Sorry for the new thread. I seem to have lost the publication notice
to anchor this message to.

I have a few mostly nitpicky comments. I probably wouldn't send them
except that this one sticks out a bit:

2.1 says,

    Both the "spf" and "dkim" results may optionally include a
   "human_readable" field

however the Appendis A schema has "human_results"


Now that I'm here.. .other picky remarks, in random order of pettiness--

The intro says

    DMARC Aggregate Reports MUST contain two primary sections
    ("metadata" & "data" below) ...

I find it interesting that these named sections are never really
referred to by those names again. i.e. those section names aren't called
out. Most other references are made with other terms, not section
names. The meaning of "metadata" and "data" can, of course, be
inferred.

----

Relatedly, the terms "element" and "field" seem to be used
interchangeably (not to mention "sub-element"). I think I'd like it a
bit more if xml elements were called one thing (like "element") and not
"field," especially where "field" sometimes has other meaning like in
"header field." There's no bright line between those terms, and I must
say I have not really been completely confused by this, so it's probably
a trivial remark. Still.

----

This kind of nomenclature issue comes into play a bit in references to
"report-id" as a field when there is a report_id element in the scheme
(and indeed is called "report-id" in a comment in the schema). And I did
find this a little confusing.(*)

----

2.1.2 says 

    ... there is a preference as to which signatures are included.

and then there's a numbered list of signatures followed by

    A report SHOULD contain no more than 100 signatures for a given
   "row", in decreasing priority.

two things about "priority" vs "preference" (did I mention the
pickiness)?:

The list says there is a preference, but it doesn't really say what the
preference is, nor what the numbers mean in relation to a priority or a
decreasing in priority. Maybe add "in the following order" in the
introduction to the list. The summary line could say something like

    A report SHOULD contain no more than 100 most preferential
    signatures for a given "row."

-mm- 

(*)even having read multiple interations of this document
multiple times.

_______________________________________________
dmarc mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to