diff --git a/README.txt b/README.txt index 9829c11cf2168d5a06bc970c45244146acfdda4f..eb476e8cafc38cc8b7c1b00495eee6101328d9de 100644 --- a/README.txt +++ b/README.txt @@ -19,9 +19,30 @@ CONTENTS Our artifact is a Coq formalization of the model of our Iris logic, together with a proof of adequacy (establishing that the model is - faithful wrt the operational semantics), and a proof of soundness of + faithful wrt the operational semantics) and a proof of soundness of the primitive rules of the logic wrt the model. + NOTE: We have just mechanized the *soundness* of the *primitive* + rules of Iris in Coq. We have not mechanized the proofs of derived + rules (i.e. those derivable from the primitive rules), nor have we + mechanized the case study or other examples that are proven within + the logic. Proof outlines for the latter are given in the appendix + that accompanied the POPL submission, and will be fleshed out even + further for the final version of the appendix. + + The reason we focused on the primitive rules is that those are the + rules whose soundness is proven by direct appeal to the semantic + model of Iris. For space reasons, we did not want to present the + semantic model of Iris in any detail in the paper, but we still + wanted to give the reader confidence in the results of the paper. + With our Coq mechanization in hand, the reader can safely ignore the + semantic model and instead focus on how to *use* the primitive rules + of the logic (to derive more sophisticated rules or prove + interesting examples). + + Mechanizing Iris proofs is a very interesting and important + direction for future work, but it is beyond the scope of the paper. + The folder is organized as follows: @@ -90,3 +111,4 @@ OVERVIEW OF LEMMAS Fork htFork The main adequacy result is expressed by Theorem soundness_obs. +