Hi Alessandro,
please find my comment below.
Il 21/02/2022 18:25, Alessandro Vesely ha scritto:
Hi Mario,
On Mon 21/Feb/2022 11:34:14 +0100 Mario Loffredo wrote:
Il 21/02/2022 09:54, Alessandro Vesely ha scritto:
On Mon 21/Feb/2022 08:30:53 +0100 Mario Loffredo wrote:
Il 20/02/2022 13:17, Alessandro Vesely ha scritto:
On Wed 16/Feb/2022 15:54:01 +0100 Mario Loffredo wrote:
For what is worth, I would proceed as in the following:
1) If a contact with the abuse role exists:
1.a) If at least one pref parameter exists, I would return only
the most preferred email address;
1.b) If the pref parameter is missing, I would return all the
email addresses in order of appearance.
2) If a contact with the abuse role is missing, I would return no
abuse email address
Hope It could be helpful.
Yup, that's an applicable hint. I'd suggest to publish it as a
default profile in the next RFC. (I don't know how to get
profiles, let alone parsing them. Having a default profile allows
to make sense of APNIC's preferences.)
Don't think this is a matter of an RFC.
Why not?
Because, since the pref parameter can be added to many VCARD
properties that can be included in the RDAP response, what is stated
in https://datatracker.ietf.org/doc/html/rfc6350#section-5.3 seems
enough to me.
Special meaning of the pref parameter with regard to the
specification of the abuse mailbox should be decided by each registry
autonomously.
That said, it seems to me that the APNIC's response is not ambiguous
as it appears clear (at least to me) which is the preferred abuse
mailbox.
Yes, having seen that a pref parameter exists, its usage can be
inferred. I didn't know about it (and haven't yet fixed the program
which extracts it from the RDAP response.)
What I suggested, it can be valid as a general rule applicable to
every RDAP response provided that the server returns a contact with
the abuse role.
So the missing paragraph about "pref" could be filed as an erratum
to RFC7483?
What should such a missing paragraph say beyond what is stated in
section 5.3 of RFC6350 ?
Perhaps, the first paragraph of Appendix A.1 could have had an
additional sentence saying something like so:
For automated usage, it is recommended that, if preference is
specified,
only the most preferred element be used.
Sorry but, honestly, I don't see the need for that.
IMO, it would make more sense to recommend servers to specify a
preference when more instances of the same property are provided.
But specifying a preference is a possible option for the servers to
convey that a given value is considered the most suitable in a collection.
Therefore, from server side, I would add nothing. As I said in my
previous mails, additional requirements can be defined by each RDAP profile.
For example, since the proposal of provisioning the abuse mailbox seems
something accepted by every RIR according to what is described in
https://www.ripe.net/participate/policies/proposals/2019-04, RIRs can
agree on a
common way to include it in the RDAP response (maybe they're just doing
this, please members from RIRs have a voice if you think !)
From client side, clients can simply make their own decisions based on
the information returned by the server.
Anyway, I invite the other WG members to manifest their own opinion.
Best,
Mario
Best
Ale
--
Dr. Mario Loffredo
Technological Unit “Digital Innovation”
Institute of Informatics and Telematics (IIT)
National Research Council (CNR)
via G. Moruzzi 1, I-56124 PISA, Italy
Phone: +39.0503153497
Web: http://www.iit.cnr.it/mario.loffredo
_______________________________________________
regext mailing list
regext@ietf.org
https://www.ietf.org/mailman/listinfo/regext
