Inconsistent NIC handles

Dear all,

I have just come accross a problem with the RIPE database.
I am sorry to bother you if this has already been discussed.

What I noticed is an inconsistency problem between the different databases that
keep the same type of information (in this case _person_ objects).

Theoretically the NIC handle is supposed to uniquely identify the person.
The problem I noticed is that somebody (inadvertently) registered a person
with an Internic-type NIC handle (i.e. without the "-RIPE" suffix) in the
RIPE database. The result is an inconsistency between the two databases
(RIPE and Internic). One gets different answers, depending on whom one asks the
question.

I can think of two approaches to solve this problem:

- First is checking for the "-RIPE" suffix in the "nic-hdl" attribute when
updating a person object (whenever nic-hdl is not empty)

- Second is checking (whenever possible) for inconsistencies with the
apporpriate database (in our case one could check the person object in the
Internic database).

I think the first approach can be too restrictive (it would not allow for
redundant data between the databases, which otherwise can be useful for saving
bandwidth on frequent queries). This would nevertheless have the benefit of
preventing people from registering the same thing in several places (and forget
to update some of them eventually). Probably the badwidth saving can be better
addressed with the sites that keep the large databases 'mirroring' each other.

The second approach can be too complicated in case one would like to check more
than just the name of the person. There might be problems with the name as well
in case of several 'first' names for example.

Nevertheless I feel some basic check would be welcome...

Best regards,
Janos Zsako