| 20:08 | <mhofman> | bakkot: I was wondering if there was any update to the status of https://github.com/tc39/proposal-freeze-prototype. We have a defense in depth use case that could probably use this feature. |
| 20:18 | <bakkot> | mhofman: no updates. the place that i left it was, I have not figured a good way to represent the three states "fully extensible, extensible by new properties but not by setting prototype, not extensible" in a way which plays well with the MOP/proxy traps, like isExtensible. I haven't found anything that I liked and there may not be a way to do it. |
| 20:39 | <mhofman> | I'm not fully up to date on the issues. Is there any fundamental problem besides adding new proxy traps? |
| 22:09 | <bakkot> | the problem is that isExtensible is an existing proxy trap, which returns a boolean, and which is used for both "can add properties" and "can change prototype". so what should IsExtensible mean in a world where those two things are not always identical? |
| 22:53 | <rbuckton> | Huh... We define ScriptEvaluation in the spec but don't actually reference it from anywhere in the spec. |
| 23:16 | <ljharb> | the problem is that |
| 23:29 | <bakkot> | rbuckton: yeah, that's the entry point, which is left to hosts; see https://html.spec.whatwg.org/multipage/webappapis.html#calling-scripts |
| 23:29 | <bakkot> | we should probably call this out though |
| 23:29 | <rbuckton> | we should probably call this out though |