Proposal 2 for ICANN: Get Broad-Based Input by Crowdsourcing Each Stage of Decisionmaking

This is the second 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). 

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

The legitimacy of a 21st century global institution operating in the public interest depends on whether those affected by the decisions the institution makes are included in the decisionmaking process. Especially in the case of the Internet and of ICANN, to be legitimate, anyone must have easy and equitable access to help shape the policies and standards of the Internet that ICANN helps facilitate.

Using a variety of web, SMS-based and in-person participation tools, ICANN should test a wide array of alternative mechanisms for getting broad-based input in identifying and framing issues, drafting solutions, gathering relevant information to translate solutions into implementable policies, as well as commenting after the fact and participating in oversight and assessment. ICANN should use some of these tools in conducting its Public Forum at ICANN meetings, in which people can “make comments and ask questions on the main topics at each meeting directly to the Board and in front of the rest of the community.” [1]

What Do We Mean by Broad-based Input and Crowdsourcing?

Crowdsourcing traditionally refers to the act of a company or institution taking a function once performed by employees or volunteers and outsourcing it to an undefined (and usually large) network of people. [2] Crowdsourcing can be done in-person or online and serves as an important technique for broadening participation in that it involves the use of networked groups to expand the toolkit for problem-solving.

Crowdsourcing may be used in a variety of contexts and domains. For example, crowdsourcing tasks (sometimes referred to as peer production) involves spreading tasks in small bits or “chunks” of work across a crowd (e.g., Zooniverse); crowdsourcing ideas (sometimes referred to as ideation) essentially means conducting a distributed brainstorm; crowdsourcing funds (or crowdfunding) involves rallying a crowd to contribute small amounts of funds to a collective project or to help complete a goal.

Crowdsourcing can be done to broaden meaningful and global input at all stages of decisionmaking, from issue spotting to agenda setting to decisionmaking to implementation and review.

Why Does This Proposal Make Sense at ICANN?

In order to work in the global public interest ICANN must provide channels for and facilitate broad-based participation. Enabling global engagement in ICANN decisionmaking must be easy (i.e. provide accessible, legible, multilingual and low-bandwidth options) and be equitable (i.e. present fair opportunities for participation facilitated in a manner so that no one player, group or interest can dominate the decisionmaking process). At the same time that ICANN must ensure people of all nationalities and interests can join ICANN’s discussions easily and effectively.” [3] This means ICANN should proactively work to identify who in the global community is affected by its decisions and who has the expertise to bear to help solve a given challenge. Finally, to be truly inclusive, ICANN must “enable online collaboration to support distributed work for effective participation without physical attendance.” [4] At present, however, ICANN faces a variety of challenges related to these objectives, including:

  • A lack of truly global participation in working groups and as active participants in the policy development processes at ICANN;
  • A lack of metrics for resource allocation and limited data on understanding whether supporting organization and advisory committee (SO/AC) issue-framing processes are more or less effective than topically-based issue-framing [5];
  • Silo’d work departments and lack of effective communication within and across SOs and ACs [6];
  • A lack of meaningful “early engagement.” Because SO/ACs are not formally required to dialogue when producing “Issue Reports,” sometimes an SO/AC will not join an important dialogue until much later, resulting in wasted time and jeopardizing the legitimacy of outcomes. [7]
  • No formal mechanism to staff cross-community working groups [8];
  • A lack of easy-to-use mechanisms for anybody to access ICANN’s work and to participate at various stages.  For example, ICANN currently lacks useful online tools that allow for a staggered work process of people working from different places at different times. [9]

Experimenting with new techniques for getting broad input can help address these challenges. Specifically, using open, innovative and collaborative tools for reaching out to the existing community – and beyond – to help in issue framing, agenda setting, solution development, implementation and review, ICANN will be able to:

  • Create new networking channels and introduce new global players into ICANN;
  • Allow for the formation of relationships and allow participants to set agendas and collaborate on topics as they move into the “drafting” stage of decision-making;
  • Better prioritize issues and vet importance to a variety of different stakeholders by using ranking and feedback tools. This is particularly important for ICANN, where many involved have needs that are not fully defined or often vary depending on the issue;
  • Introduce new avenues for participation in stages of ICANN decision-making previously reserved for entrenched or elite participation;
  • Mitigate certain individual biases – e.g., the tendency to want to confirm prior assumptions, see non-existing patterns or be influenced by framing – by collecting and then aggregating a wide range of viewpoints on a particular issue [10];
  • Better tap a dispersed pool of expertise on subjects or issues that affect ICANN’s work (e.g., cybersecurity), but are not directly within ICANN’s remit;
  • Better facilitate a process by which relevant stakeholders can work together and talk together to solve key issues.

