UBC Library and Archives

Data Management Working Group Report and Appendices Barsky, Eugene 2011

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
Barsky_Data_Management_Report_Appendices.pdf [ 100.39kB ]
Barsky_Data_Management_Report_Appendices.pdf
Barsky_Data_Management_Report.pdf [ 78.88kB ]
Barsky_Data_Management_Report.pdf
Metadata
JSON: 1.0077923.json
JSON-LD: 1.0077923+ld.json
RDF/XML (Pretty): 1.0077923.xml
RDF/JSON: 1.0077923+rdf.json
Turtle: 1.0077923+rdf-turtle.txt
N-Triples: 1.0077923+rdf-ntriples.txt
Citation
1.0077923.ris

Full Text

1  The University of British Columbia Library Data Management Working Group Report- Appendixes  August 2010.   Appendix 1 - Terms of Reference Background Ubiquitous computing has had a significant impact on the process and products of scientific research. Instruments are gathering ever-increasing volumes of data, and newcomputing techniques are enabling researchers to querydata in newways. Right nowin Canada, and in particular in UBC, the vast majorityof research data is being lost. One of the major stumbling blocks for this newe-Science is the need to find strategies for the long-term storage, preservation, and access to research data. Currently, UBC Librarydoes not have an active strategyfor curating the data produced through research carried out at the institution. However, as part of its 2009 strategic plan, UBC has made a number of commitments to research excellence including “develop[ing] a campus strategyfor making UBC research accessible in digital repositories, especiallyopen access repositories.” (Universityof British Columbia President's Office, 2009) Purpose The Library’s eStrategycommittee has established a working group to advise the eStrategy committee regarding research data curation at UBC, in particular as it relates to the potential role for UBC Library. Members: · Eugene Barsky(co-lead) · Tom Brittnacher · Hilde Colenbrander · Margaret Friesen · KathyHornby · Paul Joseph · JoyKirchner · MaryLuebbe · Joerg Messer · Katherine Miller · Shawnna Parlongo · Sarah Romkey · Barbara Sobol · Bronwen Sprout (co-lead) · Lea Starr · Rudi Traichel · Erwin Wodarczak Responsibilities To consider and advise the eStrategycommittee on: 2  o whether UBC Libraryneeds to be involved in the process of data management o current data management practices and initiatives in UBC (Vancouver and Okanagan) – UBC IT, UBC Office of Research Services and others o current data initiatives in comparable university libraries and identifypotential partners on campus and in other Canadian and international libraries for data management initiatives o possible pilot project for data management and to provide a report to the eStrategycommittee on its findings and recommendations as above. Constitution The group’s co-chairs will be Eugene Barskyand Bronwen Sprout. Membership will be made up of interested librarians and librarystaff. Frequency of meetings The group will meet once a month for approximatelysix months. 3  Appendix 2 - Literature Review: Afterreviewing the professional literature it is clear that academic libraries do have a role to playin curating data generated byresearchers at their various institutions. Over the years librarians have been instrumental in establishing and maintaining institutional repositories and social science data collections. While doing so, we have developed expertise in areas such as data selection, metadata standards, access and search functionality, managing the data lifecycle and digital preservation. We can and should leverage this knowledge to expand our services to include research data created byresearchers at our own institutions. The keys to success are: § surveying local researchers’ data preservation needs and practices § establishing strong partnerships with researchers to establish metadata standards and preservation best practices § ensuring that our cyberinfrastructure is both scalable and provided with long- term commitment and funding Selected Annotated Bibliography Inge Angevaare. TakingCare of Digital Collections and Data: 'Curation' and Organisational Choices for Research Libraries. Vol.19, 2009. Focuses on issues surrounding digital preservation and curation for research libraries. Stresses the fragilityof digital data and the care and expense required to preserve it long- term. Suggests three options for libraries: Store data on existing cyber infrastructure and hope for the best; find a third partyto host their data; build a proper digital archive in-house. Suggests that research libraries mayrevive their perceived value bycurating researchers' digital datasets themselves. Ends with a do & don't list for research libraries' digital initiatives. Do: surveyresearchers' needs and practices; ensure your cyber infrastructure is scalable; establish selection criteria; provide researchers with metadata instruction and support; focus on access. Don’t: Store data on local computers or outdated storage devices; neglect to back up data frequently; entrust your institution's data to a custodian without a proven track record. Christine Borgman, Jillian Wallis and Noel Enyedy. Little Science Confronts the Data Deluge: Habitat Ecology, Embedded Sensor Networks, and Digital Libraries. Vol.7.Springer Science & Business Media B.V, 2007. Addresses the "data deluge" threatening to overwhelm researchers in the sciences. Focuses on habitat ecology, an example of “little science.” Details the value added services that libraries can provide to a data repositoryprogram aimed at little science: Establishing metadata standards, determining a standard communication framework for the exchange of metadata, developing technical tools to support/create metadata and providing the means to preserve metadata. Provides extensive analysis of the role of data in science and the ways in