Re: [dnssd] Confirming consensus from DNSSD Privacy discussion in Bangkok

[Christopher Wood <[e-mail]>]

On Wed, Feb 27, 2019 at 12:18 PM David Schinazi
<[e-mail]> wrote:
>
> Thanks for writing this up Christian.
>
> I went through the minutes of our last in-person meeting in Bangkok and while we did not
> explicitly ask for consensus then, I got a feeling that the sense in the room was:
>
> 1) Is the server mode important for the scenarios that we are thinking about?
>     - No one in the room came out strongly in favor of server mode
>     - Server mode would add complexity
>
>     My personal recommendation: we declare server mode out of scope for now.
>
> 2) Do we care about compatibility with the existing DNS-SD formats?
>     - No one in the room came out strongly in favor of this
>     - It was mentioned it would be nice to have but not critical
>     - This would cause some security tradeoffs regarding hash size
>
>     My personal recommendation: we declare compatibility with existing formats out of scope for now.
>
> 3) Do we want private discovery to work as a system component, or as an app library? (a.k.a. per device or per app.)
>     - We had briefly discussed this at IETF 100 in Singapore following Stuart's discussion of requirements:
>         https://tools.ietf.org/html/draft-cheshire-dnssd-privacy-considerations-01#section-3
>     - I recall there was support in the room for having finer granularity such as per-app
>
>     My personal recommendation: we focus on per-app for now.
>
> Based on the above, as chair, I'm proposing we move forward with a reduced scope solution for now,
> that is without server mode, using binary formats, and runs as an app library.
> To clarify, this wouldn't mean we won't try to build a follow-up solution for those requirements later,
> this is just meant as framing to help us make progress towards a concrete proposal.
>
> Please share your opinions on this potential approach.

+1

This sounds reasonable to me!

Best,
Chris