Re: [dnsdir] Dnsdir telechat review of draft-ietf-homenet-front-end-naming-delegation-18

[Michael Richardson <[e-mail]>]

    as> Reviewer: Anthony Somerset
    as> Review result: Ready with Nits

    as> Section 3.2 = "SHOULD remain pointing at the cloud provider's server IP address
    as> - which in many cases will be an anycast addresses."

    as> I don't believe its correct to include this assumption about anycast addresses
    as> and is largely irrelevant to the content of the draft so i don't believe there
    as> is value in keeping the text after the hyphen

I see your point.
I feel that there is some relevance to pointing this out.

One of important aspect of reminding people about this is to indicate that it
should be surprising if queries to these addresses actually return different
time views of the zone.  It can take some minutes for all anycast hosts to
update.

A second important aspect is that the address that queries go to is not,
because of anycast, the same as the place where the updates go.

I don't feel strongly about this, I just think that we wrote this down for a reason.

    > The intro is very long and talks about things that don't get explained until
    > much later in document and could cause some confusion, it may be better to make
    > the intro more concise and move some of these aspects into the relevant
    > sections.

It grew as a result of reviews.
you are saying we overshot, sure.

    > Section 1.2 - to me this would flow better if it was its own section after the
    > solution is explained

okay.


--
Michael Richardson <[e-mail]>   . o O ( IPv6 IøT consulting )
           Sandelman Software Works Inc, Ottawa and Worldwide