which researchers do or could re-use data. Conclusions: Scientists would ideally like to share 4  databut lack tools for data analysis; lack metadata standards; and most importantly, lack career incentives to preserve and share their data. Rowan Brownlee. Research Data and RepositoryMetadata: Policyand Technical Issues at the University of SydneyLibrary. Vol.47, 2009. The author describes recent efforts bythe libraryat the Universityof Sydneyto establish metadata standards for its research repository. Currentlythe repositoryhouses three small, but significant data collections from the departments of geoscience, arts & archeologyand biological sciences. Data are typicallyheld in databases such as filemaker, MySQL or Excel. The primarychallenge at USL has been to develop metadata standards that are sustainable and support findability, while also retaining the "granularity" of the depositor's supplied metadata. Details the institution's choice to applyDublin core metadata standards to its data deposits - which have thus far been mapped from the depositor's supplied metadata bythe appropriate subject librarian. The author suggests that this activitywould ideallybe mapped bya cataloguer-subject librarian-researcher team. G. Sayeed Choudhury. Case StudyinData Curationat Johns Hopkins University. Vol.57, 2008. Discusses institutional repositories (IR) and the gap between their promise and their reality- particularlyfocusing on the "crisis" in scholarlycommunication that onlyseems to exist for librarians. Posits that researchers are generallysatisfied with the current scholarlypublishing models and therefore have little motivation to support IRs and/or the open access movement. Suggests that IRs maybe revitalized as data curation centers but concedes that the scale and complexityof "big science" renders its contributions unsuitable for a local repository. The author suggests IRs might instead choose to partner with like-minded institutions to share storage, preservation and dissemination services and/or limit submissions to select portions of a large project. Finishes bystressing the importance of using open source software platforms. TracyGabridge. The Last Mile: LiaisonRoles inCuratingScience and EngineeringResearch Data. Association of Research Libraries, 2009. Focuses on the librarian's role in creating valuable data services; soliciting contributions to librarydigital repositories and curating data deposits over the long term. Points out the expertise that alreadyexists in terms of social science data services - though to date this has largelybeen comprised of datasets from external agencies. Argues that librarians could leverage this knowledge to expand services to include faculty-generated data. Identifies key skills librarians have to applyto curation duties: Selection, planning data life cycle; teaching preservation best-practices; collecting and disseminating the data appropriately; help establish/maintain metadata and data standards. Anna Gold. Cyberinfrastructure, Data, and Libraries, Part 2. Vol.13, 2007. Describes traditional types of data curated bydata librarians: Social science, GIS and bioinformatic data. Suggests that librarians can leverage this experience and expertise to extend services to include scientific and technical data. Discusses potential roles that 5  librarianscould playin supporting e-science: Selecting & acquiring datasets; offering long- term data preservation services; and helping develop best-practices to assure data longevity. stresses that librarians' skill-set will require that theylimit their reach to small and medium- size research and that subject specialization maybe required bylibrarians to assist researchers most effectively. Cautions that data services cannot expand without long-term commitment to staffing and funding. Stuart MacDonald and Luis Martinez Uribe. Libraries in the ConvergingWorlds of OpenData, E- Research, and Web 2.0. Vol.32.Information TodayInc, 2008. Considers the ways in which the ever increasing amounts of data being created will affect both the research lifecycle and the librarians who will be expected to support this e-research. Highlights the benefits of open access to research data and details tools, technologies and organizations which were keyto the open data movement at the time this article was published. Discusses possible strategies that librarians might employto support e-research, such as providing data curation services via institutional repositories. Michael Witt. Institutional Repositories and Research Data Curation ina Distributed Environment. Vol.57, 2008. Describes the challenges that arise from the "data deluge" generated bye-science: Masses of unorganized data are created, used once, preserved locally if at all and eventually lost or discarded. Discusses traditional roles librarians have played in disseminating research and suggests roles librarians might have in curating research data, such as creating metadata, providing access to datasets, appraisal of submissions, and providing the means to preserve and disseminate submissions. Describes Purdue Libraries' (PL) digital repositories for archives and documents and explains howknowledge developed over the course of establishing those units informed the development of PL's distributed data curation center. Outlines current research/studyat Purdue aimed at improving the center's services and ensuring its long-term sustainability. 