This is the first of a series of 16 draft proposals developed by the ICANN Strategy Panel on Multistakeholder Innovation in conjunction with the Governance Lab @ NYU for how to design an effective, legitimate and evolving 21st century Internet Corporation for Assigned Names & Numbers (ICANN). The MSI Panel has been specifically chartered by the ICANN President & CEO to:
- Propose new models for international engagement, consensus-driven policymaking and institutional structures to support such enhanced functions; and
- Design processes, tools and platforms that enable the global ICANN community to engage in these new forms of participatory decision-making.
Please share your comments/reactions/questions on this proposal in the comments section of this post or via the line-by-line annotation plug-in.
From Principle to Practice
For ICANN to be an effective institution operating in the 21st century it needs to be smart. This means it needs access to the best possible ideas in forms and formats that are useful and relevant to the decision at hand from sources inside and outside the institution. ICANN should, therefore, together with the other Internet governance organizations, adapt expert networking technologies for identifying and making searchable technical expertise worldwide, where expertise is defined broadly to include not only credentials (such as formal engineering and computer science degrees), but also technical experience and skills (e.g., as evidenced by GitHub commits or answers on Q&A sites), as well as interests (e.g., as measured in response to questions on Quizz.us).
What Are Expert Networks?
Expert networks are platforms or communities that provide individuals with the tools for representing information about their expertise (e.g., “scholarly works, research interests, and organizational relationships” ) and for enabling easy search of that expert information. Instead of looking for answers from an undefined crowd (crowdsourcing widely), expert networking seeks to “involve experts on particular issues and problems distributed anywhere in the world”  (crowdsourcing wisely). For instance, expert networking tools such as VIVO – an interdisciplinary network of research scientists  – “when installed and populated with researcher interests, activities, and accomplishments … [enable] the discovery of research and scholarship across disciplines at that institution and beyond.” 
Such networks are being deployed in a variety of fields and contexts, from academic and research networks like VIVO to industry-specific networks (e.g., for data scientists or for advertising and marketing creatives) to skills-based collaboration communities (e.g., TopCoder for computer developers).
Why Expert Networking at ICANN and Across the I* Organizations?
Foundational to ICANN as an institution is its open nature of welcoming broad-based input; ICANN appeals to the global community, allowing anyone to join a working group or participate at ICANN’s triannual global meetings. But ensuring the stability, security and operability of the DNS includes multi-faceted and often highly technical work requiring specialized knowledge and skills. And some have assessed that ICANN’s current working group (WG) model for developing consensus around how to solve such complex problems “often appears to be lacking – especially when dealing with complex issues compounded by widely disparate points of view and/or strongly held financial interests in particular outcomes.”  Moreover, many issues at the forefront of the Internet governance debate today are “new” – previously unaddressed or nonexistent – and lack the governance mechanisms for finding solutions (e.g., privacy). Many issues are intractable or contain extremely nuanced policy and technical implications. Finally, there are no institutional or cross-institutional frameworks for addressing Internet governance issues comprehensively.
In such a case, for ICANN to be smart and thus effective, it should use a distributed yet coordinated approach to tap expertise for new and complex problem-solving. Specifically, leveraging expert networks has potential to:
- Increase diversity, reduce redundant participation and remove vested interests from stakeholder groups and working groups at ICANN. 
- Move ICANN from a representation-based to expertise-based organization. In fact, leveraging expert networking technologies would enable ICANN to organize its participants topically rather than by constituencies that are defined by interest. This could help streamline and depoliticize the solution development process and avoid redundant work.
- Inspire and incentivize collaboration within and across silod ICANN structures.
- Save time and resources by crowdsourcing technical know-how wisely rather than widely. This is especially important given the complex and sometimes opaque nature of ICANN’s work and the often-times slow-moving policy development processes, which serve as barriers to meaningful participation globally.
- Provide ICANN a means of locating needed, but previously dormant, specialized expertise to solve problems facing the DNS.
- Empower netizen-experts with the willingness to participate in ICANN decisionmaking to engage. Experimenting with incentives – e.g., reputation points, prizes or badges – will ensure ICANN finds the best means for reaching those most willing to bring their expertise to bear for ICANN.
- Help match those with the skills and knowledge to bear to particular problems and needs – from figuring out how to mitigate name collisions to how to support internationalized domain name variants within the DNS to how to best balance data privacy and data security in configuring the next generation system of Whois.
