Re: [DNSOP] [Ext] Re: New draft for consideration:

On Sun, Mar 24, 2019 at 4:49 AM Warren Kumari <warren@kumari.net> wrote:

>
>
> On Sun, Mar 24, 2019 at 11:46 AM Paul Hoffman <paul.hoffman@icann.org>
> wrote:
>
>> On Mar 24, 2019, at 11:18 AM, bert hubert <bert.hubert@powerdns.com>
>> wrote:
>> > It may be good to add a note that "DoH is the protocol as defined in
>> > [RFC8484]. The operation of this protocol by browser vendors and cloud
>> > providers is frequently also called 'DoH'. DoH-the-protocol is
>> > therefore frequently conflated with DoH being used to perform
>> > DNS lookups in a different fashion than configured by the network
>> settings
>> > (see DaT and DaO)."
>>
>> A much better outcome would be that people who are saying DoH when they
>> mean DaT or DaO would use the new terms. That is, this is a forward-looking
>> document because we're making up new terms.
>>
>
> <no hats>
> This is likely to be an annoying comment, but I don't really like the DaO
> "acronym", simply because I'm not sure how people will pronounce it -- I
> could see people mishearing "DaO" as "DoH", or the other way round --
> unfortunately I don't have a better suggestion. Is it just me who has this
> issue?
>

It probably isn't just you.
Here's a couple of suggestions to maybe canonicalize some of the terms, and
make them easier to distinguish/say:

DoTR (rather than RDoT): DNS over TLS, Recursive
DoTA (rather than ADoT): DNS over TLS, Authoritative. Or, some kind of
online game played by Millennials, presumably. :-)
DoN (rather than DaO): DNS on Nonstandard
DoS (rather than DaT): DNS on Standard. Risks confusion with Denial of
Service, if there is no provided context (but generally context will exist,
so...)

In anticipation of crazy ideas I might bring up, maybe we can agree on
compounding of lower-case "o" usage, with left-to-right meaning
left-encapsulated-in-right.
E,.g, DoTo53 would be "DNS over TLS, carried via some manner of encoding
within the payload of a Do53 message".

Brian

>
> </no hats>
> W
>
>
>
>>
>> > Secondly, I understand the technical need for the wording of the
>> definition
>> > of DaO.  But I had to read this all a few times before I understood that
>> > 'DaO' includes what I've referred to as DoC (DNS over Cloud). I think
>> > definitions should be easy to understand because otherwise they don't
>> > function.
>>
>> I fully agree; proposed changes to this wording are quite welcome. It's a
>> new term, after all.
>>
>> > I'm also not too hot for conflating "user consciously changes
>> > /etc/resolv.conf or equivalent" with "application makes the choice for
>> the
>> > user".
>>
>> The split here is more "someone changes from traditional without the user
>> knowing, when the user cares". If you have a better way to express that,
>> that would be great.
>>
>> > Perhaps we should talk about 'Per-application stubs'? Because this is
>> the
>> > nub.
>>
>> Maybe, but I'm hesitant to make the break that way because some
>> applications' stubs use the traditional resolver, others don't. I would be
>> hesitant to conflate those two.
>>
>> > I'm willing to write text once we have discussed this a bit.
>>
>> Thanks!
>>
>> --Paul Hoffman
>> _______________________________________________
>> DNSOP mailing list
>> DNSOP@ietf.org
>> https://www.ietf.org/mailman/listinfo/dnsop
>>
>
>
> --
> I don't think the execution is relevant when it was obviously a bad idea
> in the first place.
> This is like putting rabid weasels in your pants, and later expressing
> regret at having chosen those particular rabid weasels and that pair of
> pants.
>    ---maf
> _______________________________________________
> DNSOP mailing list
> DNSOP@ietf.org
> https://www.ietf.org/mailman/listinfo/dnsop
>