Page History

ISSN as NAAN, part 2

NAANs originally were for institutions (eg, National Library, Portico, University), but over time, CDL loosened the guidelines (eg, University Library instead of an entire University). With EZID, CDL began to follow  the lead of the publishers and DataCite, assigning ARK NAANs the same way that DOI Prefixes were assigned (eg, departments in organizations, laboratories, virtual organizations such as consortia), and even to individual journals.

Broadly, NAANs (like Prefixes), were given to groups that asked and could plausibly claim to be significant, stable sources of identifiers. Add to that, that some organizations asked for more than one NAAN (Ithaka/Portico). 

Still undefined for NAANs: what happens when (a) an org goes away or merges, (b) another org asks/offers to take over a frozen namespace, (c) wants to change its NAAN?

How often would (c) happen and how should we respond? In perspective, if an organization asks for an 8-digit NAAN (whether or not based on an ISSN), how do we respond? Are NAANs-as-ISSNs more or less likely to change than other NAANs? Could we not contain ISSN instability with the same rule that contains NAAN + define organization to be an institution, a department, or possibly anyorganized and stable name assigning effort
+ then describe what to do when orgs change (eg, name), or journals get new
ISSN (old NAANasISSN would have to be honored)(regardless of origin) instability?

ARK as URN or URI

On Tue, Aug 13, 2019 at 10:44 PM Mario Xerxes Castelán Castro <mario.xerxes.castelan.castro@gmail.com> wrote:

Hello. Are there plans to register “ark” as an URI scheme or under the existing “urn” scheme? If there are not, I suggest that it is considered. For some purposes where URI are expected it would make sense to be able to use the bare ARK, for example: XML namespaces as in <ark:/12345/example>. That would give the full persistence assurance that ARKs have, not limited by any particular resolver as is the case if one uses an ARK with NMAH like <https://n2t.net/ark:/12345/example>

Response:

Registering "ark" as a URI scheme or URN namespace has been considered in the distant past, and it may be worth considering again. I'd be interested in hearing feedback about this.

Here are some considerations. As for registering "ark" as a URN namespace, I'm not sure what it would accomplish other than to permit expressing ARKs in the form,

    1. urn:ark:/12345/67890

in addition to the usual form,

    2. https://n2t.net/ark:/12345/67890

While the urn form (1) is shorter than the usual form (2), it can only be made resolvable by using an even longer form, such as,

    3. https://n2t.net/urn:ark:/12345/67890

Regarding registering "ark" as a URI scheme, what would an ARK look like and where would resolution take place? Would it be this?

Form:
    ark:/12345/67890

Implicit resolution via:
    https://n2t.net/ark:/12345/67890

It's a nice idea, but (a) does it prevent de-centralization (a feature of ARKs) and (b) given that this sort of thing was tried unsuccessfully in the past with DOIs and Handles, what argument would make ARKs successful here?

alternatives to google for arks-forum

On Tue, Aug 13, 2019 at 10:44 PM Mario Xerxes Castelán Castro <mario.xerxes.castelan.castro@gmail.com> wrote:

Incidentally, I did not find an option to subscribe to this group using an existing e-mail address, so I had to register one with Google. I find that very inconvenient as I would rather using Google services as much as feasible.