others. But those are relatively modest advances over theorems and patterns already established in iset.mm.