[db-wg] Locking unmaintained PERSON and ROLE objects in the RIPE Database

Dear Denis,

Thank you for your request. Please allow me some time to look into this
matter. I will provide you and the list with my comment next week.

Kind regards,

Athina Fragkouli
Head of Legal
RIPE NCC



On 14/04/16 13:31, denis wrote:
> Hi Piotr
> 
> On 14/04/2016 08:41, Piotr Strzyzewski wrote:
>> On Thu, Apr 14, 2016 at 01:07:01AM +0200, denis wrote:
>>
>> Denis
>>
>>> You really are barking up the wrong tree here.
>>
>> Nice try, but you misinterpret my intentions.
> 
> This is a very condescending remark.
> 
>>
>>> On 14/04/2016 00:31, Piotr Strzyzewski wrote:
>>>> On Thu, Apr 14, 2016 at 12:00:45AM +0200, denis wrote:
>>>>
>>>> Denis
>>>>
>>>>>> Taking your arguments stated above, I once again (this time more
>>>>>> clear)
>>>>>> say that rolling back the change to allow changes to the PERSON
>>>>>> object
>>>>>> name would _not_ have fixed the problem.
>>>>>
>>>>> Maybe not, but it would have been a harmless change. Your action has
>>>>
>>>> This change has been discussed and expected in this community at least
>>>> from the year 2000.
>>>
>>> For a start this database with this (broken) data model did not exist in
>>> 2000. It was released to production in April 2001.
>>
>> Which do not proof that there was no desire/discussion/need for change
>> of the person name even earlier:
>>
>> https://www.ripe.net/ripe/mail/archives/db-wg/2001-January/001512.html
>>
>>> And has been checked by You in 2014 with RIPE NCC's
>>>> legal team. I haven't seen any technical, legal, procedural (nor any
>>>> other) objections raised by You under issue 221
>>>> (https://github.com/RIPE-NCC/whois/issues/221).
>>>> What has changed during last 1.5 year?
>>>
>>> Issue 221 was about changing the PERSON object name. It was agreed by
>>> the
>>> RIPE NCC legal team that they saw no problems at that time. What has
>>> changed is that you have now identified a hijacking issue. Rolling back
>>> that change would have prevented 'easy' hijacking by simply changing the
>>> name of an unmaintained PERSON object to your name. By locking the
>>> objects
>>> they now have to 'get' ID in that name. As you pointed out that seems
>>> to be
>>> easy, although I have no idea how to do something like that.
>>
>> The hijacking issue was there for years. I'm just surprised that it was
>> not raised by you during the discussion of issue 221.
> 
> Again this is another condescending remark. You imply that I should know
> everything at all times. If you knew about this hijacking issue that you
> say has been around for years, then as a co-chair of this working group
> why did you not mention it at the time?
> 
>>
>>> You seem to be determined to 'prove' I am wrong suggesting rolling
>>> back the
>>> name change would fix this issue. But you don't seem to accept that the
>>> action you have taken has also NOT fixed the problem but caused many
>>> more
>>> serious problems.
>>
>> Please refrain from suggesting that I have done something.
> 
> The whole tone of this thread is one of attack and aggression towards
> me. You are not discussing the issues but you are discussing me. You are
> relentlessly pursuing my comments and looking back in history to prove
> me wrong, instead of trying to move forward. You are doing exactly what
> you accuse me of.
> 
>>
>> Moreover, keep saying the mantra about causing many more serious
>> problems is neither the proof of this thesis nor the solution to
>> anything. If you know/see something which could seriously improve the
>> quality of the data, security model, business rules, etc, just bring it
>> on the table.
> 
> Sorry but you are missing the point here. It is not a thesis and I have
> stated the serious problem. I was part of the Data Protection Task Force
> many years ago. I spent a lot of time with the RIPE NCC's lawyers
> discussing the data protection issues around the RIPE Database.
> 
> The action you have taken HAS already created a data protection issue.
> There are now about a million personal data sets in the database that
> the data subjects cannot change. The RIPE NCC has stated it will not
> unlock any of these objects as they cannot be sure who they refer to.
> The RIPE NCC, as Data Controller of this database has a million personal
> data sets in their database that they cannot assert contain accurate
> data and they have prevented the data subjects from updating their
> personal data. The RIPE NCC also, even as Data Controller, cannot do
> anything to fix or replace these personal data sets.
> 
> Now I will bring some suggestions to the table to fix this issue. First
> I would like the RIPE NCC legal team to review this situation and
> publish to this list their considerations. Secondly I suggest the RIPE
> NCC unlocks these objects, as it makes no difference to the hijacking
> situation that you say has been around for years anyway. Third I suggest
> the RIPE NCC aggressively pursues the members who reference these
> unmaintained personal objects and pushes them to either maintain them or
> replace them.
> 
> Finally I would say that for an issue that the DB WG co-chairs have
> known to have been around for many years, I don't see why it needed a
> secret, back room discussion and a sudden announcement that the RIPE NCC
> has locked a million objects without any community discussion.
> 
> As I am not receiving the emails in this discussion from the DB WG
> mailing list, I think you may have already blocked them. So I have cc'd
> Athina directly for a legal comment.
> 
> cheers
> denis
> 
>>
>> Piotr
>>