Re: [DNSOP] [apps-discuss] Draft of interest in DNSOP: draft-ietf-dnsop-attrleaf

[Dave Crocker <[e-mail]>]

Hi, folks,

I've been reminded that the attrleaf draft is still pending.  I've 
re-submitted the -00 draft as -01, just to restart the timer on the 
document.

On reviewing the discussion history, I see some items for the list that 
I believe weren't resolved...


On 8/28/2016 4:04 PM, Dave Crocker wrote:
>>> For URI records RFC 7553 says they're either named the same as SRV
>>> records, or they use enumservice names from the Enumservice
>>
>> Declaring a namespace as the union of two, independently-maintained
>> registries is a very efficient way to encourage -- actually in
>> theoretical terms, it guarantees -- collisions.
>
> Patrik's and John 's postings notwithstanding, I'm still concerned about
> the proposed way of handling this, namely to rely on IANA to do a manual
> check of the two registries the URI RR might call on.  First, it does
> not seem reasonable to me to impose that burden on the IANA staff and
> second a manual process like that is almost certain to produce errors.

This needs discussion and resolution by the working group.

Again, my concern is with handing IANA an on-going task -- making a 
check every time they update the Attrleaf table -- that requires their 
being perfect at detecting collisions with one or more other tables.

If the wg decides this is acceptable, I'd like to document the basis for 
the decision.


> On 8/3/2016 9:05 PM, John R Levine wrote:
>> I suppose, but since the two registries exist and the URI RFC says to
>> use both of them as _name, that horse has left the barn.
>
> I think it left the theoretical barn, but not the practical one.  We've
> been told that there is not yet any uptake in the URI RR.  That makes it
> plausible to modify its spec to eliminate this problem.

Can we get a working group assessment of the real-world uptake of the 
URI RR?


> Really, the burden of trying to have on-going coordination for the URI
> RR, between two different registries is worth finding a way to avoid.
> The problem is that I am not sure what to suggest that will work for URI
> usage.
>
> Suggestions?

Suggestions are still solicited.

d/

-- 
Dave Crocker
Brandenburg InternetWorking
bbiw.net