You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Global facts formulated with obtains with one alternative and local facts formulated with obtain have the form (⋀x. P x ⟹ Q x ⟹ … ⟹ thesis) ⟹ thesis. Often, we prove such facts by showing ?thesis, which is thesis, using the local fact that, which is ⋀x. P x ⟹ Q x ⟹ … ⟹ thesis. However, the standard proof method, which is in particular invoked when proof comes with no initial proof method, turns such an eliminator-style goal into the subgoals (⋀x. P x ⟹ Q x ⟹ … ⟹ thesis) ⟹ P ?x, (⋀x. P x ⟹ Q x ⟹ … ⟹ thesis) ⟹ Q ?x, etc., which are weaker variants of P ?x, Q ?x, etc. Therefore, it is possible to not put - after proof and then just pick a concrete witness x and show P x, Q x, etc. individually, instead of putting - after proof and proving the whole eliminator-style fact, which involves referring to that. This alternative approach generally results in smoother proofs. We shall correspondingly change proofs of facts of the above-mentioned form wherever this improves the readability of the code.
The text was updated successfully, but these errors were encountered:
Global facts formulated with
obtains
with one alternative and local facts formulated withobtain
have the form(⋀x. P x ⟹ Q x ⟹ … ⟹ thesis) ⟹ thesis
. Often, we prove such facts by showing?thesis
, which isthesis
, using the local factthat
, which is⋀x. P x ⟹ Q x ⟹ … ⟹ thesis
. However, thestandard
proof method, which is in particular invoked whenproof
comes with no initial proof method, turns such an eliminator-style goal into the subgoals(⋀x. P x ⟹ Q x ⟹ … ⟹ thesis) ⟹ P ?x
,(⋀x. P x ⟹ Q x ⟹ … ⟹ thesis) ⟹ Q ?x
, etc., which are weaker variants ofP ?x
,Q ?x
, etc. Therefore, it is possible to not put-
afterproof
and then just pick a concrete witnessx
and showP x
,Q x
, etc. individually, instead of putting-
afterproof
and proving the whole eliminator-style fact, which involves referring tothat
. This alternative approach generally results in smoother proofs. We shall correspondingly change proofs of facts of the above-mentioned form wherever this improves the readability of the code.The text was updated successfully, but these errors were encountered: