Grants:PEG/WM Movement/Flow Funding Pilot Project/Report

Report accepted
This report for a Project and Event grant approved in FY 2012-13 has been reviewed and accepted by the Wikimedia Foundation.
  • You may still comment on this report on its discussion page, or visit the discussion page to read the discussion about this report.
  • You are welcome to Email grants at wikimedia dot org at any time if you have questions or concerns about this report.


Compliance and completion

edit
Did you comply with the requirements specified by WMF in the grant agreement?
YES
Is your project completed?
YES
Did you use any of the grant funds?
YES

Activities and lessons learned

edit

This section describes what the grantee did, and what the grantee learned from implementing the project. This section should be useful to others implementing similar projects and is an opportunity for the grantee to reflect on the project's performance.

Activities

edit
After six months of intense discussion and development together with the Wikimedia Foundation and the Wikimedia community, TSB coordinated from November 2012 until July 2013 the Flow Funding Pilot Project.
This pilot project was created to test an innovative model of fund allocation called “Flow Funding”, entrusting Wikimedia volunteers with decision-making power to allocate financial resources to initiatives they autonomously identified and selected for being aligned with Wikimedia's strategic goals.
All Wikimania Scholarship recipients and Grants Advisory Committee members in 2012 were surveyed. And from this group ten volunteers, 40% female and 50% from the global south, showed interest and were available to participate as "flow funders" in the Pilot Project. Each flow funder participant was then entrusted to allocate between $500USD and $2,000USD to initiatives they would autonomously identify and fund.
Flow Funder Name Gender Location Languages
Isla Female Cape Town, South Africa en-N
Thuvack Male Johannesburg, South Africa ts-N,en-4
Ilario Male Lugano, Switzerland it-N,en-4,fr-3,es-2,de-2
Lpagola Female Córdoba, Argentina es-N, en-3
Solstag Male São Paulo, Brazil pt-N, en-4, es-3, it-2, fr-1
Kiril Simeonovski Male Skopje, Macedonia mk, sh-4, en-4, de-3, bg-3, ru-2
Ganesh Paudel Male Kathmandu, Nepal ne-N, hi-4, en-4, sa-2, bh-2, new-1,
IbrahimPsy111 Female Minnesota, USA en-N
West.andrew.g Male Philadelphia, PA, USA en-N, es-1
poupou Female Berlin, Germany de-N, en-3
Throughout the project, flow funders engaged with each other, with the Wikimedia community and the Wikimedia Foundation through an open mailing-list and a Flow Funding Portal created on Meta. Additionally, four open IRC meetings (February, March, April and May) and one face-to-face meeting at Wikimania 2013 were facilitated to discuss with participants and the community the latest flow funding updates, challenges and opportunities.
As a result of this pilot initiative, four projects were funded, totalling approximately USD6,500 directly transferred from the WMF to the final recipients (the flow funders never handled any money themselves).
Funded Project Name Grantee Name Flow Funder Amount requested Status Report
Faculty of Economics Skopje Igor Sazdovski Kiril Simeonovski 1846 USD / 1420 Euros Completed Report
Fotofahrt im Heißluftballon Martina Nolte poupou 1235 USD / 950 Euros Completed Report
Medievalia 2013 Pietro Di Lorenzo Ilario 1945 Euros Completed Report
Transite Felipe Baenninger Solstag 1500 USD / 3000 BRL Started

