Mark S. Miller

I am confused about the separation between this PR and https://github.com/endojs/endo/pull/1986 , Both contain all three files. I would have expected one (that one) to define the benchmarking framework, and...

Now that #1986 is merged, if it is ok with the other reviewers, I'd prefer that this PR be rebased on master if it is not already based on a...

> #1986 is already incorporated, but it would be a good idea to land #1991 and then rebase this PR (which already shares its packages/base64/test/test-main.js) on top of the resulting...

>  Oops, sorry! I had no idea. Thanks for fixing.

> WeakMap can't be truly polyfilled. That is correct but it amazing how close we can come: https://github.com/google/caja/blob/master/src/com/google/caja/ses/WeakMap.js Fortunately, we rarely need to polyfill it anymore. https://kangax.github.io/compat-table/es6/ (see WeakMap row)

Indeed. Our WeakMap polyfill is open source. You and they are welcome to it. See also @kriskowal 's port and refactoring at https://github.com/drses/weak-map . This if the more useful one...

Third seems better to me, but second is acceptable. I dislike the first because it would cause us to invest more effort into engineering new features for sloppy mode. The...

Or to take option 3 of @ljharb 's list at https://github.com/tc39/proposal-decorators/issues/204#issuecomment-453763265 : * similar to non-simple parameter lists, inline decorator functions would be strict, but would throw if the thing...

I do not subscribe to the so-called 1js policy and I do not agree with it. There is no committee-wide consensus on this policy. I have consistently resisted adapting new...

Yes, early error. Exactly as for non-simple parameter lists.