6  Appendix 3 - Data Management Questions for Researchers on Campus  Name of Respondent:_______________________________________________________ Position:_________________________________________________________________ Dept/Admin. Unit: _________________________________________________________ Contact information:________________________________________________________ UBC Libraryhas convened a working group to examine the potential role that the library could playin data management and preservation. We are at an exploratorystage of our work and have developed a short questionnaire to determine what is alreadybeing done at UBC, what areas researchers are concerned with, and where we might focus our efforts. Your input will inform the development of a wider surveyof researchers at UBC and will impact the directions that the Librarytakes regarding data management. Note: The following questions refer to research data, not administrative data: 1. What data do you store in your (faculty/department/administrative unit)? 2. What format is the data in? 3. Who has access to this data? 7   4. If the data is shared, bywhat method is the data shared? Follow-up exploratoryquestions: 5. Under what conditions (if any) would you be willing to share your data? (For example, the researcher wants to be notified each time there is a request for the data; the data can onlybe shared within UBC; conversely, the data can be shared beyond the university, etc.) 6. What assistance/tools/funding would you need to prepare your data for use byother researchers (including preparing metadata)? 7. Anyother comments? 8  Appendix 4 -  UBC Librarys Data Management Questions for IT and other Campus Managers Name of Respondent:________________________________________________________ Position:_________________________________________________________________ Dept/Admin. Unit: _________________________________________________________ Contact information:_________________________________________________________ UBC Libraryhas convened a working group to examine the potential role that the Library could playin data management and preservation. We are at an exploratorystage of our work and have developed a short questionnaire to determine what is alreadybeing done at UBC, what areas researchers and administrators are concerned about, and where we might focus our efforts. Your input will inform the development of a wider surveyof researchers at UBC and will impact the directions that the Librarytakes regarding data management. Note: The following questions refer to research data, not administrative data: 1. What data do you store in your faculty, department or administrative unit? 2. What format is the data in? 3. Who has access to this data? 4. If the data is shared, what is the method used to share it? Does your method allow for the re-use and re-purposing of data byothers? 5. Do you receive requests for data management and support that you cannot meet? If yes, what are the obstacles? 9  6. Howmuch storage is required for the data in your unit? Howis it backed up? 7. Howmuch staff time is required to support data infrastructure and management in your unit? 8. Howis the data currentlybeing described/catalogued and what standards, if any, are being used to do this? Who has access to the descriptive information? 9. Are you aware of anydata repositories outside of your faculty, department or administrative unit (either on or off campus) that are, or potentiallycould be, used by researchers in your unit? If yes, which one(s)? 10. What data management services would you suggest should be offered centrally?   1 The University of British Columbia Library Data Management Working Group Report August 2010 Executive summary In January2010, the Libraryformed a working group to consider and advise the eStrategy Committee on whether UBC Libraryshould be involved in the process of data management, to investigate current data management practices and initiatives at UBC, to research current data initiatives in comparable university libraries and identifypotential partners for data management initiatives. The Terms of Reference for the group is appended in Appendix 1. The group was co-led byEugene Barskyand Bronwen Sprout, and reported to the eStrategy committee. Nineteen libraryprofessionals volunteered to serve in this verydiverse group, demonstrating a strong interest amongst UBC archivists and librarians about this topic. Group members conducted a literature reviewand interviewed UBC researchers and administrators to find out about current data management practices at UBC. The group’s findings demonstrate that there is an institutional need for data management both practicallyand as mandated bygranting agencies, and suggest that the Librarycan position itself as a bridge between researchers and users byproviding metadata and infrastructure support. In terms of the Library’s efforts, data management aligns well with other digital initiatives such as cIRcle and the newDigital Initiatives unit, as outlined in the Library’s strategic plan in the areas of Accelerate Research and Engage with Community. As our literature reviewrevealed, data management is a relativelynewarea for academic libraries. There is an opportunityto establish ourselves as leaders in Canada in data management, and to collaborate nationallyand internationallyon data management issues. In light of the significant opportunities for the Libraryin this area, the group recommends that a limited term, internal position be created for a data management librarian to continue the group’s work and to further university-wide engagement.  Background Ubiquitous computing has had a significant impact on the process and products of scientific research. Instruments are gathering ever-increasing volumes of data, and newcomputing techniques are enabling researchers to querydata in newways. Right nowin Canada, and in particular in UBC, the vast majorityof research data is not being preserved. Currently, UBC Librarydoes not have an active strategyfor curating the data produced through research carried out at the institution. However, as part of its 2009 strategic plan, UBC has made a number of commitments to research excellence including “develop[ing] a campus strategy for making UBC research accessible in digital repositories, especiallyopen access repositories.” (Universityof British Columbia President's Office, 2009) The Library’s eStrategycommittee established a working group to advise the eStrategy committee regarding research data curation at UBC, in particular as it relates to the potential role for UBC Library.   