Community Insights/2018 Report/Legal Department

Community Engagement Insights 2018 Report: Support & Safety

The Wikimedia Foundation's Legal department oversees all legal matters for the Foundation and focuses on key initiatives to help support the communities, consistent with our Foundation goals and values. They deal with a wide range of issues and projects, including public policy, trademark and copyright law, international law, employment law, litigation, fundraising and grant law, domestic and international contracts, privacy law, ethics, internet law, and non-profit corporate governance.

In participating in the survey, the Legal department is interested in answering the following questions:

  1. What does engagement look like for the Wikimedia movement in legal and public policy work, in terms of awareness and preferences for communication channel?
  2. What opinions do Wikimedia contributors hold, so that the Legal department can align with community expectations, desires and values?[1]
  3. What keeps Wikimedia affiliates from getting involved with public policy and advocacy work?



Results edit

1. What does engagement look like for the Wikimedia movement in legal and public policy work, in terms of awareness and preferences for communication channels? edit

The Legal department uses communication channels to connect with Wikimedians about to legal or policy issues in the movement. Participants in the survey were asked about their awareness of these various communication channels.

 
2017 Wikimedia Foundation Transparency Report

In general, we find that the Wikimedia Foundation blog is more well known than the other sources, as 72% of the participants reported being aware of the blog (LA01). Compared to editors and other contributors, organizers (affiliates and program leaders) are significantly more aware of the public policy website, mailing list, and blog.[2] No statistically significant differences were found between the two audiences in their awareness of Twitter. Awareness of the public policy website did not change significantly by audience compared to last year.

Participants in the survey were asked about their awareness of the Transparency Report (LA06). The Legal department publishes a transparency report twice each year to show the ways in which the Foundation protects user privacy and defends against censorship, which are essential to the success of the Wikimedia mission.

On average, awareness of the Transparency Report is higher among Wikimedia organizers (affiliates & program leaders) and developers than among editors. We found that 14% of high-activity editors and 9% of low-activity editors were aware of the report, while 45% of volunteer developers and 51% of affiliate and program organizers were aware of the report.[3] These differences were statistically significant.[4]

In general, it appears that a wiki page may be a favored channel for learning about Wikimedia's policy work among Wikimedia community audiences (LA10). As the Legal department would like to engage contributors in legal and policy work, they asked participants their preferences for which channels they prefer to use. Due to routing difficulties in the survey, unfortunately, response rates for this question were low, so the results are general.


2. What opinions do Wikimedia contributors hold, so that the Legal department can align with community expectations, desires and values? edit

Questions relating to public policy & advocacy edit

The Legal department wanted to hear perspectives from contributors about the legal environment in the movement. The legal team asked which policy areas were most important among contributors.

From various policy areas, "Access to Knowledge" was the policy area most selected as being important for Wikimedia (LA03). For each of the audiences, the responses to this question look fairly similar across the groups: privacy and intermediary liability were rated lower, and access to knowledge and copyright were rated higher. For volunteer developers, copyright and access to knowledge were selected about the same, while for all other audiences, access to knowledge prevailed as the most important. The answer options for this question were not randomized in the survey, which may have introduced some bias. These data have not been checked for statistical differences.

Within the policy area of "Access to Knowledge", respondents scored "internet access" the highest and "offline wikipedia" the lowest (LA11). Unfortunately, there was an error in the question. The question asked was "In your opinion, how important are the following aspects important to Wikimedia?", and it should have read "To what extent are each of the following aspects of access to knowledge important or not important for Wikimedia?" Although the results should be interpreted with caution, the results may prove useful in how participants ranked between the choices, despite the error in the question. Among low-activity editors who responded to the question (n=31), internet access, access to factual information, and safe participation in knowledge online scored the highest means. Telecommunication regulation and offline Wikipedia seemed to score lower. The high-activity editors who responded to the question (n=52) rated these items similar to low-activity editors, except they seemed to rate open-access publishing higher as well. Among developers who responded to the question (n=14), the highest median responses were internet access and safe participation online, with offline Wikipedia having a lower score. Among program leaders who responded to the question (n=17), internet access, open-access publishing, access to factual information online, and safe participation in knowledge online had the highest medians, while offline Wikipedia was lower.[5]



Questions relating to Wikimedia project content edit

The Wikimedia Foundation's Legal department provides legal support to the Foundation, and, as part of that support, sometimes seeks to learn more about the community’s views on various policy issues.

