Hi Denis, Thanks for your feedback, I will try to answer some of the points you raised.
On 29 Sep 2025, at 07:28, denis walker <ripedenis@gmail.com> wrote:
Hi David
Thanks for the update. I think there are several misunderstandings here. Let me offer an alternative view from someone who has spent decades studying and debating these documents :)
On Sat, 20 Sept 2025 at 11:46, David Tatlisu <dbwg@tatlisu.eu> wrote: Hi Everybody,
Depending on the interpretation of ripe-508, email addresses might already be a mandatory but unenforced part of PERSON objects.
First of all ripe-508 is a RIPE document but it is not a RIPE policy. So anything it says is for guidance only.
In the context of Internet number resources, the policy states:
'document' states
In case the Status is either "Allocated" or "Legacy", the following information is also mandatory: […] Contact information for matters of an administrative nature, and for matters of a technical nature. This information consists of an email address and a telephone number
OK so we are talking about allocations here. It gets a bit confusing here because of the history and mistakes in documentation and the communities 'lost' understanding of the whole subject of contacts (the wider picture I referred to).
When I completely re-wrote the RIPE Database documentation in about 2014 I added the attribute property of 'required' [1]. We had always had this property but never defined it. This is where an 'optional' attribute is 'required' to be present in some cases because of (software) business rules.
This is still the case. An 'optional' attribute can become 'required' due to some business rules. If an attribute is 'required' then it's always required regardless of the business rules.
For the INETNUM object "org:" is a 'required' attribute. (I am sure I did change this in the object templates. But in the documentation of the INETNUM object [2] and in the 'whois -t' output it has gone back to 'optional'.)
The inetnum 'org:' attribute must be listed as 'optional' in the object template because it's only 'required' in some cases (due to the business rules). The business rules for "org:" are listed in the documentation: https://docs.db.ripe.net/Database-Support/Database-Business-Rules "On inet(6)num objects for allocations or End User assignments made by the RIPE NCC, the "org:" attribute is mandatory." In the "Description of Attributes Specific to the INET6NUM Object" section, I think that "org:" is incorrectly described as required, i.e. ""org:" – single valued to make sure that only one organisation is responsible for this resource. This is a required attribute. In some cases, there are business rules to ensure that it is present. If the inet6num object is (jointly) maintained by the RIPE NCC then the object must have an “org:” attribute." Also similar text in the "Description of Attributes Specific to the INETNUM Object" section. I think we should correct this to state: "This is an optional attribute. In some cases, there are business rules to ensure that it is present."
... (Just as an aside, I noticed that the RIPE Database documentation's description of the status 'AGGREGATED-BY-LIR' [2] is completely incompatible with the description in the address policy [6].)
The documentation text for INETNUM AGGREGATED-BY-LIR was mostly copied from the existing INET6NUM description, because the policy proposal 2023-04 mostly used the same wording as in the IPv6 policy. I reviewed the documentation description for INETNUM AGGREGATED-BY-LIR and noticed we don't state that the "assignment-size:" is optional for INETNUM, we will clarify this. Also we don't allow two levels of "AGGREGATED-BY-LIR" in inetnum, but only one. We will correct this. Please let me know how we can otherwise improve the text (e.g. should we quote the policy text directly?). Regards Ed Shryane RIPE NCC