Identifiers Webinar Q&A
Below are listed questions that were submitted during the NISO Identifiers
Webinar. Answers from the presenters will be added shortly. Not all the
questions could be responded to during the live webinar, so those that could
not be addressed at the time are also included below.
Feel free to contact us if you have any additional questions about library, publishing, and technical services standards, standards development, or if you have have suggestions for new standards, recommended practices, or areas where NISO should be engaged.
What's in a Name?: Identifiers for Institutions, Public Identities, and
Webinar Questions & Answers
October 29, 2008
- In your opinion, which of the various identifiers will "prevail"
- What is your perception of publisher perception of these identifier
initiatives? Beyond GDSN, what kinds of things are you hearing from
publishers in their involvement in these standards?
- Can you speak to the sustainability of vendors-based IDs vs. national
or international IDs?
- To guarantee interoperability, is NISO developing an ontology or
metadata framework that variant id systems must comply with?
- How is work in author registry related to LCNAF, and what is NISO
doing to help in this area?
- Is the NISO WG focused just on institutional IDs or also author
Answer (Karen Wetzel, NISO): NISO's I2 working group (www.niso.org/workrooms/i2) is focused on institutional identifiers, though NISO is looking to see what kind of role we might play in the area of author identifiers -- there is a lot happening in this area right now, and so we need to consider how a standard or recommended practice might fit in that space. Any feedback that you have or suggestions would be most welcome: www.niso.org/contact
- How is the persistence of the ResearcherID (Thomson)
- For ResearcherID: LCNAF (the Library of Congress Name Authority File)
is also a very good source for author identifier. My question here is what
mechanism you have in place to work with a culture (in the case of science,
engineering environment) where authors do not generally provide full name
when a work is published.
- For ResearcherID: are the work references ones that are non-Thomson
- How do you ensure uniqueness if a publisher batch uploads
- Are all the authors in Researcher ID working in the sciences? Is
there an interest in expanding to the humanities and social sciences?
- Are all the authors in ResearcherID those who registered themselves?
Do you add any others at Thomson Reuters?
- Is there any collaboration going on between I2 and
Answer (Grace Agnew & Tina Feick): A major part of our workflow is to examine standards and initiatives that are either complementary or may form part of the solution for a robust institutional identifier framework. We will be looking at standards such as ISNI and initiatives such as ResearcherID, OCLC NNAG and the Nimas project to ensure that we have considered how they might address or interoperate with our problem space. Some of our working group members are researching these initiatives. Scenario development teams will be looking at these to inform their scenario development and as I2 stakeholders. We also plan to invite initiatives that impact many scenarios, such as ISNI and ResearcherID to participate in a monthly conference call with the I2 working group so we can explore synergies directly.
- Is it possible to embed a feed from ResearcherID.com within another
page, rather than just linking back to the ResearchId.com website (with
reference to the RID website)?
- How many researchers are registered in ResearcherID? What's the
discipline and geographical distribution? Also, can you guarantee ID
persistence? Are you working with social networking apps? Linkedin for
- For the publications that are not indexed by ISI, will they have any
citation information in ResearcherID database? And how many researcher
records are stored in the database now?
- If I as a publisher upload a name of someone already in your registry
but with a different email address, do you assign another ID to this
- In the past you have presented on a partnership between University of
Queensland and ResearcherID. How does that fit into the general
- Do we assume ISNI system can have dozens of roles for a single
person, as one participates in various organizations with various roles? Can
all roles for a person be retrieved?
- Is ISNI going to work with any of the authorities work being done
nationally (OCLC names group) and internationally, with the FRAD work from
the International Federation of Library Associations and Organizations
- What is the humble user-the librarian, author, publisher-to do while
you are all figuring out interoperability? Pick a standard and hope that you
pick the right one?
Answer (Grace Agnew and Tina Feick): This is an excellent question. There is a tremendous amount of work occurring in the identifier standards arena, at various stages of development and adoption. Our recommendation would be to study the landscape and identify the standards initiative(s) that seem most relevant to your organization or program. If a developing standard that seems most applicable is not available in a fairly sturdy draft, we suggest contacting the working group chairs responsible for its development and asking advice for developing a transitional scheme for your immediate use. Identifiers serve a number of purposes, from unambiguous identification to efficient machine processing, but one of the most critical purposes is to enable collaborations through the use of globally unique identifiers. If a national/international standard is under development but not yet available, choose a lightweight transitional approach that you can easily migrate to the standard. And do consult with, and alert, the relevant standards developers. It is their responsibility to develop a standard that can easily replace and interoperate with legacy practices, but they can’t discover every legacy practice out there. You have to tell them you exist. They should also be able to advise you on a “safe” interim strategy.
- Are all the identifiers being discussed "dumb numbers" like the ISNI,
or do some have semantic elements? If so, what semantics?
Answer (Grace Agnew and Tina Feick): The I2 Working Group has not selected an identifier standard yet, but we are strongly predisposed toward an opaque or “dumb” number, for the simple reason that identifiers should be both scalable and durable, even outlasting the thing that was identified, since the identifier can then point to the historic record of what once existed. Roles and characteristics of the “thing identified” on the other hand are not particularly durable. A case in point would be “Grace Agnew,” currently a librarian at Rutgers University who might leave to become an IT administrator at Princeton University. If the identifier assigned to Grace Agnew included a sub-element for the assigning organization (Rutgers), and even perhaps a sub-element relevant to the subunit of that organization (library) or Grace’s role in the organization (librarian), both of these elements can cease to have meaning. Not only are individuals peripatetic, but organizations, such as publishers, merge or cease to be going concerns. If an identifier has a prefix that indicates whether the thing identified is a person or a publisher, what do you do with the individual who might be a librarian by day and run a digital publishing firm by night? Do you assign two identifiers? We feel that the semantics are best left to metadata. In the previous case, there would probably be a metadata instance for the individual as librarian or researcher and a metadata instance for the individual as publisher and the opaque identifier plays a critical role in cross referencing the two as one entity, but with different roles for different contexts.
- Is the ISNI draft available?
- Can ISNI be assigned to entities that are subjects of publications
(what other roles beside creators, publishers are included)?
- What could be the business model for INSI, etc.-charge per use?