Gathering Data: NISO E-Resources Management Forum

Oliver Pesch
EBSCO Information Services

"Gathering Data"

Starts with The Deal:
Buying something - a product, with resources and an interface
For whom? - a consortium, library, or location
Make sure - do a trial
How much? - acquisition process
Terms? - license or other terms of use
Expose product for access
Support with administrative interface and staff
(315 - # things DLF ERMI report says need to track)

Types of data:
Entities: - Library (16); eResource (33); Consortium (14); Trial (9); Acquisition; License; Terms; Access; Administration; contacts
(and so forth - didn't catch all #s - plus data gathered about the workflow

Sources of data:
-Library
-Publisher/Provider
-Agent/Jobber
-Consortium (financial, not physical)
-A-Z/KB supplier (you or a vendor)

Current data harvesting opportunities:
usage data  - COUNTER, SUSHI
E-Resource information - bib, MARC records from content providers and KB vendors

Current data standards:
COUNTER - makes usage stats consistent, credible, and comparable
revision 3 next year focusing on consortium reports, XML, SUSHI
Vendors must now be audited to be labeled COUNTER compliant
Consistent usage data provides home of ERMs to offer usage consolidation

SUSHI (NISO Z39.93 - waiting on ANSI)
Standardized Usage Statistics Harvesting Initiative
Automates harvesting of usage data using web 2.0 approach
ERM/Usage consolidations - can automatically connect to and retrieve usage data from any content provider with a SUSHI server
Challenges - adoption; lack of consistent identifiers; adoption (up to libraries to demand)

ONIX SPS
Serials Publications and Subscriptions
Used for communicating information about subscription products
designed for communicating price catalogs
Adv: allows some financial data; for title lists to be included in packages

ICEDIS/EDI
Series of formats to communicate order and activation data (designed for computer tapes)
Latest revision expands message to include IP addresses, other components appropriate for eresources
current format built on fixed data model - discussion to upgrade to XML
Adv: used by many pubs, agents and ILS systems

ONIX SOH
Serials Online Holdings
Communicates holdings information about electronic resources
includes coverage, URLs, embargoes, etc - stuff needed for link resolvers. Now stretching to include more granularity
Good for transfer of holdings from one platform to another

TRANSFER
UKSG formed to address problems caused by transfer of titles b/t publishers
set of guidelines for pubs to follow; ensure libraries have continuous access
considering a central repository of "transferred" titles

ONIX PL (License Expression Working Group - NISO group)
XML schema allowing terms of a license to be exchanged in a machine-readable form
work also being performed on a license editor; one goal is to allow negotiation by exchanging and editing a license
Captures terms of a license, but doesn't necessarily map to elements in ERMI license element
Interpretation of license still needed by staff or publisher

SERU
Shared E-Resource Understanding
alternative to a license - librarians and pubs agree a license is not necessary
documents expectations of behavior on part of pub, libraries, and users (makes purchase of an e-resource a lot like a print purchase because expectations are clear on all sides)
But - won't fit every deal; doesn't grant every right a library may want

ERMI-2 ILS Acquisitions and ERM Interoperability
beginning stages

Summary:
ERM intended to be single site to access all need to know about an e-resource
data needs are varied and complex
DLF ERMI data dictionary lists 315+ data elements (~158 are library's; s/b able to get ~146 from publishers and vendors) and 25+ entities
Data comes from many sources
Unlikely that one source will supply all data
Some automated feeds exist, many more are possible
Standards are key to interoperability and smooth data exchange

(Point from questions -
Will be interesting to identify "most important" 25 or so ERMI elements that yield greatest return on investment of time to enter -
costs you money to answer questions; costs you money to enter data; some data elements are in the long tail and won't yield enough return to make them worthwhile to complete)