Could CLAIM be added to the keywords for bibliographic cross-reference ?

[Glauco]

The command

HELP WRITE BIBLIOGRAPHY

doesn't list "claim" as a keyword (and in fact, it looks like it's not accepted by the markup verifier).

But books, papers and thesis often use such word, so it may be useful to add it to the next metamath version.

Thanks

Glauco

[Norman Megill]

That's a good idea.  Give me about a week to release a new version with this keyword.  If you think of any others that would be useful let me know.

Norm

[Glauco]

I'm not sure if "case" is a good candidate: in most cases it would be something like "case 2 of theorem 3.1 ..." (that's already covered) but I guess it could also be used outside of a theorem proof. I don't know if we also want such kind of reference.

Glauco

[David A. Wheeler]

On Sun, 5 Apr 2020 08:10:56 -0700 (PDT), Norman Megill <n...@alum.mit.edu> wrote:
> That's a good idea. Give me about a week to release a new version with
> this keyword. If you think of any others that would be useful let me know.

I agree. Let's make sure we update the "conventions" section to match.

Norm: Please let us know if you add anything else, some tools (like my vim extension)
will need to tweaked.

--- David A. Wheeler

[Alexander van der Vekens]

"Claim" would be a very helpful keyword - it is used by Huneke, and I will revise the references in my comments on the friendship theorem accordingly if it is available.

I have no additional suggestions.

Alexander

[Norman Megill]

"Claim" has been added in metamath.exe version 0.182.  I didn't add "Case" because it seemed to me that it would almost always be accompanied by another keyword, and in "Theorem 1.1 Case 2" it would cause the theorem number to be omitted.  But I'm willing to reconsider that.  (The algorithm scans backwards from the [author] tag until a keyword is found.)

On Sunday, April 5, 2020 at 2:09:21 PM UTC-4, David A. Wheeler wrote:

On Sun, 5 Apr 2020 08:10:56 -0700 (PDT), Norman Megill wrote:
> That's a good idea.  Give me about a week to release a new version with
> this keyword.  If you think of any others that would be useful let me know.

I agree. Let's make sure we update the "conventions" section to match.

The current list of keywords is always given by "help write bibliography".  It might be better if the conventions reference the help command rather than duplicating the list, otherwise there is a risk of it getting out of sync.

I am also wondering if it might be better to specify the list in the $t comment (the one with the htmldefs) in set.mm rather than having it hard-coded in metamath.exe.  Any opinions?
 

Norm: Please let us know if you add anything else, some tools (like my vim extension)
will need to tweaked.

How would this be affected if we put the list into the $t comment?

Norm

[Thierry Arnoux]

For « Case », I have been using « Case 1 of Theorem 10 in [Reference] p. 100 ». This fits the current format enough to be recognized as a valid bibliographic reference, but still includes the “Case” information.

I don’t think cases are often numbered independently of any other statement.
BR,
_
Thierry