The committee has just announced 12 fresh grants over a wide range of themes. Committee member Pine told the Signpost that in his opinion "all are good projects, and both the committee and the WMF staff exercised due diligence in managing value for money. In particular, the staff did a pretty thorough job on the audio interviews [with applicants]." This round presented a good group of grantees, he said, and in his view only one unsuccessful application was a pity, but stands a very good chance "if it returns to us with a different set of conditions".
Pine told us that while all proposals have merit, two strike him as especially interesting. One is WikiTrack, which aims to expand worldwide the scope of a facility already introduced to four Indian-language Wikipedias. This initial version has enabled users of mobile devices to track recent changes, the diffs of those changes, their watchlist, and user contributions. The facility has already proved very popular, and the new grant—with a budget of only US$2500—will bring significant technical improvements and consolidate WikiTrack for all languages on Android into a single app. Applicant Hari Prasad Nadig, on the Kannada Wikipedia, is himself a programmer and has contributed to the free and open-source software movement in India for more than a decade.
But for Pine, the standout is Women and Wikipedia, funded at just over $8000. Nothing yet has proved effective in addressing the gender gap. The new approach is to enable triangulation using both qualitative and quantitative data, so we know better where we stand in designing future interventions and technologies. The qualitative element, crucially, will include women's narratives about their wiki experiences, gathered in-person or through online modes such as skype. Methods will include interviews, small focus groups, observations of editing and mentoring events (many of them being arranged for the upcoming Wikimania in London), and an online survey, and content analysis of movement-related pages. The focus will be the English-language communities, but one goal is to form a robust basis for adaptation to other-language communities. A final report is scheduled for January 2015, and there may be related publications in academic journals.
There will be two participants: Amanda Menking, an information science PhD student at the University of Washington, Seattle, who has been guest lecturer this quarter in University classes in which she spoke about Wikipedia and systemic bias; and David McDonald, her doctoral advisor and a faculty member at the University's Information School, whose research focuses on how to make large-scale collaborative systems more effective for users. David has published "numerous papers on different aspects of Wikipedia".
The Signpost put it to Amanda Menking that it's still unclear how narrative analysis might point to strategic action to solve gender asymmetry. Might her findings help us to prioritise or contextualise strategic action? She said that although quantitative data alone tells a story, choices are always made regarding what types of data to collect/ignore, keep/trim, and how to analyze it. Wikipedians' stories of their experiences, frustrations, and motivations, as well as a careful read of how the community constructs the story of the gender gap, will help to situate and contextualise quantitative data.
| “ | A qualitative piece provides an explicit and transparent framing of these choices and a richer understanding of how these choices have been made. I think a narrative analysis may also function as a strategic action in this case. That is, community members may feel a sense of value in the sharing of their stories and they may find a sense of connection and understanding around the topic of gender asymmetry. Sometimes, we simply need a safe space in which to share our thoughts and feelings so that a dialogue can begin to take place. This isn't just "talk"; this is the beginning of action—and an act in and of itself. I do know that some gaming communities have decreased gender and/or race related hostility by slightly increasing technical barriers (e.g., by turning chat off by default so that players must make the effort to turn it on), and considering possible design solutions is very interesting. However, it's really early on. I've just begun to collect data, and there's quite a "story" here to understand and share. |
” |
We asked Amanda whether WMF plans to work toward introducting real-time talkpage exchanges between editors—instant messaging and even audio—might create a more agreeable interpersonal environment for women; or perhaps might be perceived as risky, even threatening. Will her results shed light on how to link women’s inherent preferences and fears to technological development for the communities? She said: "I would love to include questions about these kinds of design decisions and innovations", but that it is a modest research project and might be a longer-term goal.
From the perspective of large-scale collaborative systems, how different are the challenges faced by the English Wikipedia and other WMF sites in attracting and retaining female contributors compared with those faced by other such systems? David replied:
| “ | The problem of balanced gender representation pervades many distributed contributor systems (like Wikipedia, GitHub, Reddit, YouTube, and Flickr). A few systems stand out, such as Pinterest, where women dominate or are representative, but they are rare enough that one could enumerate them. On Wikipedia the issue is a bit extreme even given the general skew across "similar" systems. I make that statement based on the prior survey work by the WMF, and some studies where researchers have attempted to data-mine gender info from user pages. Why gender representation is so skewed is somewhat unclear at this point and one of the things we would like to figure out. Focusing on women who are participating is one way to begin to get a picture. It might be equally interesting to understand something about the women who are not there or those who have quit. But in a modest research project with a modest budget, we wouldn't get to answer every question ... |
” |
Other new IEG grants are: Reimagining Wikipedia mentorship, a "choose-your-own-adventure approach to skill learning that incorporates the 1-to-1 interactions of mentorship" ($22,600); a suite of systems to accelerate the translation and integration of medical articles into as many languages as possible ($10,000); tools for WMF Armenian-language projects (zoom into proofreading segments, easy cropping of scanned book-page images, mass parsing, and section-name monitoring) ($7600); the development of optimised online categories, including the possible use of Wikidata as a unified category system across WMF sites ($9750); a pilot program to engage digitally-literate senior citizens in editing the Czech Wikipedia (~$8000); a project to bring to fruition a pronunciation recording facility for Wiktionary ($1980); an open-access reader to deliver a complete workflow from online open resource to editor, allowing editors to access highly relevant open-access research sources ($6500); an online catalogue to make Telugu content more accessible "to Wikimedians and other open-source activists" (~$1700); Wikiquiz, an app for fun engagement with readers of the Chinese Wikipeida ($1070); and Promoting Wikivoyage, a welcome first funding for that new sister project, to promote it through local tourism bureaus (seemingly an under-ask at just $600).
Discuss this story