Hi Stuart,

I don't have a copy of the official standard, but from the documents on
the ISNI website I remember that there are name variations and 'public
identities' (as the lemma on Wikipedia also uses). I'm not sure where the
borderline is or who decides when different names are different identities.

If it were up to me: pseudonyms are definitely different public
identities, name changes after marriage probably not, name change after
gender change could mean a different public identity. Different public
identities get different ISNIs; the ISNI organisation says the ISNI system
can keep track of connected public identities.

Discussions about name variations or aliases are not new, of course. I
remember the discussions about 'aliases' vs 'Artist Name Variations' that
are/were happening on Discogs.com, e.g. 'is J Dilla an alias or a ANV of
Jay Dee?' It appears the users on Discogs finally went with aliases, but
VIAF put the names/identities together: http://viaf.org/viaf/32244000 -
and there is no ISNI (yet).

It gets more confusing when you look at Washington Irving who had several
pseudonyms: they are just listed under one ISNI. Maybe because he is dead,
or because all other databases already know and connected the pseudonyms
to the birth name? (I just sent a comment asking about the record at
http://isni.org/isni/0000000121370797 )


[Here goes the reference list…]

Hope this helps :)

Groeten van Ben

On 15-06-14 23:11, "Stuart Yeates" <[log in to unmask]> wrote:

>Could someone with access to the official text of ISO 27729:2012 tell me
>whether an ISNI is a name identifier or an entity identifier? That is,
>if someone changes their name (adopts a pseudonym, changes their name by
>to marriage, transitions gender, etc), should they be assigned a new
>identifier?
>
>If the answer is 'No' why is this called a 'name identifier'?
>
>Ideally someone with access to the official text would update the
>article at 
>https://en.wikipedia.org/wiki/International_Standard_Name_Identifier
>With a brief quote referenced to the standard with a page number.
>
>[The context of this is ORCID, which is being touted as an entity
>identifier, while not being clear on whether it's a name or entity
>identifier.]
>
>cheers
>stuart