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
In basics.tex, line 1372 (proof of 2.6.4 in the book), the proof concludes with
Thus, it remains to show $x = (\proj1 x, \proj2x)$.
But this is the propositional uniqueness principle for product types, which, as we remarked above, follows from \cref{thm:path-prod}.
\end{proof}
where "thm:path-prod" is Theorem 2.6.2, and is being used to deduce the propositional uniqueness principle for product types.
Is it really necessary to invoke 2.6.2 to do this? Is the uniqueness principle for product types not a more elementary fact that can be proved just using induction on the product type, rather than 2.6.2 which also uses induction on identity types? Like in Coq below (definition and setup borrowed from https://github.com/HoTT/HoTT/blob/master/theories/Basics/Datatypes.v).
Record prod (A B : Type) := pair { fst : A ; snd : B}.
Arguments pair {A B} _ _.
Arguments fst {A B} _ / .
Arguments snd {A B} _ / .
Scheme prod_rect := Induction for prod Sort Type.
Lemma propositional_uniqueness A B (x: prod A B) : x = pair (fst x) (snd x).
Proof.
induction x.
simpl.
reflexivity.
Qed.
Thanks,
Tom
The text was updated successfully, but these errors were encountered:
In basics.tex, line 1372 (proof of 2.6.4 in the book), the proof concludes with
where "thm:path-prod" is Theorem 2.6.2, and is being used to deduce the propositional uniqueness principle for product types.
Is it really necessary to invoke 2.6.2 to do this? Is the uniqueness principle for product types not a more elementary fact that can be proved just using induction on the product type, rather than 2.6.2 which also uses induction on identity types? Like in Coq below (definition and setup borrowed from https://github.com/HoTT/HoTT/blob/master/theories/Basics/Datatypes.v).
Thanks,
Tom
The text was updated successfully, but these errors were encountered: