On Wed, Apr 11, 2018, at 22:24, internet-dra...@ietf.org wrote:

> A diff from the previous version is available at:
> https://www.ietf.org/rfcdiff?url2=draft-ietf-regext-epp-fees-11

I see that this includes some of my latest comments
(but not all of them, for example required attributes still
do not have use="required" in the schema, and I have not seen
an explanation about that), so thanks for taking them into account.

However I still think there is a big problem for the "standard" attribute.

For me even this new version does not define it really.

Where it appears it says now:
 The <fee:command> element(s) contain(s) a "name" attribute (see
   Section 3.1), an OPTIONAL "standard" attribute with a default value
   of false (0) (see section 3.7), an OPTIONAL "phase" attribute, and an
   OPTIONAL "subphase" attribute (see Section 3.8).  The <fee:command>
   element(s) MAY have the following child elements:

so it references the section 3.7, which is new.

But the problem is that this section only speaks about classes
and the <fee:class> element in responses. There is again nothing there about
the "standard" attribute of a <fee:command> in requests.

There is specifically a need to explain things in relation to last line
of this section. What happens/What does it mean/Is it allowed or not
when fee:class=standard but standard=false
or when fee:class=premium but standard=true?

Especially since both elements are optional.

I believe that this lack of explanations can cause major interoperability
problems, even more so because this attribute was added very late in the draft 

Maybe people in favor (I am not) of this attribute
should chime in and provide some useful text to add in the specification.

I do not think it can pass LC without this fixed.

  Patrick Mevzek

regext mailing list

Reply via email to