Thanks to the people who replied to my query about those working on
extended "find it/get it systems." There are some very interesting
projects out there; here's my promised summary of replies:
==============================================
University of Minnesota
We're considering some very similar strategies here at the University of
Minnesota, although we're a few months away from implementing a link
resolver. We established some good groundwork in efforts to implement a
cookie-based central authentication approach to single login.
John T. Butler <[log in to unmask]>
==============================================
University of Pennsylvania
The Penn Library is experimenting with tools to perform some of the
services you envision. We're working on ways for patrons to automatically
build a personal profile and customized view of library assets based on
their past use of resources, rather than laboriously build a profile
manually.
The user would opt in to have us analyze their usage logs and build an
automatic and dynamic profile of their favorites as well as suggest other
resources based on our knowledge of their "community of interest". The
user could also modify their profile manually, and incorporate other
MyLibrary and Portal features into their custom interface.
Although federated searching is our goal, we agree that native search
interfaces are needed by researchers. We've experimented (pretty much
unsuccessfully so far) with combining Z39.50, XML Gateway and HTTP "screen
scraping" technologies to try and build an interface that is suitable for
researchers rather than undergraduates.
Much of this is still in the planning and testing stage so we do not
pretend to have solved all the problems.
Roy R. Heinz <[log in to unmask]>
==============================================
Library of Congress
In developing our virtual reference services, CDRS (Collaborative
Digital Reference Service) and QuestionPoint, our vision has been a
"one-stop-shop" service. This service would include a discovery element
with broadcast searching of a variety of databases and a fulfillment
option with document delivery options such as ILL. Our most recent
research is also pointing in the direction of portals and open URL
technology. As you mentioned the broadcast searching offered by most
portals is still based on a common user interface and the Z39.50
standard. Two vendors, MuseGlobal,
http://www.museglobal.com/Libraries/, and WebFeat
<http://www.webfeat.org/prism.html>, are offering products that
support multi-protocol searching based on Z39.50, HTTP, telnet, and
proprietary protocols.
In addition to our ongoing development of QuestionPoint the Library of
Congress also has a Portals Applications Issues Group that is
investigating current portal/open URL technology and its applications to
our library services.
Mitzi Cole <[log in to unmask]>
==============================================
California Digital Library
The California Digital Library (CDL) has developed a multi-functional,
multi-part system for providing content services for the UC community. It
is possible for a member of the UC community to search the Melvyl union
catalog as well as remote databases on vendor sites (e.g., PubMed or CSA
or Ovid), and produce ILL requests for materials found in those databases
with a minimum of additional effort.
The CDL's Request system utilizes the OpenURL standard to receive
citations into its SFX server and from that server to offer links to full
text and document delivery or ILL services. This is done with a separate
"request" service that acts as a target for SFX and receives an OpenURL
from it, then queries the Melvyl database (using ISSN or ISBN) to retrieve
UC holdings for the items. The Request service can retrieve profile
information from the Melvyl (Aleph) system, and authenticates the patron
against that person's local campus circulation system, then parses the
personal and bibliographic data into a structured email message that is
loaded into a centralized VDX tracking system. Requests for materials
that cannot be filled within the UC system are then forwarded from VDX to
OCLC using the ISO ILL protocol. Patrons may query the VDX system using
their campus library barcode number to view the status of previous
requests.
Mary Heath <[log in to unmask]>
==============================================
Colorado Alliance of Research Libraries
We are working on a link resolver, public search interface and ejournal
system that has some of the elements you describe. The system is called
Gold Rush.
If you are interested take a look at our marketing website at
http://grweb.coalliance.org
If you would like to see our public interface to the link resolver go to
http://goldrush.coalliance.org. Be sure to pick a library in the "Change
Library" menu (on the top right) so you can see what patrons at a
particular library would see.
George Machovec <[log in to unmask]>
==============================================
For those who asked what Hopkins is doing: nothing yet, I'm afraid. We
have ideas for what we'd like to acheive, and wanted to poll the DLF
community to see if work is already being done in this area. As we
evaluate existing projects and products (and our available staffing) I'll
certainly keep the DLF community informed about our activities.
Thanks again for your feedback!
- Nathan
On Tue, 25 Feb 2003, Nathan Robertson wrote:
> Colleagues,
>
> We at Hopkins are interested in the possibility of automated systems that
> would go a step further than the current capabilities of SFX-like link
> resolvers. This envisioned service would enable users, wherever they are
> in the virtual library universe, to click on a "get" button which will
> invoke a service that will automatically search available full-text
> services, if not found then route the request to the catalog system to
> automatically place a request for print copies, and if not found in the
> catalog it would generate an ILL request for the item--all with minimal
> intervention from the user (or from staff!).
>
> This system would also be integrated with a "my library" portal, allowing
> users to track past search history, check on the status of ILL requests,
> save citations results, etc. etc.
>
> The existing services that come close to this functionality rely on a
> portal and unified search interface to manage the authentication and
> storing of personal data and preferences.
>
> We want to avoid forcing users to use a unified search interface to get
> these services, because unified search interfaces all necessarily
> sacrifice some of the important functionality of the native interfaces.
>
> We have spoken with a few library system vendors about their project plans
> in this area. We would also like to know if any of the DLF member
> libraries are working on projects that tackle some or all of this scope.
> Are any of you working on such projects?
>
> (If you respond directly to me, I will summarize results to re-post to the
> dlf list).
>
> - Nathan
>
> --
> =+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
>
> N.D.M. Robertson [log in to unmask]
> Database Analyst/Programmer vox (410) 516-8886
> Sheridan Libraries fax (410) 516-5339
> The Johns Hopkins University
|
