Semi-Permeable Boundaries Among Institutions: Facilitating the Flow of Between Service Settings

1
Semi-Permeable Boundaries Among Institutions
Facilitating the Flow of Between Service
Settings

• Libbie Stephenson, ISSR,
• University of California, Los Angeles
• libbie_at_ucla.edu

Jon Stiles, UC DATA, University of California,
Berkeley jons_at_berkeley.edu
2

• Semi-Permeable WHAT?

Starting Point Data support occurs in a variety
of institutional settings. Those settings may
and probably do differ in terms of mission,
clientele, resources and focus. These
differences can be a strength, in that services
can be tailored to local context and needs, but
can also be isolating and unnecessarily limit
services to users. Question How do some
services wind up in particular settings, how does
that affect end use, and how can institutions
work to bridge barriers that limit end use?
1/6/2015
2
3
What we plan to cover
? Local History Development of secondary data
support at UCLA and Berkeley ? 1960s, 1970s,
1980s, and beyond ? Changing roles ?
technology, expertise, mission, resources, turf,
AND data producers ? internal,
inter-organizational, external factors ?
Models of collaboration ? Cross-unit
collaboration and challenges
1/6/2015
Stephenson/Stiles 08/06/2008
3
4
Data Services is about relations between
producers and intermediaries intermediaries and
data intermediaries and other intermediaries
intermediaries and users and users and data
01010101010 01 000 11
Intermediaries
01010101010 01 000 11
Producers
Users
01010101010 01 000 11
Environment
1/6/2015
4
5
Evolution of Data Services Landscape 1960s

• General Environment
• Increasing use of surveys
• Technology supportive of machine-readable data
  expensive, barriers to entry
• Producers
• Key institutional players (Census Bureau, large
  survey/research organizations, NSF/Funders).
• Users
• More interest and use (demand)
• Fairly specialized community, content focused
• Local Environment very important
• Lateral Institutions
• Activities bundled not easily broken up
• Data
• Largely survey based. Dynamic and developing
  environment.

1/6/2015
5
6
UCDATA and ISSR 1960s

• Content focused collections
• Strong links with researchers with
  content/methodological knowledge
• In-house consumption, small clientele
• Training an important component
• Technology
• Berkeley
• ? International Data Library Reference
  Service (IDLRS -1962)
• ? NSF Funds active outreach /acquisition (
  1964)
• ? CSSDAUCLA
• ? Political Behavior Archive (PBA-1961)
• ? Library receives NSF funding for CIS
• ? Survey Research Center Archival Data
  Library (1964)

7
Evolution of Data Services Landscape 1970s- 80s

• General Environment
• ? Thin Edge of the Wedge 1970 STFs in
  Depository Libraries
• ? Continued development of computing/storage
  technology
• ? Bibliographic control through MARC
  descriptive cataloging
• ? IASSIST formed
• ICPSR and national archives gain prominence
• Archives
• ? Unbundling of support components
• ? Complementary activities at Libraries,
  archives, computing centers
• Influence on data producers to provide better
  documentation

1/6/2015
7
8
Two different avenues of development UCDATA and
ISSR1970s 1980s

• UCDATA
• ? Census Service Facility broad dissemination
  and services
• ? Increased focus on State Data, Field Poll
  Collection
• ? Records in library catalog begin in
  mid-1970s
• ? Census State Data Center network 1979 ?
  Strong Census-related development through 1980s
  
• ISSR
• ? Library acquires 1970 Census limited
  do-it-yourself service
• ? ISSR established data archivist hired
  census transferred
• ? ISSR Data Archive is de facto central campus
  unit
• ? Extensive campaign to preserve
  faculty-generated data

1/6/2015
8
9
Evolution of Data Services Landscape 1990s to

• ? Increase in collaboration and joint projects
• ? Over-lap of clientele, data formats and
  services
• ? Variety in organizational operating models for
  libraries
• and archives
• ? New cohort of professionals have increased
• technological skills
• ? Potential of opportunities using Internet seems
  endless

10
Two different avenues of development UCDATA and
ISSR

• Berkeley
• ? Mission expanded and name change in 1990s
• ? Collaborative projects with Library others
• ? Library and archive develop services in
  parallel
• UCLA
• ? Data services provided by ISSR
• ? Involvement in IASSIST
• ? ISMF developed join IFDO

11
What does history tell us?(One reading)
Secondary Data Mission involves (at least) 4 sets
of relations Producer relations User
relations Institutional (Local -Lateral)
Relations Data Relations Change at
institutional levels emerges from Internal
factors (expertise, funding, interest, etc) Other
institutions (archives, producers, private
sector) Big environment (technology, user
demands)
1/6/2015
Stephenson/Stiles 08/06/2008
11
12
Part II
Changing Roles of Practitioners
Operational models
1/6/2015
Stephenson/Stiles 08/06/2008
12
13
Changing Roles Who provides the services?

• ? Data discovery
• ? Statistical advice
• ? Technical assistance
• ? Data visualization support
• ? Access to files, documentation and tools
• ? Cataloging and metadata
• ? Data curation and preservation
• ? Physical storage space
• ? Virtual storage space
• ? Staff, training, programming, licensing,
  funding

users
producers
infrastructure
1/6/2015
13
14
Changing operational models
Levels
Single ? Local multi-unit ? Federated ? Consortial
Independent ? membership/consortial ? national
mandate (heirarchical)
Structures
Modes
Collaboration ? Separation ? Hierarchy
Players
Amazon, Google and the individual data creator
15
Part IV
Barriers Tools
1/6/2015
Stephenson/Stiles 08/06/2008
15
16
Pros and cons to models

• Separation
• Collaboration
• Hierarchical

1/6/2015
Stephenson/Stiles 08/06/2008
16
17
Collaborationbarriers and tools

Constructive tools ? SWOT? Competing Values
Framework
Barriers ? Institutional culture ? Turf ?
Political power plays ? Financial constraints ?
Technological capacity ? Workforce limitations
1/6/2015
17
18
Multiple-points-of-access-model

• Goal provide best services and resources
  possible ? Develop shared expertise across
  units ? Collaborative collection building
• ? Develop access and data use tools
• ? Provide support for data visualization
• ? Use metadata standards to enhance data
  discovery

1/6/2015
18
19
Summary and conclusions

• Models for services and support are increasingly
  complex
• Politics, turf, finances require skill and
  temerity to navigate stakes are higher
• Players do not possess common skill set, or
  common vocabulary nor common goals/objectives
• Payoffs are high extended scope, projects

1/6/2015
19