...
For a resolver to work, its hostname must be carefully chosen so won't ever need to be changed. Memory organizations, some of them centuries old, tend to have hostnames well-suited to be resolvers. Some well-known, younger resolvers are n2t.net (the ARK resolver), identifiers.org, doi.org, handle.net, and purl.org.
How
...
does ARK compare to DOI, Handle, PURL, and URN?
These are all major persistent identifier schemes (or identifier types). Superficially they have much in common, starting with
The short answer as that these are all major types of persistent identifiers, and among them, ARKs are the only mainstream, non-siloed, non-paywalled identifiers that you can register to use in about 24 hours. Over 500 registered organizations have created an estimated 3.2 billion ARKs in the world, and no one has ever paid for the right to create them.
That's not to say that making ARKs persistent is cost-free. Keeping any identifiers persistent burdens every provider with the costs of content management, hosting, monitoring, and forwarding. You can do those things yourself or with help from a vendor, but with ARKs you are not charged separately for your identifiers and you are not locked in to a special-purpose resolution silo that also locks out other identifiers.
Is there a longer answer?
Yes, but first let's dispense with what ARKs, DOIs, Handles, PURLs, and URNs have in common. As seen in these examples, they have similar structure:
https://n2t.net/ark:/99999/12345
...
https://purl.org/12345
https://???<various>/urn:99999:12345
They all start As seen in these examples, they all begin with the protocol (https://) plus a hostname, followed by the Name Assigning Authority (99999, 10.99999, or purl.org), which is the organization that created a particular identifier. Finally there's the name, or local identifier, that it assigned (12345). More
Here are some more things these identifier schemes (types) have in common:
- They all fail to stop the major causes of broken links: loss of funding, natural disaster, war, deliberate removal, human error, and provider neglect.
- They all burden the end provider with the responsibility to update forwarding tables as URLs change.
- They all give access to any kind of thing, whether digital, physical, abstract, person, group, etc.
- They all identify content that is subject to change on future visits.
- They all break regularly and in large numbers (thousands and more).
- They all use very simple redirection built in to every web server since 1995 1994 and provided for free by hundreds of URL shortening services.
- They all (as a result) leave you wondering if you need them at all, and if so, at what cost.
How do ARKs differ from identifiers like DOIs, Handles, PURLs, and URNs?
The short answer is that ARKs are the only mainstream, non-siloed, non-paywalled identifiers that you can register to use in about 24 hours. Over 500 registered organizations have created an estimated 3.2 billion ARKs in the world, and no one has ever paid for the right to create them.
That's not to say that making ARKs persistent is cost-free. Keeping any identifiers persistent burdens every provider with the costs of content management, hosting, monitoring, and forwarding. You can do those things yourself or with help from a vendor, but with ARKs you are not charged separately for your identifiers and you are not locked in to a special-purpose resolution silo that also locks out other identifiers.
In what other ways are these identifier types different?
Here's a longer list:So how do these identifiers differ? Here's a short list.
- When (not if, because all things pass) the
https://protocol and the hostname cease to exist, only ARKs and URNs indicate the type of identifier that remains. - For DOIs, Handles, and PURLs, you are required to use their respective resolvers. ARKs and URNs, permit you to use your own resolver.
- To create DOIs and Handles, you are required to pay a membership fee and, for DOIs, per-DOI charges. There are no fees for ARKs, PURLs, and URNs.
- Although you can use your own or a vendor resolver for your ARKs and URNs, all ARKs can be resolved via n2t.net, making it the closest thing to a "global ARK resolver".
- For URNs there is no single global resolver. In order to register to create URNs, you must apply for a URN namespace.
When should I use ARKs compared to DOIs, Handles, PURLs, and URNs?
There is nothing are no simple answers. Identifiers are abstract and tricky to talk about, so it helps to beware of common fallacies. Nothing inherent in ARKs, DOIs, Handles, PURLs, or URNs that make makes them more or less suitable to identify any kind of thing in any fit for any particular field, domain, or sector. In that sense they are all equally suitable.
Where they differ are in the nature of services and sociology and buzz. xxx
With an identifier resolver and administrative management service, they can all provide the core service of resolution (for that matter, so can any properly managed URLs).
The concrete differences that we think we experience, such as restrictions on identifier metadata (descriptions) and landing pages, are not properties of identifier strings, but properties of resolution and management services extended to or withheld from those strings. The foundation of these services is a reliable database that stores each identifier and its metadata elements (eg, creator, title, date, redirection URL) to describe the identified object. Other services include things like link checking, duplicate detection, report generation, and searching.
Typically, services are built as siloes ("walled gardens") to serve a particular identifier type (eg, Handle, DOI, or PURL) to the exclusion of other types. Such practice takes extra work and violates basic principles of openness, so the N2T resolver and EZID (identifiers made easy) management interface were designed to work with all identifiers. Work put into any new feature can be efficiently leveraged across all types, which sometimes creates surprising flexibility; for example, ARKs are sometimes stored in EZID with "DOI metadata", and every DOI stored in N2T can benefit from "ARK resolution features" such as inflections and suffix passthrough, which are not available via the main DOI resolver (doi.org). Here are some concrete differences in metadata
There are also softer differences, such as community attitudes and trends, based on numbers and "buzz". If we did free association in 2019, one might hear:
ARKs – libraries, archives, museums
DOIs – published literature (Crossref), published data (DataCite)
Handles – xxx
PURLs – linked data and semantic web
URNs – UUIDs, European libraries
any of these identifier It is hard to generalize how people use these identifiers. DOIs, for example, used to be known primarily as identifiers for scientific and scholarly publications, with a mature community and service offering around "Crossref DOIs", but newer kinds of DOIs, such as those from DataCite and EIDR, are changing the nature of the DOI.
...
Those are special kinds of persistent identifiers. ORCIDs only identify researchers, and they link to research works using ARKs, DOIs, etc. ORCIDs look like
...