Lessons learned

edit
What lessons were learned that may help others succeed in similar projects?
The flow funding pilot was a great opportunity to test and evaluate the flow funding model to Wikimedia's conditions and needs. The overall engagement and contributions of volunteers remained high despite of a few setbacks identified below. The funded projects illustrate how volunteers were able to identify by themselves some interesting and innovative initiatives that may help achieve Wikimedia's goals in a faster and more distributed manner. As a coordinator of the pilot project, I strongly recommend everyone to take this experience as an example of how to decentralise Wikimedia's funding processes and how to leverage volunteers' engagement even more, all in alignment with Wikimedia's values.
I will quote a text written by one of the flow funder candidates to summarise what I consider the most valuable (shared) learning of the flow funding pilot project: "Being a flowfunder takes a lot of time and effort. He (or she) has to go after the projects, they won't come to him (or to her) that easily. The Flowfunder must be knowledgeable and judicious about (Wikimedia's) mission. The Flowfunder must be creative and help the proponents think how to adapt their work to Wikimedia. Flowfunding means inserting yourself in a rich context and engaging projects, not fishing for them."
What went well?
  • Recruitment/selection of flow funder candidates: It was a sensible and transparent approach to select/recruit flow funders candidates for the pilot project among Wikimania scholarship recipients and Grant Advisory committee members. The pilot group with 10 participants was diverse, representative and well balanced. Nonetheless, the recruitment of candidates for a larger scale programme would have to modified to be open and accessible to a more diverse group of candidates.
  • Advertisement of the flow funding initiative: The flow funding initiative was widely advertised (several mailing lists, "Village Pump", Kurier, talk pages, WMF official communications, Wikimania agenda etc).
  • Channels for communication: The public channels of communication created for the pilot project (mailing-list, collective portal on meta, flow funder's individual portal pages, IRC meetings and Wikimania presentation) offered enough space for open and transparent communication between flow funders, WMF, local communities and pilot project coordinator. Further thought would be necessary to better coordinate and integrate both the centralised and the decentralised communication channels (collective portal on meta and flow funder's individual portal pages).
  • WMF grantmaking structure: The WMF staff have always responded in a timely manner to flow funders requests and concerns.
  • Time frame of the pilot project: The period of six months was enough to create all the conditions for the flow funders to understand and experience the opportunities and limitations of the flow funding model. More time would have enabled more flow funders to identify and fund projects, but it would have defeated the purpose of timely understanding the dynamic of the model before a potential roll-out phase. Nonetheless a larger flow funding programme, not the pilot project, should consider extending the amount of time for the flow funders to identify and fund projects - between 9 and 18 months would be possibly enough .
What did not go well?
  • Pilot start delay: It took six months of negotiation with the WMF before financial resources were allocated to fund a revised and smaller pilot project to test the flow funding model. Considering the innovative quality of the pilot project it is understandable that some negotiation was needed before the project could be implemented, but the delay in receiving a final approval had an impact on the number of candidates available to act as flow funders. Initially 16 candidates showed interest, but only 10 were available when the pilot started. Among the ten remaining candidates some had less time availability when the pilot started than when they first agreed to participate - impacting their participation level.
  • Pilot budget cut: The final approved budget for the pilot excluded resources to pay for a project coordinator and facilitator. The flow funding proponent agreed to facilitate the pilot as a volunteer, but the number of hours per week to coordinate the initiative were substantially smaller. Furthermore, the budget for training the candidates was also cut, which impacted the depth of candidates' understanding of the initiative, the flow funding concept and the proposed model.
  • Funding tool and email address: The funding request tool and the Wikimedia foundation address to submit the funding requests did not fully work throughout the pilot project.
  • Profile of flow funder candidates: Candidates for the pilot project were selected from a pool of Wikimania scholarship recipients and Grant Advisory Committee members. None of these two groups did necessarily provide candidates with a characteristic fundamental to act as a flow funder: be able to source funding proposals. When flow funders understood that they were expected to identify projects by themselves, some became very proactive, but some disengaged - possibly because they didn't feel comfortable to act that way.
  • Funding criteria change: The funding criteria were changed by the Wikimedia Foundation after the pilot project had already started. The new criteria included a more restrictive understanding of conflict of interest (COI) and payment of labor. Originally the pilot criteria was designed to accommodate a wide range of projects in order to evaluate the profile of fund-seekers not served by the current funding options. The alteration of the criteria had an impact on at least three fund requests. These fund requests were submitted by two flow funders in accordance with the original criteria, but unfortunately had to be declined by the WMF. This change of parameters created some confusion among flow funders, but at the same time allowed for a constructive and transparent debate about COI and payment of labor.
  • Funding upper and lower limits definition: The approved version of the flow funding pilot restricted each individual funding request to an amount between USD 500 and USD 2,000, compared to the original proposal of up to USD 20,000. It is understandable to establish a minimum threshold to balance the number of requests and the back-office processes necessary to implement them. According to some flow funders' evaluation this balance should have strived for an alternative (and simpler) process in order to allow for a greater number of smaller requests. The upper funding threshold, on the other hand, should not have been so low. By defining USD 2,000 as the maximum funding request, the WMF limited the size and number of projects that were eligible to be funded. Flow funders shared on their reviews and evaluations a strong desire to be able to identify and fund larger projects. Among other reasons, some flow funders believed that the effort to request USD 500 is not different from larger funding requests, but the leverage and influence they would have to align a potential project to Wikimedia's specific goals would be much greater with more discretionary funding.
  • Pilot official language: English was(is) used for the communication between flow funders, Wikimedia Foundation and the Pilot coordinator. This was a necessary condition considering the limitations of scale, time and funds of the pilot project. Nonetheless, it is worth noting that this language barrier have possibly limited the participation and contribution of volunteers from non-speaking countries. Flow funders were able to communicate in they preferred idioms with potential grantees, but when a funding request was submitted to the WMF much of the communication and reporting would then have to occur in English. For a larger scale flow funding programme groups of flow funders could be organised by idioms and the WMF should consider a multi language structure.
What would you do differently if you planned a similar project?
A similar project would achieve greater results with training opportunities being offered to flow funder candidates, a hired contractor to coordinate and facilitate their contribution and a higher discretionary upper limit to fund projects.
Finally, when recruiting flow funders, more emphasis should be placed on informing candidates about the characteristics of the flow funding model - volunteers need to be previously informed that they have to be extremely proactive and that the overall experience will require a considerable amount of their time.

