My work includes community-sourcing and scholar-sourcing.* This includes collaboration at scale, with different numbers of people involved (generally dozens – hundreds) and with different types and levels of activity and contribution for all involved.
However, I don’t say that my work includes crowdsourcing. Crowdsourcing is a familiar term for many folks, but it seems to be familiar without much definition or nuance. Wikipedia’s entry for crowdsourcing includes many references to the community, which can be specific or open. Wikipedia’s entry also notes key potential problems with quality and ethical concerns for contributors for crowdsourcing. In “Pixel Dust: Illusions of Innovation in Scholarly Publishing“, Johanna Drucker explains how tight constraints can prevent problems (incurring enormous overhead in managing contributions and quality, which can result in a net loss), and the difficulty of designing and implementing these controls:
A handful of projects, like the Jeremy Bentham transcription, or the New York Public Library’s menu decipherment, were expertly designed, highly constrained, and made effective use of contributions by the public. The redesign of scholarship to allow for participation is an enormous undertaking, not yet much beyond prototypes, none of which have yet proved fully viable except the wiki.
I’d like to emphasize Drucker’s spot-on statement: “The redesign of scholarship to allow for participation is an enormous undertaking, not yet much beyond prototypes.” The redesign of scholarship for larger participation that is beneficial requires positive return on investment (so, the value of the contributions is higher than the processing and validation overhead) and positive impact for the community (the system-process must support necessary ethical standards for contributors).
Drucker’s comments are on a different scale and type, focusing in this quote on crowdsourcing from the entire online world. By changing the framing to be collaboration at scale to scholar-sourcing and community-sourcing, there are successful models of practice for scholarly project participation that may be informative for the larger scale collaboration with crowdsourcing.
While there’s no Wikipedia entry for community-sourcing, the Community source entry includes an explanation of Community Source as a type or model of Open Source software development, often used in academia, where institutions commit resources together. The entry includes a longer quote which is excerpted below. While the focus of the quote is on software development, much of the process of community-sourcing is directly applicable to scholarly work and the successful design for participation in scholarship.
The Community Source Model in Higher Education (Excerpt from Wheeler, 2007 in References)
The Community Source Model is a hybrid model that blends elements of directed development, in the classic sense of an organization employing staff and resources to work on a project, and the openness of traditional open-source projects like Apache. […] The distinguishing feature of the Community Source Model is that many of the investments of developers’ time, design, and project governance come from institutional contributions by colleges, universities, and some commercial firms rather than from individuals. […] Community Source Model projects generally operate as follows. Several institutions realize they are trying to solve a similar problem—need for a research administration system is a recent example. After some discussions and resulting agreement on project objectives, timelines, and philosophy, the institutions pool their resources under a project board of institutional leaders.
The community source model has commitments at the institutional level, with participants then coming from the institutions. For scholarship, community-sourcing can include contributions from institutions, groups, and individuals, with different supports needed for these different communities.
In my work on building socio-technical systems (people, policies, technologies) to support scholarly work (including all aspects of research, teaching, and service), I increasingly focus on building community and connecting communities, or what can also be called the “human infrastructure.” I’m very lucky to be able to focus on the human-community needs. I’m not fighting with a bad technology or bad conceptual model or wrong-scaled process. I’m extremely lucky to have great technical collaborators and great technologies like the SobekCM Open Source software for digital libraries, data curation, and digital scholarship. SobekCM developed in a community source process, beginning with communities from GLAMs (galleries, libraries, archives, and museums) with many explicit and formalized commitments for different grant projects and programmatic needs supporting the software development over time (SobekCM presentation including these communities). In addition to support for the technical aspects with the software, the community provided user testing, feedback, definitions of need and problem statements, and expert consultation for how to make the software work for different critical user needs (technologies, patrons, public users from around the world, contributor community users, scholars, etc.).
The GLAM communities for SobekCM focused first on concerns of preservation and access. The GLAM communities needed more life to the system even to support just preservation and access because the system had to connect with their internal user needs and processes, as well as institutional needs. The GLAM communities required support for user roles and permissions for:
- institution-level: all institutional materials and collections, additional aggregation-grouping levels within and below the top institution-level
- types of activities and levels of action for each user
- roles specific to particular collections (as with curators for a single collection who should have all rights to that collection, but nothing more)
- roles at the item-level for contributors working and approved for an item only or only the items they submit and digitally curate in the system (e.g., used for each scholar in an institutional repository)
- roles for managing rights to specific collections (again, as with curators who should control access for assistants and volunteers)
- roles for managing all of the users for the institution; used as the overall group-level user manager (when it’s best to have this done by one person or group) and/or more of a coordinator for the higher level user roles (when a more distributed structure is best, as often the case for large institutions with many curators)
- roles for full system administrators with all technical backend controls
SobekCM supports these user roles and functions at the individual, group, and institutional level (and meta and sub groups and institutional levels) provides the infrastructure to support community-sourcing and scholar-sourcing activities. For instance, some institutions have leveraged the technical infrastructure to engage with scholars on known problems, like limited metadata (or item information) for materials in their collections. This is the opening to a conversation on shared needs, goals, and concerns. Rather than the crowdsourcing model applied to a smaller scholarly “crowd”, the process itself changes in terms of collaboration for collaboration at scale. The process of one group engaging with another on a known, shared problem then opens the discussion on needs, problems, and goals from another perspective. This model of practice has supported positive results for a myriad of small needs, small projects, ongoing activities and led to the creation of rich scholarly resources (e.g., Haiti: An Island Luminous, About Face), new types of scholarly practices (e.g., dLOC and a DOCC, Distributed Online Collaborative Course, on Panama and West Indian migration: Panama Silver, Asian Gold: Migration, Money, and the Making of the Modern Caribbean), new scholarly resources with new and changed communities (e.g., the Vodou Archive), and much more. My work on community building includes the work to support these as specific entities and as new models, for what they present and how they inform others. This also includes learning from scholarly best practices outside of digital technologies, to ensure the technologies support needs and are not the driving agents with technology for technology’s sake. Community-sourcing and scholar-sourcing may be productive for considering scholarly activities and crowdsourcing, even if only as they transmit existing and evolving practices from scholarly communities.
* Thanks to Mark Sullivan for suggesting the term scholar-sourcing at the Unearthing St. Augustine Meeting on January 31, 2014.