Thanks for the comments on this item. We do realize that there might be a few cases where the possibility raised actually happens: That the original person referenced got deleted some time in the past and a new one with the same name was introduced into the database. However, contacting the holders of all the affected objects is really not feasible and I think also not practical (the order of magnitude is 10^5 objects that would get touched). We could actually send out all these emails and write a mail robot to process the responses. However, with the respondent being a human being, the chances are that a lot of mails would not be be able to be automatically processed. And a lot of people would not even reply. Also, many people not affected by the problem (being the correct reference) would get (unsolicited) emails. I would therefore not be able to extract a valid conclusion from the process. In the meantime, not carrying out the proposed modifications can only make the situation worse (as more person get registered in the database the chances of any two having the same name increase, and old person objects can get deleted). I think at this point there is not a perfect solution for this situation but I would rather deal with the few complaints about an object pointing to someone incorrect (and then act on them) rather than let the situation keep on deteriorating. One of the solutions to catch some of the misassignements (at least for inetnums) is to run a cross check of the database objects against the registry database. So, in spite of the concerns raised (valid as they are) I would still vote to go ahead on this one as proposed. What do people think? Regards, Joao Damas DB Group RIPE NCC