Project goal and measures of success

edit

Project goal

edit
The Flow Funding Pilot Project was designed to assess if and how a Flow Funding model could contribute towards building a stronger community of Wikimedia volunteers while leveraging the growth of the Wikimedia Movement. Three critical aspects were assessed in the Pilot Project with respect to the Flow Funding model within a Wikimedia context:
  1. Supply of reviewers to collaborate and work in a decentralized way
  2. Demand from fund-seekers not served in the current option set
  3. Potential problems of gaming the system
Did you achieve your project goal? How do you know your goal was achieved?
Yes, in a period of six months the pilot project demonstrated that out of a sample of 150 Wikimedia active contributors:
  • 8 volunteers actively collaborated and worked in a decentralised way to identify and fund projects + 2 volunteers remained inactive
  • 4 new projects were funded in accordance with the funding criteria and funding restrictions of USD 500 (min) and USD 2,000 (max) + 2 new projects were declined
  • The risk of gaming the system was avoided by always promoting an open and transparent communication between flow funders, WMF, grantees and the WM movement

Measures of success

edit

1. Supply of reviewers to collaborate and work in a decentralized way

Before the Pilot started, around 150 Wikimania Scholarship recipients and Grants Advisory Committee members were surveyed to gauge their interest in acting as flow funders. Out of this total, 10 volunteers showed interest and were available to participate in the Flow Funding Pilot Project. By the end of the Pilot phase, 3 new volunteers and 2 participants in the pilot phase spontaneously showed interest to (continue to) act as flow funders.
The table below illustrates the commitment and development of the ten flow funder participants monitored throughout the Pilot Project:
  • Participation in meetings, edits on Meta and announcements on mailing list
  • Contribution to the identification and reviewing of grassroots initiatives
  • Reporting back on grassroots initiatives, including the pilot initiative
Flow Funder Participation Contribution Reporting Evaluation
Isla Inactive
Thuvack Active
Ilario Project Report Active
Lpagola Active
Solstag Very active
Kiril Simeonovski Very active
Ganesh Paudel Inactive
IbrahimPsy111 Active
West.andrew.g Very active
poupou Very active