Implementation Within ICANN

Here are some initial crowdsourcing pilot ideas that ICANN could test over the course of the next year:

Formalize Up-Front Issue Framing by Using Open Brainstorming Tools to Identify and Rank Issues.

  • Recognizing that issues can be identified by anyone from anywhere, ICANN can use web-based tools (e.g., Google Moderator or IdeaScale) to create a structured channel for input to be used in parallel to current ICANN processes (in which SOs/ACs submit “Issue Reports” to highlight possible issues that need ICANN’s attention).
  • Such a tool should be accessible, require little bandwidth, be easy to use, and accommodate multilingual participation.
  • It should also be interactive. Participants should have the ability to not only see what others have submitted, but can also vote and comment on submissions to rank and prioritize them.
  • Open brainstorming sessions should be limited in time for efficiency, and they should be analyzed and summarized when they close.
  • In some cases ICANN should consider leveraging incentives – e.g., cash prizes or professional advancements or recognition – for participation. For example, ICANN could invite those who participated in the “brainstorming” phase of issue-framing to also participate in the “drafting” phase of a solution-proposal.

Leverage SMS-based Tools for Input

  • ICANN should pilot the use of SMS-based polling/survey tools to supplement its existing channels for input.
  • To be inclusive of a global community mandates ICANN offer low-bandwidth solutions for participation. In many places this means mobile, not broadband. ICANN could, for example, invite people to send text messages to a website that is simultaneously being used by online and physical participants of the Public Forum to see and rank what questions people have on a given topic.

Leverage Existing Multistakeholder Fora

  • The likelihood, or even possibility, of ICANN creating globally and sectorally  representative structures, without replicating existing organizations whose primary mandate is to do this, are remote. Considerable resources are already mobilized globally to bring together participants in multistakeholder forums to participate and collaborate on issues and challenges facing the Internet. While these have gone a long way in overcoming the biases in favor of one group or in exclusion from other ICT governance fora, most of these fora acknowledge the gaps in representation or participation either by one sector or dominance of another (whether government, private sector or civil society).
  • Therefore ICANN could, by allocating appropriate resources to such meetings, contribute to fulfilling their multistakeholder mandate and at the same time leverage the concentration of diverse interests,  groups, individuals and countries to contribute to processes seeking to make ICANN more representative, transparent and accountable. Specifically, ICANN could:
    • Create a more formal and continuous ICANN presence in other face-to-face multistakeholder Internet governance forums such as the Internet Governance Forum (IGF) or Internet Society (ISOC).
    • This relationship could be formalized through transparent agenda setting and consultation processes with clear lines of accountability on how the outcomes of consultations are implemented within ICANN structures at global and regional meetings of such bodies.
    • ICANN could support the participation of historically underrepresented groups at the meetings of organizations that are already building multistakeholder fora and in so doing could raise awareness regarding opportunities to participate within ICANN with relevant communities of interest in such forums.
    • In this way ICANN could support and improve multistakeholder participation, and thereby make claim to designated time within the program to canvas, consult and report back on ICANN issues through panel discussions and more technical side meetings.
  • Such an experiment could be instituted immediately through the piloting of proposed processes at forthcoming regional multistakeholder meetings of IGF.

Case Studies – What’s Worked in Practice?