ICANN can use expert discovery and networking tools to better target requests for participation in all stages of ICANN decisionmaking. This could be especially useful for helping to staff working groups  (in the solution development stage) and review teams (in the evaluation and review stage). The use of technologies that enable real-time translation could also help motivate participation from regions beyond North America and Europe. 
Implementation Within ICANN
While we believe using expert networking technologies would help ICANN become a truly smart and thus effective institution – we believe that testing this hypothesis is vital. Moving this proposal from principle to practice is key. With that said, here are some initial steps ICANN could take to begin piloting this proposal:
Phase 1: Hone Research & Assessment Agenda
- Here are some initial research questions to study and test that ICANN should review and expand on given particular organizational and community needs:
- What kinds of expertise are most helpful to identify?
- Where can ICANN find people with the kinds of skills and knowledge and experiences identified above?
- What are the ways in which the needed expertise can be represented and collected? What can ICANN learn from the following:
- Reputation-based systems (e.g., Linkedin Recommendations); credential-based systems (e.g., ResearchGate); experience-based systems (e.g., StackOverflow); self-reported systems (e.g., Catchfire).
- What are different ways ICANN could target calls to participate once needed expertise has been identified?
- What kinds of incentives for participation make sense? Which may work best depending on the problem at hand?
- How does the level of expertise impact people’s willingness to collaborate?
Phase 2: Create or Build-out Ontology
- ICANN should create a standardized ontology for describing skills and categories of expertise needed at ICANN and across the Internet governance ecosystem. ICANN could start by building out a VIVO-like ontology  (described briefly below).
- To capture this information, ICANN could begin by developing different versions of questionnaires to determine the best ways to accurately capture expertise data. These should be distributed to all currently active ICANN community members as well as to other Internet governance organizations, community groups and listservs for self-reporting. The budding 1Net community is another potential data source, as is the ICANN Labs Peer Advisory Network.
Phase 3: Create Framework for Absorbing Expert Input
- Identify which ICANN structures or groups would be best to pilot expert networking technologies. Reach out to these groups to discuss where in their work leveraging expert networks will be beneficial and get agreements to run parallel processes alongside current practices for testing.
- Determine how and where ICANN will use expert input when identified issues are cross-institutional or interdisciplinary.
Phase 4: Operationalize/Pilot
- Run parallel pilots coordinated by different internal groups and using different techniques for identifying and motivating participation to test what works and to enable analysis and comparisons.
- As pilots progress, ICANN should explore strategies for creating a linked data infrastructure, to connect and make searchable the skills and expertise of individuals across all Internet governance organizations.
Potentially Relevant Expert Networks/Communities – ICANN Experts in Hiding?
ICANN should similarly explore integration with other popular international, regional and local sources of relevant expertise as well as open datasets on publications and grants. This would help ICANN test whether tapping into existing databases is effective in supplementing and vetting self-reported data. It also has potential to help locate currently non-active individuals who may have the requisite skills and interests that could be brought to bear for ICANN.
Some potentially relevant networks and communities include the below. These examples also provide certain functionalities that ICANN should study and possibly emulate in creation of any independent network.
- Epernicus – An expert network and knowledge-sharing platform for science researchers. Epernicus captures recently added expertise, provides interconnected communities for different disciplines and has related software that allows research organizations to create their own internal expert network.
- Kaggle – An expert network and competition platform for data scientists. Kaggle incorporates user tiers to highlight engagement milestones (rather than more granular points and leaderboard functionalities). Something similar could be applied at ICANN – perhaps a tier for newcomers, explorers, researchers and leaders.  Kaggle also provides users with goal-based incentives.
- Stack Exchange – A question and answer forum to get expert advice on a diversity of topics. Stack Exchange gives users the ability to upvote questions and answers, provides distinct, topic-based Q&A sites within the larger Stack Exchange framework, including the computer programming Stack Overflow community (note that a simple search for “ICANN” on Stack Exchange brings up thousands of results of Q&A threads related to ICANN and ICANN’s work).
- Technical Expert Network (TEN) – A platform for finding and contracting international technical expert consultants. TEN allows users to tap experts in the network for different technical skills: interviews, surveys, moderated discussions, consulting, proposals and collaboration, and recruiting. It also provides the ability to articulate preferred types of projects.
