Herb Sutter

Editors call: What is magical depends on the domain, so it is not possible to define what is magical independently of context. An enforcement rule has to take that into...

Editors call: We're in a halfway state where we had hoped Standard C++ would add contracts but the standard has not yet done so. We definitely do want `gsl::span` bounds...

Editors' call: [GSL.assert](http://isocpp.github.io/CppCoreGuidelines/CppCoreGuidelines#gslassert-assertions) currently requires `Expects` to terminate the program if the condition is false. This is intended to mean `std::terminate` which permits a `terminate_handler` to do things like log...

Clarification: `Expects` should `terminate`, not `abort`, just to be clear.

Editors' call: `gsl::joining_thread` will be superseded by C++20 `std::jthread`. For C++17, people should use `gsl::joining_thread` and we won't rename it now to avoid disturbing existing users since it's going away...

Editors' call: Add a note to `joining_thread` that points to C++20 `std::jthread`.

Editors call: Thanks, this sounds good and we will follow up to add a rule.

Editors call: ES.87 should really be saying to avoid redundant comparisons only for pointer types (against not-null/null) and bool types (against true/false). Those are the motivating cases, and then in...

Editors call: While intrusively requiring `shared_ptr` can be a solution in some cases, we don't think it's something we should say as general advice because that means aliasing and having...

Editors call: Would you please write up some pros and cons?