When an existing theorem directly matches a goal, which of the following is preferred?
have "..." using my_theorem .
have "..." by (rule my_theorem)
I prefer the latter, but that's mostly personal preference.
I try to skip those steps...
note H = my_theorem[of ...]
would also work
The rule version has the advantage to have an implicit standard applied, which can be useful to discharge assumptions
There's also the "fact" method for that sort of thing. Can take a theorem as a parameter, or can figure it out automatically if there's a suitable fact in the context
Last updated: Oct 12 2025 at 20:20 UTC