- TopCoder – A programming expert network, collaboration engine and contest platform. The network provides users a “reputation score” that is listed on their profile page, alongside a set of various statistics regarding their participation in TopCoder challenges. The network also uses competitions and tournaments to drive and incentivize engagement.
- VIVO – An interdisciplinary network of research scientists. VIVO allows users to tag their research areas, publications and research communities, and provides users the ability to browse expertise by People, Organization, Publications or Research. The network also provides linked, graphical representations of co-author and partnership networks and its creators are developing a central VIVO interface linking organization-specific implementations using semantic web methodologies  and an open ontology.
Case Studies – What’s Worked in Practice?
ICANN could also learn from the following case studies, whereby expert networking technologies have been deployed to help solve real and complex challenges in a variety of public interest contexts.
- Kaggle & 311 – The data scientist network has been used successfully to convene a challenge to “quantify and predict how people will react to a specific 311 issue,” taking into account factors such as urgency, citizen priority and location.
- NASA – NASA has successfully used Innocentive – an online platform that broadcasts carefully defined problems to a community of experts and researchers – to find a solution for conducting “non-invasive measurement of intra-cranial pressure,” a physiological condition that results from space-travel.
- Peer to Patent – A history initiative of the U.S. Patent & Trademark Office (USPTO), Peer to Patent leverages the use of citizen experts in examining and vetting patent applications. It is an online system that connects an open network of scientists and engineers with the aim of improving the quality of patents issued by the USPTO. In its initial pilot in 2009, community experts supplied information and research based on their relevant area of expertise and patent examiners retained final decision-making authority to grant or deny an application based on the legal requirements. Peer to Patent pilots also exist in Australia, Japan, South Korea and the United Kingdom.
- U.S. Federal Drug Administration – The FDA is currently experimenting with using VIVO to help the agency more quickly identify those with technical know-how and experience who could help determine whether a new medical device is safe.
Open Questions – Help Bring This Proposal Closer to Implementation?
- What institutional and cultural barriers – such as entrenched processes – could pose challenges to implementation?
- What techniques could we use to measure the impact of expert networking against existing models of decision-making at ICANN?
- At what stage of the decisionmaking process or policy development process would using expert networks be the most beneficial for ICANN? For developing recommendations? For developing implementation strategy?
- What types of expertise, if any, are currently lacking within ICANN?
- What physical or organizational communities already exist that comprise individuals with relevant expertise for ICANN? For example, ISOC, WSIS, IGF, etc.
- What topics or structures in ICANN lend themselves best to using expert networking?
- How can we be sure that expert input from any region or in any language can be absorbed into ICANN decision-making?
- What do current ICANN community members believe are the greatest motivations or incentives for participating in ICANN decision-making?
- How can expert networks for I*s be built-out to include the kinds of peripheral expertise that are not necessarily obvious? For example, what if a systems biologist has a better idea for how to organize the distributed DNS than a systems engineer?
- What would the framework of accountability for decisions being made by experts look like?
1. Börner, Katy, Michael Conlon, Jon Corson-Rikert, and Ying Ding. “VIVO: A Semantic Approach to Scholarly Networking and Discovery.” Synthesis Lectures on the Semantic Web: Theory and Technology 2, no. 1 (October 17, 2012): 1–178.
2. Dutton, William H. Networking Distributed Public Expertise: Strategies for Citizen Sourcing Advice to Government. One of a Series of Occasional Papers in Science and Technology Policy, Science and Technology Policy Institute, Institute for Defense Analyses, February 23, 2011.
3. “About.” VIVO.org.
5. Accountability and Transparency Review Team 2. “Report of Draft Recommendations for Public Comment.” ICANN.org. December 31, 2013 at 61.
6. Ibid. at 62 (highlighting public comments on the policy development process that called for “The need for wider participation and cross-community interactions. . . . [and] The need for participation by groups without business-related incentives for participation.”).
7. For a full review of the qualitative and quantitative current state of participation in ICANN working groups, see ibid. at 31-48.
9. See Bibliometrics by Librarians and Information Professionals National Institutes of Health (NIH) Library. “Professional Network Analysis and Expertise Mining at FDA” at slide 9. January 31, 2013.
10. These example categories were developed based on input from comments to the Panel’s engagement platform.
11. W3C. “Semantic Web Activity.” June 19, 2013.
12. Bibliometrics by Librarians and Information Professionals National Institutes of Health (NIH) Library.
“Professional Network Analysis and Expertise Mining at FDA” at slide 9. January 31, 2013.