- Количество слайдов: 68
FLOW: building Federated Libraries On the Web
A Partnership Approach: Research Science-Library-National Center Alliance Karen Baker (UCSD/SIO/LTER) Anna Gold (UCSD/Library) Dave Hart (UCSD/SDSC) Mike Vildibill (UCSD/SDSC) Jean-Yves. Le. Meur (CERN)
Federating Bibliographies: Personal to Project to Network l l l Multi-tier organization unit An individual, a research project and a research network have bibliographic needs in common. They represent nested tiers across which information can flow when technology and metadata standards are partnered to provide an accessable, interoperable digital framework
Publications: Multi-Tier Approach: Organizational Unit Identity
Why FLOW ? : l l We are oriented toward research and discovery as a process, not a product. The process observes protocols that enable federation and interconnectivity The process also allows personal or institutional differentiation or “streams” (both into and out of the federated processes) The process uses the web as a common interface
Immediate needs in I-B: l Collect data via desk top and/or web forms l l l Publications Research grants People / contact information Retrieve data in multiple ways from RDMS Create public exposure for data to enhance its impact
These needs have a lot in common with: l l LTER Network Bibliography NPACI/SDSC Bibliography/Citation Tracking Structural Engineering Bibliographies Plus many other implementations of Open Archives Initiative (OAI) and e-prints systems
A Partnership Approach: Research Science-Library-National Center Alliance Knowledge Systems Federated Library LTER CERN SDSC Xxx NCEAS DLIB LIBRARY
Communications Infrastructure The LTER Network
Example 1: l LTER Network: Using End. Note to gather structured bibliographic data from disparate sites. Using proprietary software as means to share data for discovery and networking.
Example 2: l NPACI/SDSC: Regularly compiles bibliographic records and citation counts, in order to testify to research impact of NPACI funding. Gathering is manual, no means to share or enable discovery.
Example 3: l Structural Engineering Department: faculty maintains large End. Note files of references on his research on a dedicated workstation. Wants to share, enable interactive submission, and enable discovery by research partners overseas.
Common Needs: l Gather structured resource information l l In batches (End. Note, Web of Science searches) One-by-one (individual submissions) l Share collections of information with research networks l Discover: relevant research, partners, updated results, linkages
Gather/share/discover WHAT? l l l l Journal publications Preprints Book chapters Presentations Conference papers Technical reports And: photographs, media, data, people, etc.
Implementation desiderata…. l l l Standards-based Open Flexible Fast Search within or across collections
Functional desiderata…. l l l l Modify and update submissions Provide full text via local or remote files Search by fields or full text Citation counting Handle various media and types of data Provide for review process(administrative, peer, etc. ) Customization and personalization Alert services
In short: A generalized tool that can be easily adapted to the needs of any research organizational unit: individuals, groups, organizations, and networks.
Solutions? l l Does this infrastructure already exist? Maybe: some options to explore: l l Open Archives Initiative (e-prints) CERN Document System (CDS)
The CERN Document System l l l One of several infrastructures / systems to be explored under grant proposal to NSF (K Baker, A Gold, Winter 2002) Compatible with open e-prints initiatives in research communities (OAI) but… Independent of eprints / OAI priorities (CDS custom modules and programming support are in place now)
CERN CDS: l does this system meet Integrative Biosciences needs?
CDS capabilities: l l batch upload bibliographic citations submit and modify submissions support differentiated collections or “catalogues” that can be searched separately or together implement other modules, including: personalization, alert services, output and file format options
Gather l Batch input through CDS “Uploader”
Full Record (Web of Science):
Full Record (SDSC Thrusts):
Share l l Submit citations, links, and / or full text for review or publication Modify submitted items/information Notify others of new submissions Share custom lists (“baskets”) and share with others
Test Document Submitted:
Submit New Document:
Submit, Screen 2:
Submit, Screen 3:
Submit, Screen 4:
Transfer files screen:
Upload File and File Type:
Add More Files (Optional):
Submit Revised Version:
Full Text Options (converted from Word to PS, PDF)
Send to Distribution List:
Discover l l l Search: one collection or many Output options (also custom views, baskets) Browse
Results list (“printer-friendly”):
Results list (XML):
Your Baskets (login):
Display / Share Baskets:
Your Searches/Email Alerts:
Set Up Alerts:
Steps to Implement CDS: l l l Sign a collaboration agreement/contract with CERN technology transfer department Transfer/install/configure/customize full software on our server Load all data on our site Test software with the real data Move to production when we think it's ready.
