Re: [scim] Name Internationalization

Thanks, Bert. So to clarify, in the SCIM/vCard mapping proposal, multiple
name components will be *comma-separated* in the corresponding SCIM
singular attribute? In our case, we are initially considering *space-delimiting
*the components such that this would effectively serve as a "formatted"
familyName/givenName/etc. and could potentially accommodate contributing
systems that only store single-valued name attributes as well.

With a comma-delimited value, all SCIM consumers must be aware of this
special format when dealing with the attribute. With a space-delimited
value, consumers may not be able to retrieve the individual values, but can
handle the name consistently without attempting to parse and regardless of
the origin of the data (i.e. whether it came from a system that only stores
a single value or multiple values).

I also noticed that the SCIM/vCard draft includes this as an open issue,
proposing that it may be more appropriate to update the SCIM schema to
handle multiple values. Is this a topic that's currently being discussed?

We are currently adopting the SCIM 1.1 specification, but are trying to
consider potential changes for SCIM 2.0 in our implementation.

On Mon, Mar 25, 2013 at 12:41 AM, Bert Greevenbosch <
Bert.Greevenbosch@huawei.com> wrote:

>  I stumbled across the same issue when doing v02 of the SCIM/vCard
> mapping draft. See section 6.****
>
> http://datatracker.ietf.org/doc/draft-greevenbosch-scim-vcard-mapping/****
>
> ** **
>
> Indeed in vCard multiple surnames, given names and additional names are
> possible.****
>
> ** **
>
> Best regards,****
>
> Bert****
>
> ** **
>
> *From:* scim-bounces@ietf.org [mailto:scim-bounces@ietf.org] *On Behalf
> Of *Marc Blanchet
> *Sent:* 2013年3月25日 6:35
> *To:* Shelley
> *Cc:* scim@ietf.org
> *Subject:* Re: [scim] Name Internationalization****
>
> ** **
>
> might want to look at vCard  (RFC6350).****
>
> ** **
>
> Marc.****
>
> ** **
>
> Le 2013-03-22 à 22:59, Shelley <randomshelley@gmail.com> a écrit :****
>
>
>
> ****
>
> As a SCIM service provider, we are trying to determine the best approach
> for accepting and managing names that may be federated from global
> consumers.
>
> Were there any considerations made in the SCIM core schema for using
> multi-valued attributes for the individual name components? The use of the
> "familyName" and "givenName" as opposed to "firstName"/"lastName" helps
> minimize a western/US-centric approach, but using three individual,
> singular attributes for these name components still hints at a particular
> name format that may not be global.
>
> For example, in many countries, individuals have a given name and two last
> names (rather than first name, middle name, last name). Does SCIM provide
> any recommendations for how to represent this using the existing name
> components? For example, are consumers expected to consolidate all last
> names into the single familyName attribute to accommodate this scenario?
> Likewise, there are many other cases [1,2] that don't quite cleanly fit
> into these singular name components.
>
> The "formattedName", "displayName", and "nickName" attributes help to
> mitigate some concerns around formatting names and addressing users, but
> we're still trying to iron out how to accept identity data from varying
> contributing sources as well as enable consumers to obtain the discrete
> name components.
>
> [1] http://www.w3.org/International/questions/qa-personal-names
> [2] http://en.wikipedia.org/wiki/Personal_name
> _______________________________________________
> scim mailing list
> scim@ietf.org
> https://www.ietf.org/mailman/listinfo/scim****
>
> ** **
>