2  Literature review The group focused on researching the data management requirements for major funding agencies, major journals and publishers, comparable research institutions, and COPPUL. Major findings § Academic libraries have begun to experiment in data curation and some best practices are beginning to emerge. Keys to success include: o Surveying researchers’ data needs in advance of a curation project o Leveraging librarians’ expertise to work with researchers to establish standards for metadata and preservation o Ensuring that cyber infrastructure is scalable and funded for the future o Considering where the data is best hosted. Manyof our colleagues are using their IRs to store and curate their datasets and research. § Major funding agencies in Canada, the US, the UK and Australia all require that projects must either provide a plan for data management as part of their funding applications, or agree to abide bythe agency’s pre-existing policy. § Most journal publishers do not yet have a policyin place for rawdata preservation – regarding this as the responsibilityof authors. An exception is the biomedical publishers who are requiring rawdata to be made available for other users. § COPPUL is just beginning to examine this issue via the Digital Curation and Preservation Task Group. Bronwen Sprout is UBC Library’s representative to the COPPUL Digital Preservation group. Moreover, librarians possess a unique set of skills for metadata and data organization; however, we might lack other skills, e.g. computer science or infrastructure building background which will need to be acquired through academic or on-the-job training. The complete literature reviewis available in Appendix 2. Interviews with Researchers We interviewed several UBC researchers about their current practices and future needs regarding research data management. Our findings indicate that current storage and distribution capabilities varywidelybetween departments, with no consistent standards and practices in place. Some researchers are even utilizing resources at other universities to satisfytheir needs for data computation and storage, as their needs are not being met through UBC. In none of these cases were metadata files being created; researchers typically do not have the time or inclination to produce such documentation on their own. One professor noted that graduate students would benefit greatlyfrom a storage and management program for their research data. Such a program could also be used as a data management teaching tool for students. The set of questions posed to researchers is available in Appendix 3.  3 Interviews with IT managers Due to the decentralized nature of IT management at UBC, a varietyof administrators were approached to gain insight into common practices and potential areas that the librarymight address regarding data management. A presentation was made to UBC IT Managers, a formal committee representing disciplines across campus. A representative from the Office of Planning and Institutional Research and UBC Okanagan IT Research staff were interviewed. Preliminaryfindings include the following: data management at UBC is conducted at a departmental, research group, or individual researcher level; file management, metadata creation, and cataloguing standards are irregularlyapplied and there is appetite for the libraryto provide leadership in this area; privacyand securityare major concerns for researchers and administrators with open sharing of data a tertiaryconcern; consortia/collaborative endeavours for data storage, such as WestGrid, exist and could provide either a model or the foundational infrastructure to build data services upon. The set of questions posed to IT managers is available in Appendix 4. Recommendations We recommend that the Library, working with the Director, LibraryDigital Initiatives, create a short-term (secondment) leadership opportunityfor a UBC librarian or archivist to further explore the issues and followup on the data management group. We recommend either a full time position for 6 months or 0.5 part time position for 12 months that will be backfilled. Specifically, we recommend this position be responsible for: § Establishment of a pilot project, complete with a project charter to test the waters for a Libraryrole in data management. We recommend a discipline-based approach to a pilot project to consider the differing data needs/practices amongst the various disciplines at UBC. § Creation of training programs and/or studygroups to enhance the knowledge base of a wider pool of staff members about metadata and preservation best practice § Development of a toolkit for researchers to assist in their efforts at data management § Continue to develop the connections with IT managers and explore opportunities with other Universityadministrators to identifya possible role for the libraryin data management. § Structurally, the newlyformed UBC-wide ScholarlyCommunications Steering Committee mayprovide a venue for these activities. There is also a natural connection with cIRcle and the Digital Initiatives unit.

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
United States 4 0
Russia 2 0
China 2 39
Spain 1 0
France 1 0
United Kingdom 1 0
City Views Downloads
Lewes 3 0
Unknown 3 4
Beijing 2 0
Barcelona 1 0
London 1 0
Ashburn 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}

Share

Share to:

Comment

Related Items