Re: [urn] DOI URN namespace registration

[John Levine <[e-mail]>]

In article <[e-mail]> you write:
>"Hakala, Juha E" <[e-mail]> writes:
>> [...] I have been asked to write a namespace registration request for
>> DOI, and that may have a major impact on both URN and DOI
>> systems. That request needs to be processed promptly.
>
>I was going to ask, As an opening question has consideration been given
>to registering DOI as a URI scheme?  My memory is that DOIs are written
>in a format that is compatible with URI syntax.

Well, sort of. DOIs are handles, so the actual format of a DOI is
10.<number>/<LocalName> where the LocalName is arbitrary UTF-8 and is
in practice interpreted by the DOI registration agency that registered
the DOI. Since the Handle protocol never went anywhere, DOIs are now
usually written as URLs using the DOI Foundation's lookup engine:

	https://dx.doi.org/10.17487/RFC3651
	https://doi.org/10.17487/RFC3651

(The LocalName is an opaque string; in retrospect it was a mistake for
me to assign DOIs to RFCs that look like they mean something.)

Given that DOIs are usually resolved by looking up URLs, it is clear
that in practice they are compatible with URLs because they *are*
URLs. I wouldn't pay too much attention to the Wikipedia article since
any sentence that uses both "Unicode" and "case insensitive" without a
negation in between is by definition wrong.

Also see this, and the following two sections which explain
how to display a DOI as a URI and as a URN:

https://www.doi.org/doi_handbook/2_Numbering.html#2.6.1

They say the DOI name 10.123/456 can be written in the form https://doi.org/urn:doi:10.123:456.

Beyond the case sensitivity thing which I think is a red herring, I
would ask why they're asking for a DOI URN rather than a handle URN
since they're the same namespace. I think I get the politics, but it
also seems to me that the potential for multiple URN namespaces with
the same contents doesn't seem like a good idea. 

FYI, the doi.org server can resolve at least some non-DOI handles such
as the handles assigned to the ITU.  This will get you to the ITU X.509
spec:

https://doi.org/11.1002/1000/14033

R's,
John