Jeff Burdges

@ticki's matrix is completely wrong, maybe even if you check everything under the diagonal. If you've no problem with any common licenses, then you only need to worry about mutual...

Appears @agl has basically approved https://github.com/agl/pond/pull/148 so that one should get in eventually. You might want to submit any differences from https://github.com/agl/pond/pull/148 as a pull based on top of https://github.com/agl/pond/pull/148.

I'll look into this if nobody else is doing so. If however anyone needs it now they could presumably use Lil' Debi.

A priori, there should only be one revocation per contact, but there is a serious issue around double sending messages : https://github.com/agl/pond/issues/135#issuecomment-82386725 Are these all the same revocation? Or multiple...

Alternatively we might want an option to pause PANDA operation for all or some users.

Also, when did you do that @david415 ? `git log .../poly1305_arm.s` says arm assembly was added 6 May 2015. If you built it before that, then maybe it just worked....

Appears `dup3` could work. Interesting Android does not have this issue : https://github.com/agl/pond/issues/119

Is this going to be common? If not, you can do it with a trait hierarchy already, right? We can require that `x` be disjoint from `y` and `z` like...

I see. It appears the `struct` version suffices for creating multiple "disjointness families" of methods. I suppose expressing overlap helps, but maybe not soo much, and maybe duplicate files could...

I suppose the `struct` version can be quite expressive if we identify duplicate fields between `struct` field declarations together : ``` trait Foo { struct { a: T, b: T...