Re: [Gen-art] Genart last call review of draft-ietf-dnsop-attrleaf-fix-04

[Dave Crocker <[e-mail]>]

Hi.  Thanks for the detailed comments.

Some responses...

On 9/24/2018 3:00 AM, Francesca Palombini wrote:
>     The use of underscored node names is specific to each RRTYPE that is
>     being scoped.
> 
> As an non-expert in the area, I would have appreciate a ref to a document
> introducing RRTYPE.

The term is basic to DNS, with RFC 1035 cited in the first sentence of 
the Introduction:

      "Original uses of an underscore character as a domain node name
       [RFC1035] prefix, which creates a space for constrained
       interpretation of resource records, were specified without the
       benefit of an [IANA-reg] registry."

Once such a citation has been included, is a document expected to repeat 
the citation for every term used from it?  The implication is that 
someone reading this sort of document is not expected to have basic DNS 
familiarity?

However this did cause me to look for the first use of "RRTYPE" and I 
discovered that RFC 1035 has "RR TYPE" but not "RRTYPE". I'm not sure 
where first use without the space started.


>     This section provides a generic approach for changes to existing
>     specifications that define straightforward use of underscored node
>     names, when scoping the use of a "TXT" RRset.
> 
> Same for "TXT" RRset.

Same response.


> 
>     An effort has been made to locate existing drafts that
>     do this, register the global underscored names, and list them in this
>     document.
> 
> Since the effort has been done, I would have appreciated the full list here.

This is the 'fix' document, not the registry definition document.  The 
latter is cited in the first paragraph of this document's Introduction:

       "A registry has been now defined, and that document
        discusses the background for underscored domain name use
        [Attrleaf]."

That's where the list is provided.


>     An
>     effort has been made to locate existing drafts that do this, register
>     the global underscored names, and list them in this document.
> 
> Same as previous comment.

Same response.


>     An effort has been made to locate
>     existing drafts that do this and register the associated 'protocol'
>     names.
> 
> Same as previous.

Same response.


> 3.1. and 3.2. is the formatting of the updated sections (after "And is to be
> updated to the new text:") wanted? Why not use the same format as in 3.3., with
> OLD and NEW?

OK.


> 
>     +  Those registered by IANA in the "Service Name and Transport
>              Protocol Port Number Registry [RFC6335]"
> 
> Move the end quote after Registry.

ok.  Good catch.


>     +  Those listed in "Enumservice Registrations [RFC6117].
> 
> Missing end quote after Registrations.

ditto.


>     " Signaling Trust Anchor Knowledge in DNS Security Extensions
> 
> Remove the space after the quote.

ok.


>   John Levine, Bob Harold, Joel Jaeggli, Ondřej Sury and Paul
> 
> In Acknowledgements, one name is not encoded correctly.

I believe this as a bug in the xml2rfc generator used by the tools site, 
for text format, since the correct text is produced by an xml to html 
generator.


> 
>>From running the idnits tool (https://tools.ietf.org/tools/idnits/), several
> comments, warnings and one error were raised, which I snipped and pasted below
> as a summary:

What's most interesting here is that the document passed IDNits during 
submission!  (Apparently nits only works on txt documents and I only 
submitted an xml version; I'd guess the submission process does not 
general a txt version on the fly, for nits to inspect...)


>    -- The draft header indicates that this document updates RFC****, but the
>    abstract doesn't seem to mention this, which it should. (see
>    https://www.ietf.org/id-info/checklist) --> I see that the abstract generally
>    mentions "the existing specifications that use underscore naming", but I
>    think to make this correct, it should explicitely list them as well.

That actually makes no sense to me, since that would be fully redundant 
with the Updates header field that is already provided.


>    -- The document seems to lack a disclaimer for pre-RFC5378 work (See the
>    Legal Provisions document at https://trustee.ietf.org/license-info for more
>    information.)

Another victim of the long lag time.  I've updated the IPR template 
reference.  We'll see whether it's the right one...


>    == Unused Reference: several documents are included in the list of
>    references, but no explicit reference was found in the text --> if my
>    editorial comments are taken, they should fix this one.

This actually poses an interesting challenge.  The references are used 
in the Updates header field, but apparently IDNits does not look there.


>    ** Downref: Normative reference to an Informational RFC: RFC 7553

That document is a specification.  This document modifies it.  No matter 
it's standards track status, it is a normative reference to this document.


> 
>    -- Obsolete informational reference (is this intentional?): RFC 3921
>       (Obsoleted by RFC 6121)

Ack. Not intentional; just an error introduced by 12 years of lag time...

d/


-- 
Dave Crocker
Brandenburg InternetWorking
bbiw.net