2. Demand from fund-seekers not served in the current option set

In the pilot project each flow funder was entrusted with autonomous decision-making power to allocate financial resources to initiatives that showed potential for impact within the global movement's priorities and targets. The funding criteria accommodated a wide range of grassroots initiatives identified by flow funders, while excluding flow funders’ own projects or that of a relative.
The quantity, quality and impact of initiatives identified and supported by flow funders was assessed during the Pilot Project by monitoring:
  • Number of projects selected to receive funds
  • Impact within the global movement's priorities and targets
  • Community and chapter members' feedback on initiatives’ outcomes
  • Knowledge contributed to the Wikimedia movement
  • Efficiency of the proposed use of funds
Funded Project Name Amount Impact Report Feedback Evaluation
Faculty of Economics Skopje
  • Requested: 1846 USD / 1420 Euros
  • Used: 1125.50 Euros
Achieved:
  • Enlarged mk.wikipedia.org community
    • 2 free lectures teaching the basics of creating and editing articles
    • 232 students engaged most of them not having edited articles on Wikipedia before
    • Possibility of implementing Wikimedia projects at University of Ljubljana, Slovenia
  • Improved quality of the statistics articles in Macedonian
    • 1 free lecture about “Structural Vector Autoregressive models”
    • 106 articles created and expanded (more than half are mid to large size)
    • 1 Statistics professor monitored the quality of all the articles
    • Each article was required to have relevant sources
    • Interest showed by professors at the Faculty of Economics to create articles as an activity
Report
  • Wikimedia Macedonia did not support the project
  • A few members of the Macedonian Wiki-community volunteered to help
Efficient
Fotofahrt im Heißluftballon
  • Requested: 1235 USD / 950 Euros
  • Used: 1235 USD / 950 Euros
Achieved:
  • More than 2,000 pictures uploaded to Commons
    • As of July 2103 (report date), 46 pictures were used in the global main namespace of Wikimedia projects
    • One image has been considered the most valued image on Commons within its scope
  • Wikimedia Germany expressed the general readyness to support further balloon photo flights
  • New Wikipedia articles about buildings, harbour bassins, and companies in Hamburg were written (example), or drafted (example), or planned (list).
  • The contribution of two female Wikipedia contributors (flow funder and grantee) was widely disseminated
Report
  • The project has been criticized by a few Wikipedia authors. They claimed that efforts and costs of "freeing" existing flickr photos would be comparable to the hot balloon projects.
  • The results of this initiative encouraged other Wikimedians to plan successive balloon flight photo projects. Wikimedia Germany supported a successive balloon flight in Cologne in July/August 2013. Further flights in the densely populated Ruhr area were also planned.
Very efficient
Medievalia 2013 1833 Euros Achieved:
  • The project gave the opportunity to widespread Wikimedia Projects in an area where the Wikimedia Projects are used passively
  • The project created a general interest for a follow-up and for free licenses
  • The distribution of the advertising material was wide and the knowledge of Wikipedia made stronger

Awaiting confirmation of impact:

  • Links to all the materials available under free licenses
  • Details about deeper Wikimedia conference given to selected participants
  • Further development of partnerships with universities, local administrations, libraries etc
Report Awaiting feedback shared by Swiss and Italian Wikimedian communities Efficient
Transite 1500 USD / 3000 BRL Planned:
  • Promote wikimedia's mission among photographers and art and bicycle activists that have vision alignment and produce much content
  • Teach people throughout Brazil to learn how to upload to Wikimedia Commons, include images in Wikipedia, and about the values of the movement
  • Upload to Commons images of cities, people and roads all over Brazil by people reached during the trip (and also professional grade images by the grantee)
  • Open a first door to some groups of artists and activists that have a lot to contribute, that could be candidates for the grants program, but don't have familiarity with the Wikimedia Movement
  • Receive press attention and bring visibility to Wikimedia's mission
Awaiting report

3. Potential problems of gaming the system

In order to increase accountability and knowledge sharing, each flow funder candidate have openly reported back on her/his flow funding experience and on funded initiatives’ impact. Funds were directly transferred from the Wikimedia Foundation to initiatives identified by flow funders. Transactions were transparent and accessible to the Wikimedia community to appraise and comment.
Two out six projects identified by flow funders were withdrawn from the flow funding process because they did not meet the funding criteria. A conflict of interest was identified by the WMF in the project Video and Interactive Tutorials and the compensation of volunteers for their time in project Training Librarians to Become Campus Ambassadors was evaluated by the WMF as against the funding criteria. The entire discussion was open to the community to access and participate on Meta.
Provide an overall assessment of how your project went according to these measures.

SWOT analysis (Strengths, weaknesses, opportunities and threats) associated with the flow funding model:

