Re: [dns-privacy] Operating System API support for DNS security policy

[Brian Haberman <[e-mail]>]

Hi Iain,

On 8/21/19 4:28 AM, Iain Sharp wrote:
> Yes, I would be interested if someone can explain the relationship between Posix and IETF in this regard, and why the information appears in two places. It is something that is unclear to me.
> 

It is a bit of a convoluted relationship. Generally, POSIX deals with
defining the APIs for UNIX OSes. Part of that involves defining the APIs
related to networking (e.g., socket(), getaddrinfo(), ...).

The IETF has published a number of "API" RFCs that describe enhancements
that the IETF community would like to see in the POSIX suite of APIs.
Once the IETF has agreed to those enhancements and published an RFC, the
advocates for those APIs then engage with the OpenGroup to try and get
them to take up the standardization of those new/updated APIs. In some
cases, they do and in others, they don't. The advanced socket API for
IPv6 is a good example of the latter.

I view any API defined by the IETF as an "informational" request to the
OpenGroup to standardize an API with the documented capabilities. What I
don't understand is why the IETF does not have a liaison relationship
with the OpenGroup to facilitate the development of the appropriate APIs
for the technologies defined in the IETF. That is a question better
suited for someone on the IAB.

Not sure if that clarifies things or not...

Regards,
Brian