[db-wg] [anti-abuse-wg] objection to RIPE policy proposal 2016-01

HI Niall

On 07/03/2016 17:48, Niall O'Reilly wrote:
>
>
> On 7 Mar 2016, at 10:43, Suresh Ramasubramanian wrote:
>
>>> On 07-Mar-2016, at 4:08 PM, denis <ripedenis at yahoo.co.uk> wrote:
>>>
>>> The "abuse-c:" IS standardised. It is well defined and documented as
>>> THE method of defining abuse contact details in the RIPE Database
>>> according to the policy. Historically, as I mentioned in other
>>> emails, there was "abuse-mailbox:" defined in 5 object types
>>
>> Sure - but as you point out nobody much seems to be implementing it so
>> far - or at least, very few organizations.
>>
>> So yes, I’d welcome abuse-c being implemented more widely.  I’m tired
>> of hunting up contact information from comment fields, in particular.
>
>    I can imagine.  I've just taken a look at this thread
> (https://www.ripe.net/ripe/mail/archives/db-wg/2004-April/thread.html#2678)
> about the plans to
>    move away from depending on comments by defining and introducing the
> "abuse-mailbox" property.
>
>    You tell us that, 12 years or so later, you're still depending on
> comments.

Your assessment is not accurate. Yes 12 years ago someone (I won't 
mention names) came up with the idea of adding "abuse-mailbox:" as a way 
to provide abuse contact details. However it was implemented in a way 
that was bound to fail. (Yes Gert I know you feel the same way about 
abuse-c now, but the situation is completely different.)

The "abuse-mailbox was added to 5 object types with not even any 
guidelines on how to use it. So of course users started adding it to the 
PERSON object, referenced in a MNTNER object, that maintains a ROLE 
object, that is referenced in an INETNUM object as admin-c. It seemed 
like the logical place to put it by the user at the time. But of course 
it was impossible to find it programatically with any confidence it was 
the right contact. We tried to write algorithms to find these 
references, but often got it wrong.

This was combined with the still popular usage of adding the abuse 
contact in remarks. It was this mess that prompted the introduction of 
abuse-c backed by a policy and made mandatory. So abuse-c does work and 
would work better if people did not try to evade their responsibilities.

We avoided having to clean up the tooth paste by laying another floor on 
top of it :)

cheers
denis

>
> On Thu May 6 12:39:12 CEST 2004,
> I wrote
> (https://www.ripe.net/ripe/mail/archives/db-wg/2004-May/002724.html):
>
>> [...] we (for some value of "we") have to devise a least-effort,
>> greatest-effect strategy for reaching "there" from "here".
>> I keep feeling we're still looking at tactics.
>
>    I'm sorry to say that I see proposal 2016-01 as more of the same, and
> a distraction
>    from the real work.  Administrative imposition of some measure as
> mandatory won't fill
>    the gaps in the data.  Neither will having a more consistent data model.
>
>    This toothpaste has been out of the tube since before the RIPE NCC
> came into existence.
>    By now, it's all over the floor. Cleaning it up can only be done by
> crawling around
>    between the cabinets and sanitary fittings with a spatula or damp
> cloth, not by admiring
>    the architect's proposals for how the bathroom might be remodelled.
>
>    At least for for 2028 (12 years further on), we can hope that
> pervasive adoption of
>    IPv6 will have made Legacy IPv4 resources irrelevant.
>
>    Best regards,
>    Niall