# Pull Requests

#650
* There might be some issues with TLS implementation. Confirmed TLS encoder.
* Would need length variable - and added packets.
* There were no objections - decided to merge.

#651/#647
* Marta had questions about the reasons for the change. Concerned that this change would affect the proofs. But, in this case it should be okay because the context is already is included with the MAC key.
* Joel noted that this is a bit of different property with using MACs. It’s fine, but it is different.
* Joel provided background for why we did it.
* Richard is not wed to the PR though so the WG decided to close bot the issue and PR.
* But will add some text to explain why it’s there.

#658 Replace LeafNodeRef with leaf index
* Richard provided background.
* Brendan said that the extract layer of indirection doesn’t really help.
* Raphael & Knorad discussed it, but it seems fine. In security terms, indexes are not as strong because they only point to specific member for a given epoch. One the privacy side, the DS could do an external remove proposal without knowing the exact identity.
* Let’s merge it!

#675 More Clarity around validating proposals
* The big change is a change from a MUST to SHOULD.
* Brendan wondered if there is a starvation issue with federations.
* Add point about proposals only be valid in a list.
* Add some text around the SHOULD, the only reason you wouldn’t do it is that it’s not  valid.
* Looks good to merge after tweaking SHOULD text and review from Richard and Raphael.

#659 Remove new-member Add Proposal
* Looks like there is agreement to merge.
* Should back out SignaturePublicKey change.

#660
* Raphael thinks the MUST is very important.
* The current wording is a bit vague. Will revise.
* Suggestion is to close this out and work on 678.

#654
* Concern about whether AppAck is useful or not.
* Discussed motivations - this came from signal.
* Could we use capabilities to negotiate support. Would we not do this as an extension.
* There is some debate whether this required functionality vs optional.
* If you want some property from it, you need the application to make it …
* Merge after fixing conflicts.

# Issues

#640
* It’s bloat for proposal.
* Support for fixing this. Make everything full width.

#641
* Close

#645
* Upside is that you never have to worry about growing or shrinking the tree. Discrete grown or shrinking.
* @bifucation to take a stab at PR.

#661
* Discussed and agreed a good idea for random.

#667
* There is some uneasiness with loosening FS. I.e., this is removing sender FS.