ARK current use cases and examples

370 views
Skip to first unread message

Anna Neatrour

unread,
Sep 4, 2012, 12:40:40 PM9/4/12
to arks-...@googlegroups.com
I was wondering if some of the people in this group could share some examples of current or upcoming use cases involving ARKs?

Mark Phillips

unread,
Sep 4, 2012, 2:27:29 PM9/4/12
to arks-...@googlegroups.com
The UNT Libraries uses ARK identifiers for all digital objects we place in our digital library infrastructure. 

Here is an example of what our ARKs look like. 

ark:/67531/metapth123456

We provide a public view of this item through our access system called Aubrey. 


All of the views we have into our digital objects are based on these URLs.  For example a thumbnail of an object can be seen by adding /thumbnail/ to the end of the ARK


We mint our ARKs with a tool we call our number-server which handling handing out sequential identifiers for a given prefix, for example metapth1, metapth2, metapth3 and metadc1 metadc2 metadc3

I'm not sure if that is the type of use case you were looking for. 

Mark

Anna Neatrour

unread,
Sep 4, 2012, 3:23:29 PM9/4/12
to arks-...@googlegroups.com
Thanks! I'm just trying to get a general idea of how people are using ARKs, this is very helpful.

CB

unread,
Sep 4, 2012, 3:25:38 PM9/4/12
to arks-...@googlegroups.com
The University of Chicago Library uses ARKs as unique identifiers for digital archival accessions, and as unique identifiers for archived objects in its digital repository.

Michael J. Giarlo

unread,
Sep 4, 2012, 3:42:00 PM9/4/12
to arks-...@googlegroups.com
We at Penn State are using (ARK-inspired) "noids" as unique, opaque
identifiers for objects in our new ScholarSphere repository service[1].
We have a lightweight identity microservice[2] built on the Ruby-based
noid library[3] that mints and validates identifiers. The noid
identifiers look something like "6jz77ab8".

These identifiers are used within our Fedora repository (after having a
Fedora prefix prepended), and are ultimately proxied and resolved by the
ScholarSphere web application (built using technologies from the Hydra
Project[4]). The resulting URIs look like
"http://scholarsphere.psu.edu/files/6jz77ab8".

-Mike

1. http://stewardship.psu.edu/2012/05/launching-scholarsphere.html
2.
https://github.com/psu-stewardship/scholarsphere/blob/master/lib/scholarsphere/id_service.rb
3. https://github.com/microservices/noid
4. http://projecthydra.org/

John A. Kunze

unread,
Sep 4, 2012, 4:30:42 PM9/4/12
to arks-...@googlegroups.com
Hi Mike,

Great to learn know about the ruby-noid project! Does it re-implement
all of the noid Perl module in ruby (or just the minting part)? Also
(related question), does resolution involve noid's built-in resolver?

-John
> --
> You received this message because you are subscribed to the Google Groups
> "ARKs" group.
> To post to this group, send email to arks-...@googlegroups.com.
> To unsubscribe from this group, send email to
> arks-forum+...@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/arks-forum?hl=en.
>
>

Michael J. Giarlo

unread,
Sep 4, 2012, 4:53:20 PM9/4/12
to John A. Kunze, arks-...@googlegroups.com
Hi John,

It does not re-implement all of the noid Perl stuff -- just minting and
validation of check characters.

Resolution does not involve noid's built-in resolver in our
ScholarSphere app. Resolution happens roughly like this:

1. Requests to the /files/abc12xyz34 path are routed to a Rails
controller.
2. The controller "normalizes" the abc12xyz34 noid by prepending our
Fedora namespace.
3. The controller queries our Fedora-based repository using
ActiveFedora to see if the normalized identifier has been bound.
4. Do the authentication, authorization, and access control dances. etc.

-Mike

Declan Fleming

unread,
Sep 4, 2012, 5:14:07 PM9/4/12
to arks-...@googlegroups.com
Hi - we use ARKs for our DAMS too.


Declan


For more options, visit this group at
http://groups.google.com/group/arks-forum?hl=en.


--
You received this message because you are subscribed to the Google Groups "ARKs" group.
To post to this group, send email to arks-...@googlegroups.com.
To unsubscribe from this group, send email to arks-forum+unsubscribe@googlegroups.com.

John A. Kunze

unread,
Sep 4, 2012, 5:47:30 PM9/4/12
to arks-...@googlegroups.com
I'm not sure if anyone from the Internet Archive is subscribed to this
list, but I'll risk speaking to their use case. They assign ARKs to
scanned texts, mostly books (as of today they're up to 3,408,730 ARKs).

-John

--- On Tue, 4 Sep 2012, Declan Fleming wrote:
> Hi - we use ARKs for our DAMS too.
> URLS look like:   https://libraries.ucsd.edu/ark:/20775/bb18440132 
>
> Declan

Sébastien Peyrard

unread,
Sep 5, 2012, 4:45:03 AM9/5/12
to John A. Kunze, arks-...@googlegroups.com
At the national Library of France, we are using ARKs for
  • all our digitized books, still images, manuscripts and phonograms in our digital library Gallica, e.g. http://gallica.bnf.fr/ark:/12148/bpt6k71351c
  • any information package that is preserved as an AIP in our preservation system (SPAR), which will include other digital stuff like our ARC files for web archives as well, born-digital audiovisual and software resources, archive records from the library...
  • all our catalog (bibliographic and authority records), e.g. http://catalogue.bnf.fr/ark:/12148/cb308671030
  • we also intend to use it to identify our online exhibitions as well (expositions.bnf.fr)

Hope this helps!