Crowdsourcing at Various Stages of Decisionmaking

  • Dell IdeaStorm – An initiative launched in 2007, IdeaStorm allows Dell to “to gauge which ideas are most important and most relevant to” the public by enabling submission of ideas and articles by the public. The platform allows interested customers to rate and comment on ideas and has received over 16,000 ideas, nearly 500 of which Dell has implemented. [11]
  • India’s New Rupee Design – In 2009 the Indian Finance Ministry launched a public competition for new designs for the symbol for the rupee. The contest was open to all Indian residents and included a prize of 250,000 rupees.
  • Open Ministry (Avoin ministeriö) – In 2012 the Finnish government amended the national constitution so that any proposed legislation supported by at least 50,000 signatures must be put to a vote in the parliament. The Open Ministry project is a project to crowdsource legislation, which involves:
    • Ideation and Development: Proposed legislation topics need to be refined/framed into a clear proposition through discussion between interested parties.
    • Campaigning: To gain 50,000 votes, there must be a proactive outreach strategy.
    • Lobbying: Once a proposal goes to parliament there must be fine tuning and in-depth discussions with decision-makers.
    • Note, the platform on which proposals are voted on allows authenticated comments, using the same software in use by Finnish banks. [12]
  • Patient Feedback Challenge – The United Kingdom’s National Health Service (NHS) Institute for Innovation and Improvement created the Patient Feedback Challenge to generate and implement ideas to improve patient experiences at NHS organizations. Ideas were published on a web channel, and nine were chosen by an expert panel. Programs were piloted at nine participating NHS organizations and funded from August 2012 to March 2013.

Crowdsourcing via SMS-Based Input

  • Textizen – Built by Code for America, Textizen is an easy-access SMS-based tool for proactive outreach, structured input, and ongoing engagement. Without requiring people to be present at the Public Forum, a similar tool could be help to ICANN to:
    • Create custom fields to collect data with multiple question types, built-in logic, custom area codes, etc.
    • Structure and visualize data for quick insights, e.g., by exporting to CSV or using Textizen’s developer API.
    • Send follow-up texts to drive traffic and interest to a website or live meeting. ICANN could also send follow-up texts with project updates additional surveys, and event reminders.
  • Ushahidi – An open source software that allows users to crowdsource the mobile reporting of crisis information. Data collected is used for information collection, visualization and interactive mapping of crisis situations. The project began in 2007 after Kenya’s disputed presidential election as a way to provide citizens a way to share eyewitness reports of violence via email and text message to be mapped using Google Maps. [13]
  • U-report – Ureport is a free SMS-based system that allows Ugandans “to establish and enforce new standards of transparency and accountability in development programming and services” [14] by becoming “U-reporters.”
    • U-reporters share ideas on a range of development issues and the initiative consists of weekly SMS messages and polls to and from a growing community of U-reporters; regular radio programs that will broadcast stories gathered by U-report; and newspapers that will publish stories from the U-report community. [15]
    • Notably, by 2012, over 200,000 people have subscribed to the system, which started receiving more and more unsolicited messages.  Thanks to the creation of a text classification algorithm, UNICEF can categorize and sort the messages both category and by UNICEF “branches,” e.g., education, health, employment. Messages can also be ranked by severity so that UNICEF teams could prioritize messages at the top of the list.

Open Questions – Help Bring This Proposal Closer to Implementation

  • What institutional and cultural barriers could pose challenges to implementation?
  • What tools and designs would work best for ICANN considered?
  • How will ICANN perform outreach to ensure the global public is aware of new participation opportunities?
  • How will input be curated/evaluated? How will ICANN do “quality assurance”?
  • How to balance efficiency with broad-based participation?
  • What metrics will determine whether there has been “sufficient inclusivity”?
  • How able are the participants to meaningfully engage? How much of a learning curve exists?
  • How can ICANN take a benchmark of current practices in order to facilitate meaningful comparison with parallel crowdsourcing processes?
  • Which ICANN structures or groups would be best to facilitate such a pilot?
  • How can crowdsourced input on “issue framing” be incorporated into ICANN’s current practices?