Role of the partnership: l l l l Empower partners with broader vision Bridge individual, local project, & national needs Define and meet current needs Provide arena for cross-domain work Bridge from present to future needs Anticipate change, optimizing for sustainability Work from bottom up toward knowledge system
Partnership History (Where did this come from? ) l Library-SDSC Collaboration ü l l LTER-SDSC Collaboration NSF grant (Baker/Vildibill) 2001 -2003: ü l l Palmer LTER Site Information Management: Interfaces for Science and Education Interoperability SDSC Semantics Interest Group Citation Bibliography ü ü l Proposal for library services SDSC 1991 -1996 (=5584 entries) NPACI 1996 -2001 (=836 entries) LTER 1991 -2001 (=890 entries) Web of Science (=188 entries) Document Management System ü ü ü Open Archive Initiative (OAI)-protocols CDS(CERN)-base open/supported implimentation of OAI Eprints(DLIB)-open implimentation of OAI
Partnership Goals (Where is this going? ) l l l Create multi-tier citation/document libraries Create input & retrieval tools for users and administrators Assure discovery and output capabilities Design for interoperability through identification of national categories and use of national standards Design for local needs (define local categories) ü ü Term lists (controlled vocabulary) Keyword fields, ie by discipline, theme, grant support, bibliography l Design for scalability through mapping (cross-walking) l Consider sustainability of process l Critical mass to maintain momentum & obtain funding
GATHER Ingest WEB Controlled Vocabulary LOCAL GATHER Standardize (Keeper-of-the Keys) Filter *. enf QA/QC SHARE Term. Lists *. enf Tagged %0 Report %A Baker, K S %T Partnership Science %D 2001 %@ UCSD-SIO 01 -03 %2 M 1037 %4 SDSC End. Note *. enl Style *. ens DISCOVER Public Viewer As formatted output XML type Report creator Baker, K. S. title Partnership Science date 2001 identifier UCSD-SIO 01 -03 subject M 1037 relation SDSC Individual User As txt-enl-xml SDSC Infrastructure Interface As Oracle import Application Packages DTD EXPORT K. S. Baker - 01 Nov 01
Activities (Done) l l l Obtain seed funding for bibliographic prototype project Create research science/national center/library partnership Establish communication mechanism Identify local needs via terms lists used in existing biblios Make available biblio file for individual user in multiple formats Establish report output capabilities ü ü formatted list (from Endnote -individually driven) for proposals/papers formatted list for reports statistics
Activities (Doing) l Establish report output capabilities ü l l formatted list (from Oracle -dynamically driven) Investigate existing tools and alternative designs Investigate qa/qc mechanisms Integrate biblio_file with broader schema Integrate biblio files with "person_table"
Ongoing Communications… http: //pal. lternet. edu/dm/projects/semantics
Focus on Bibliographic Example
Design Details… • Reference types • Tag types %C city of conference or of publisher %0, %2 Reference type § § § -abstract, paper, report, thesis -associated, outreach, nonref -thesis: subtype Masters/Phd %2 Disciplines %4 Site id %K NPACI thrust • Term lists/categories ü public • • discipline subject ü private • • projects cross-projects
Design Details 2. . . • Specifics ü ü Duplicates - Focus on author/title uniqueness Completeness Author-People-Citation normalization Process - Keeper-of-the-keys • Endnote ü import/export interoperable ü structure for QA/QC and to format w/word ü local input (controlled vocabulary)
Activities (To. Do) l l l Reassess prototype tool for immediate focus Launch biblio effort as an in-house testbed Create Best Practices for Bibliographic Citations ü l l term lists and a keeper-of-the-keys Establish infrastructure location SDSC: Maintain repository over time Partnership: Obtain funds to support tool development Complete existing bibliography files
Discussion / Questions
References: l Semantics Interest Group: http: //pal. lternet. edu/dm/projects/semantics/xbibexample. html l CERN CDS: Search http: //weblib. cern. ch/Home/SDSC_Collection/ l CERN CDS: Submit http: //doc. cern. ch/send? ucsd l Open Eprints: http: //www. eprints. org/
END In support of science Federating bibliographies Creating an organizational bibliography Structured, parsable, loosely federated, multi-tier approach