Re: [MLS] Erik Kline's No Objection on draft-ietf-mls-protocol-17: (with COMMENT)

[Richard Barnes <[e-mail]>]

Hi Erik,

Thanks for the review.  A couple responses inline below...

On Tue, Jan 31, 2023 at 2:25 AM Erik Kline via Datatracker <[e-mail]>
wrote:

> ## Comments
>
> * Thank you very much for all of the explicitly worked examples.
>

They were helpful in the WG discussions as well :)


### S7.3
>
> * What is the reason for MUST vs RECOMMENDED difference in the two
> different
>   circumstances when validating the `lifetime` field?
>

The reason is asynchronicity.  The protocol anticipates that there may be
some time between when a message is sent and when it is received, during
which time the KeyPackage might have expired.  So the WG felt that having a
hard requirement on the receive side would cause undue complexity in light
of this requirement.



> * Consider adding a platitude about communicating nodes SHOULD use some
>   method of time synchronization...
>

Good idea, thanks.


> ## Nits
>
> ### S3.2
>
> * "top of these basic mechanism" -> "top of these basic mechanisms"
>   (or maybe "top of this basic mechanism")
>
> ### S6.3.1
>
> * "in an PrivateContentTBE structure" -> "in a PrivateContentTBE structure"
>

Can you tell we changed the names of these structs late in the process?  :)

Best,
--Richard