A majority of contributors are not aware of the topic of paid editing (LA07). Prior to asking about opinions on paid editing, we needed to understand the awareness of paid editing as an issue among Wikimedians. The mean response for high-activity editors was 2.14, or just above "slightly" while the mean response for low-activity editors was 1.61 or between "not at all" and "slightly". These results were statistically significant.[6] There were also statistically significant differences when comparing responses between English Wikipedians and other Wikimedians. On average, English Wikipedians reported a mean response of 2.13 with a median of "Slightly" (2), while non-English Wikipedians reported a mean of 1.85 and a median of "Not at all" (1).[7]

Those who reported having "slightly" or higher awareness of the topic of paid editing were asked whether they agree or disagree with this statement: "The Wikimedia Foundation should be doing more to curb paid editing on Wikipedia." (LA08). Results were mixed. 52.2% of 184 contributors agreed, while 32.1% selected "neither agree nor disagree" and 15.7% disagreed. When checking for differences between activity levels of editors, no differences were detected.



Among those who were aware of paid editing, Wikimedia contributors were also asked about compliance with the Creative Commons license (LA08). There was a routing error in this question, and the question went to all editors, Wikipedians, Commons users, and others, who who reported having "slightly" or higher awareness of the topic of paid editing. Among respondents to this question, 71.4% of 189 respondents agree or strongly agree with the statement "The Wikimedia Foundation should do more to ensure that Wikimedia content is reused in compliance with the Creative Commons license." Twenty-five percent selected "neither agree nor disagree" and 3.2 percent selected either disagree or strongly disagree. There were no statistically significant differences between contributors from any particular countries.

Wikipedia editors who participated in this survey are divided about whether a person's information should be removed from Wikipedia when that information is considered private in the person's country of residence (LA12). Wikipedia articles sometimes have information about living persons that is legally allowed on Wikipedia, but that would be considered private information where the article subject lives. Wikipedia editors were asked whether private information should be removed when this situation happens. The results are mixed. From 516 editors, about a third agree, a third disagree, and a third selected "I don't know."

In general, most Wikipedians who participated in this survey agree (65%) that content about a living person that editors believe is problematic should be removed until the problematic content is fixed. (LA15) Out of 488 respondents, about 16% reported that they disagree with this type of content being removed. The median response to this question is 4 ("agree"), and the mean is 3.6 (between "agree" and "neither agree nor disagree").



3. What keeps Wikimedia affiliates from getting involved with public policy and advocacy work? edit

Wikimedia organizers report lacking expertise, capacity and resources for doing public policy and advocacy work ((LA09)). From about 100 respondents, 92 said that they lacked their own capacity or resources, and 62 said that they lack of expertise or experience. Less often selected were a lack of support from community (36 respondents), a lack of support from the Foundation (32 respondents), their partners' capacity/resources (28 respondents), and a lack of interest (26 respondents).


Most useful results edit

  • Communications: For communicating legal information from the Foundation, it would be better to use channels that aren't designed just for public policy, but ones that are used for more broad use cases, when the goal is to reach the community as broadly as possible (LA01).
  • Limitations of public policy work: A lack of capacity, resources, and expertise from affiliates are the primary reported limitations that prevent affiliates from being involved in advocacy.
  • Content policies: There is no clear consensus about either whether a person's information should be removed from Wikipedia when that information is considered private in the person's country of residence (LA12).
  • Paid editing: Slightly more than half are not aware of paid editing. Also, of those who are aware of paid editing, slightly more than half believe that the Foundation should be doing more to curb it.
  • Policy areas: Access to knowledge was reported as most important for communities. (Follow-up research may be needed to understand how communities understand the terminology about these policy areas.)

Next steps edit

Based on the results from this report, the Wikimedia Foundation Legal department will:
  • Discuss and continue to develop the communications strategy for our semi-annual transparency report.
  • Incorporate community interest in access to knowledge as the policy team prioritizes defensive and proactive actions.
  • Experiment with existing and alternative communication channels to inform the community about policy topics.
  • Develop a roadmap for working with community members and strategic partners to establish best practices and enforcement options related to content re-use.
  • Have a conversation about how to discuss and frame "intermediary liability" and "access to knowledge" in our public policy work.

See also edit

Include links to appendix materials (SPSS output, additional graphics, etc.)

Notes edit

  1. Note: the Legal Department asked two additional questions related to legal risks and threats to Wikimedia projects and participants. The responses to those questions are not reported here, due to the sensitive topics they cover.
  2. Mann-Whitney tests for differences: Wikimedia Blog, p-value=.000; Public Policy website, p-value=0.042, Public Policy mailing list, p-value=0.015
  3. Out of 989 participants who responded to this question, 109 (11%) reported that they were not sure whether they had heard of the report. These people were excluded from this analysis.
  4. P-value < 0.000; effect size is medium = 0.397
  5. Affiliates had a very low response (n=5) to this question, so we are excluding it from the results.
  6. p-value=0.000, effect size = 0.21
  7. p-value=0.012, effect size = 0.12