Kind regards,

Sébastien


2012/9/4 John A. Kunze <j...@ucop.edu>

Anna Neatrour

unread,
Sep 5, 2012, 1:12:19 PM9/5/12
to arks-...@googlegroups.com
Thanks everybody for the replies so far! We are just beginning our implementation planning process at the University of Utah, so it really helps to see some examples of current ARK usage.
I don't want to clog up the list with too many newbie questions, but I was also curious to see if anyone was using ARKS with Rosetta?
To unsubscribe from this group, send email to arks-forum+...@googlegroups.com.

Christy Shorey

unread,
Sep 14, 2012, 8:36:54 AM9/14/12
to arks-...@googlegroups.com
We digitize material with Internet Archive and are beginning to send material for inclusion in the Hathi Trust. One of the pieces of metadata that Hathi has requested from us is the the ARKs that Internet Archive assigns.

-Christy

Ed Summers

unread,
Sep 14, 2012, 9:35:37 AM9/14/12
to Christy Shorey, arks-...@googlegroups.com
On Fri, Sep 14, 2012 at 8:36 AM, Christy Shorey <weyl...@gmail.com> wrote:
> We digitize material with Internet Archive and are beginning to send
> material for inclusion in the Hathi Trust. One of the pieces of metadata
> that Hathi has requested from us is the the ARKs that Internet Archive
> assigns.

Thanks for sharing. I think this is an excellent use case for ARK,
where the same "information object" (in this case a book) may happen
to be held by multiple institutions, and have multiple homes on the
Web. Do you happen to know if HathiTrust create a new ARK for the
items you are sending, or if they adopt the IA ARK, or if they do both
and relate them together somehow in their own metadata?

//Ed

Jeremy York

unread,
Sep 17, 2012, 2:41:12 PM9/17/12
to arks-...@googlegroups.com, Christy Shorey
We use the ARK identifier from the Internet Archive, with a pre-pended namespace identifier. There is more information about our identifiers, including an ARK id example, at http://www.hathitrust.org/deposit_guidelines#pdi.

Jeremy

Jeremy York
Project Librarian, HathiTrust

Abe Pineda

unread,
Apr 15, 2013, 4:29:12 PM4/15/13
to arks-...@googlegroups.com
The Hawaii Digital Archives is implementing ARK identifiers for our digital objects. We are currently in the works of developing this application and are also employing the NOID persistent ID service, but only in regards to generating an identifier. We haven't created and ARKs as of yet, but I am curious as to how this process plays out. Many thanks to anyone who can reply. 

Amy Kirchhoff

unread,
Apr 18, 2013, 10:04:10 AM4/18/13
to Abe Pineda, arks-...@googlegroups.com

Hi Abe ~

 

At Portico, we assign ARKs to *everything* … objects in our content model and chunks of metadata (every event has an ARK, every block of technical metadata has an ARK, etc.).  We churn out more than 100 million NOID IDs a year.

 

We haven’t done anything special about setting up the NOID service, though we run our own.  We do use different minters (ID prefixes) for different content types.

 

Good luck.

 

~ Amy Kirchhoff

 

--

Amy J. Kirchhoff

Archive Service Product Manager

Portico, JSTOR

(p) 609-986-2218   (f) 609-951-0020   (e) Amy.Ki...@ithaka.org

 

Portico (www.portico.org) is a community-supported preservation service for electronic journals, books, and other scholarly content.  JSTOR (www.jstor.org) is a digital library of academic journals, books, and primary sources. JSTOR and Portico are services of ITHAKA, a not-for-profit organization that also includes Ithaka S+R.

--

You received this message because you are subscribed to the Google Groups "ARKs" group.

To unsubscribe from this group and stop receiving emails from it, send an email to arks-forum+...@googlegroups.com.


To post to this group, send email to arks-...@googlegroups.com.

Abe Pineda

unread,
Apr 18, 2013, 8:36:09 PM4/18/13
to arks-...@googlegroups.com, Abe Pineda
Thanks for your reply Amy, but I was actually wondering as to how the actual assigning process is carried out, like how a digital object gets assigned a persistent ID. Is this done through some sort of online registration process? Again, much appreciated for your reply. 

Amy Kirchhoff

unread,
Apr 26, 2013, 10:17:45 AM4/26/13
to Abe Pineda, arks-...@googlegroups.com

Hi Abe ~

 

I caught up with my systems folks and queried them about our configuration.

 

What we’ve done is created a servlet that interacts with NOID tool.  Our code interacts with the servlet.  The servlet handles the command line requests to the NOID tool for Noids.  Our workflow code requests up to 10,000 Noids at a time and caches them.  The workflow then hands out Noids as ARKs as requested by other pieces of the workflow and goes and grabs another 10,000 Noids when it runs out.

 

As a general overview, we process publisher supplied content.  The system unpacks the directories and files and then puts the content back together again in the Portico content model.  We also transform publisher supplied XML to a preservation standard.  Every file, semantic unit, event, and metadata block is assigned an ARK.  Any process that needs to assign an ARK grabs one from the cache.  We do not “register” them as we process (we do not have a master database mapping every ARK to the object it identifies).  The final steps in the workflow are to deposit the content in the Portico archive and to drop much of our metadata (and identifiers) into an Oracle database.  Conceptually, this database is just a cache to make it easier for us to access and find content, it is not part of the true Archive.

 

I hope this helps.

 

~ Amy

Reply all
Reply to author
Forward
0 new messages