1. “ICANN Public Forum.” July 18, 2013.
2. Jeff Howe. “The rise of crowdsourcing.” Wired Magazine. Issue 14, no. 6 (June 2006): 1-4.
3. “Internationalization & Regional Development.”
4. “ICANN Community.”
5. “GNSO Improvements – Opportunities for Streamlining and Improvements.” January 16, 2014 at Proposal 9: 7.
6. Ibid. at Proposal 4: 4 (noting that current GNSO Working Group guidelines do not mandate any “required participation,” but only suggest that “a Working Group should mirror the diversity and representatives from most, if not all, GNSO Stakeholder Groups and/or Constituencies.”).
7. Accountability and Transparency Review Team 2. “Report of Draft Recommendations for Public Comment.” December 31, 2013 at 40 (“[T]here continues to be a lack of GAC early involvement in the various ICANN policy processes.”).
8. “GNSO Improvements – Opportunities for Streamlining and Improvements.” January 16, 2014 at Proposal 3: 3.
9. Accountability and Transparency Review Team 2. “Report of Draft Recommendations for Public Comment.” December 31, 2013 at A-46.
10. Eric Bonabeau. “Decisions 2.0: The Power of Collective Intelligence.” MIT Sloan Management Review 50, no. 2 (2009): 45-52.
11. “About Ideastorm.”
12. Dawson, Ross. “How Finland’s Open Ministry Is Crowdsourcing Legislation.” Getting Results From Crowds. June 5, 2013.
13. Jeffery, Simon. “Ushahidi: crowdmapping collective that exposed Kenyan election killings.” The Guardian. April 7, 2011.
14. UNICEF. “U-report application revolutionizes social mobilization, empowering Ugandan youth.” November 20, 2013.
15. See

The Tags . . .

4 Responses to “Proposal 2 for ICANN: Get Broad-Based Input by Crowdsourcing Each Stage of Decisionmaking”

  1. Tim Bonnemann February 1, 2014 at 7:48 pm #

    Nice conversation starter!

    A couple of comments:

    Crowdsourcing and stakeholder engagement are two fundamentally different things. As you rightly point out, crowdsourcing relies on an “undefined (and usually large) network of people”. In stakeholder engagement, on the other hand, very specific groups or individuals need to be involved. Which one is it? If the idea is to do both in parallel or in some other complementary fashion, I suggest to outline in more detail what that would entail.

    Secondly, when it comes to public participation (involving the public in decisions that affect them), it is absolutely critical to know upfront how any of the input generated in the process will impact the decision at hand. In your example, the decision at hand is to define an agenda, presumably. The suggested approach is to support the “up-front issue framing” process by crowdsourcing the identification and ranking of issues. Would it matter who participates in this process and whether certain stakeholders are represented? How binding would the results be? Would the decision making body commit to implement what the crowd decides or merely take their input into consideration?

  2. Jillian Raines February 2, 2014 at 8:45 pm #

    Thanks for your comments! You’ve raised some really important questions that we need to think about as we move forward. Some initial thoughts:

    One of the biggest challenges in figuring out how to apply crowdsourcing techniques to stakeholder engagement is exactly what you suggest – which phases of the decisionmaking or policy development process lend themselves well for broad calls to participate? And at which stages of the process is it crucial to target requests to participate to certain stakeholder groups to ensure they’re fairly represented? And how do we determine upfront who will be affected by a given decision to ensure they are included in any targeted outreach? (Is that question one that crowdsourcing could help answer?)

    I would think the outline for running any experiment in parallel or as a complement would depend on the topic or issue at hand. Crowdsourcing wisely rather than widely would likely look different depending on whether you’re determining how to mitigate name collisions or figuring out how to use last-resort auction funds in the public interest or determining a new approach for conducting ICANN’s public forum.

    And that leads straight to your second point – having clear expectations and an understanding around how the public’s input will be used will be crucial before one can expect people to willingly engage. That’s a really important point that should be included in any pilot or experiment blueprint. Without clear communication re: how input will potentially affect action, determining whether a crowdsourcing pilot was legitimate and successful will be difficult.

    Maybe making the decision as to whether input will result in binding action or serve as additional consideration could depend on the level and diversity of participation that results in the experiment, e.g., who and how many participated, and were there clear ‘winning’ ideas or agreement demonstrated among those who were active? While a community should not be bound to act if only one or two interests give input; some may be dissuaded from participating if they know up front their input will not influence real action. And again, the nature of the decision topic or the “ask” should be considered, since some types of issues may lend themselves better for crowdsourcing widely and wisely than others.

  3. Chuck Gomes February 7, 2014 at 9:26 pm #

    I compliment the panel for providing constructive suggestions of how this proposal could be tested in ICANN along with existing tools that could be used. They also recognized the important of cost-effectiveness. And they raise important questions that need to be answered.

    I personally think it would be good to test this proposal.


  1. Govlab/ICANN Collaboration Explores Open Policy Making - February 4, 2014

    […] The second proposal was released last week: Proposal 2 for ICANN: Get Broad-Based Input by Crowdsourcing Each Stage of Decisionmaking […]

Leave a Reply