Strength: The decentralised quality of the flow funding model was confirmed by the pilot initiative.The flow funding model showed strong alignment with Wikimedia's open, diverse and volunteer-driven movement. Based on the results of the pilot initiative, the decentralised flow funding model allowed the volunteer contingency to develop and to grow. Flow funders learned from their experience and felt empowered to further contribute to achieve the movement's goals, reaching out to new contributors and supporting fund-seekers not served in the current option set. Every activity was done in an open and transparent way, managing to completely avoid the risk of gaming the system.
Weakness: The complexity of flow funders' coordination was confirmed during the pilot initiative. Flow funders required more support to collaborate and work in a decentralized way than it was provided by the project coordinator in a limited volunteer basis. The engagement and interaction between flow funders remained low, as evidenced by the lower participation rates on online activities and meetings (i.e. Flow Funding Portal, mailing-list and IRC meetings). The project coordinator was able to provide approximately six hours of volunteer time per week during the pilot project, but more time would have been necessary in order to fully explore flow funders potential to collaborate with each other and to identify funding opportunities.
Opportunity: The wealth of resources raised by the Wikimedia movement could be channelled through a flow funding model. In recent years the movement has managed to raise larger sums of financial resources, but the funding decision processes remained centralised by the Wikimedia Foundation staff, and more recently by the limited number of members of the Funds Dissemination Committee (FDC). The pilot project confirmed that Wikimedia volunteers are available to collaborate and work in a decentralised way in order to fund initiatives not served in the current option set. By adopting a larger scale flow funding model, more knowledge, connections and volunteers could be leveraged to complement Wikimedia's existing funding processes. And Wikimedia's volunteers would count on all the additional resources provided by the flow funding model to further engage and contribute to the movement's goals.
Threaten: The heterogeneity of tax systems and bank transfer requirements and fees reduce the efficiency of the allocation of funds worldwide. During the pilot project the Wikimedia Foundation worked on a case by case approach to find the most appropriate alternative to transfer resources to grantees selected by flow funders. In at least one case during the pilot project, local tax requirements could have significantly reduced the amount of resources available to fund a selected project. A larger scale flow funding programme would reach out to a larger number of fund-seekers not served in the current option set, but would require more understanding of tax systems, transfer requirements and costs worldwide. A comparative analysis between the flow funding model and the current funding models would illustrate when taxes and transfer fees need to be paid multiple times or in larger amounts before reaching the grantees.
If you were to plan a similar project, would you measure it differently? If yes, please explain how.
No

Impact

edit

The flow funding pilot project had the following direct impact on:

  1. Encouraging innovation: The flow funding model aimed at bringing innovation to one of the movement's most strategic processes, offering an opportunity to evaluate a decentralised funds dissemination process aligned with the movement's values and goals. The pilot also enabled 4 fund-seekers not served by the current funding structure to to be identified by Wikimedia volunteers and to receive funds from the Wikimedia movement.
  2. Increasing participation: 8 Wikimedia volunteers actively participated in the flow funding pilot (3 active volunteers were women and 3 were from the global south), offering their time, connections and knowledge to the movement. They were able to identify and fund 1 new project in the global south and 1 new project lead by a female grantee.


The flow funding pilot project also had the following indirect impact on WMF's strategic goals:

Impact already achieved by funded projects Impact still planned to be achieved by funded projects
  • More than 2,000 pictures uploaded to Commons
    • As of July 2103 (report date), 46 pictures were used in the global main namespace of Wikimedia projects
    • One image has been considered the most valued image on Commons within its scope
  • Wikimedia Germany expressed the general readyness to support further balloon photo flights
  • New Wikipedia articles about buildings, harbour bassins, and companies in Hamburg were written (example), or drafted (example), or planned (list).
  • The contribution of two female Wikipedia contributors (flow funder and grantee) was widely disseminated
  • Enlarged mk.wikipedia.org community
    • 2 free lectures teaching the basics of creating and editing articles
    • 232 students engaged most of them not having edited articles on Wikipedia before
    • Possibility of implementing Wikimedia projects at University of Ljubljana, Slovenia
  • Improved quality of the statistics articles in Macedonian
    • 1 free lecture about “Structural Vector Autoregressive models”
    • 106 articles created and expanded (more than half are mid to large size)
    • 1 Statistics professor monitored the quality of all the articles
    • Each article was required to have relevant sources
    • Interest showed by professors at the Faculty of Economics to create articles as an activity
  • Opportunity to widespread Wikimedia Projects in the South of Italy where Wikimedia Projects are used passively
  • General interest for a follow-up and for free licenses
  • Wide distribution of advertising material mentioning Wikimedia
  • Promote wikimedia's mission among photographers and art and bicycle activists that have vision alignment and produce much content
  • Teach people throughout Brazil to learn how to upload to Wikimedia Commons, include images in Wikipedia, and about the values of the movement
  • Upload to Commons images of cities, people and roads all over Brazil by people reached during the trip (and also professional grade images by the grantee)
  • Open a first door to some groups of artists and activists that have a lot to contribute, that could be candidates for the grants program, but don't have familiarity with the Wikimedia Movement
  • Receive press attention and bring visibility to Wikimedia's mission

Reporting and documentation of expenditures

edit
Ten flow funders were allowed by the Wikimedia Foundation to individually allocate up to USD 2,000, totalling a budget of USD 20,000. The resources remained with the Wikimedia Foundation and were transferred directly to the grantee only if/when the flow funder made an official request.
The reporting and documentation of expenditures is available here for each funding request. In total, 4 flow funders requested approximately USD 6,500 and by the end of the Pilot Project (July 2013) the majority of the pre-approved budget remained unused and with the WMF (approximately 67%).