Candidates talk about the Meta problem, the nation-based chapter model, world languages, and value for money
The Board of Trustees answers questions at Wikimania 2010
Last week the Signpost began a two-part interview with the candidates for these positions; this week, the second and final part of the interview explores two broad themes: Meta, the site that hosts movement-wide coordination; and offline entities—the chapters and the new thematic organisations and user groups.
The 12 candidates are Leigh Thelmadatter, Milos Rancic, Liam Wyatt, Phoebe Ayers, Tom Morton, John Vandenberg, María Sefidari, Jeromy-Yu Chan, Samuel Klein, Kat Walsh, Michel Aaij, and Francis Kaguna. Kat Walsh was alone among all candidates in sending no responses to the Signpost.
In late news, the election will not begin on 1 June, as notified some time ago. Little more than 12 hours before the advertised start, the election was postponed by a week, to 8 June. The reasons given involve the inability to verify the voter list and that SecurePoll is properly functioning, and the fact that the voter interfaces have not been translated from English to other languages. Given that the Signpost had just been published, we were not in a position to make enquiries about what went wrong.
All active editors of WMF projects (with minor exceptions) can vote, and there is information on how to vote, including a direct link to the SecurePoll system, which will be active from UTC midnight start of Saturday 1 June for two weeks. Risker, a member of the independent election committee, told the Signpost that banners with direct links should appear for logged-in users during the voting period. Elections for two seats on the Funds Dissemination Committee (FDC) and its ombudsperson are being held concurrently. Candidates for these positions are interviewed in this week's "News and notes".
Meta: "a specific kind of mess"?
Meta was a child of the very early wiki environment in 2001, a time when there was little more than a threadbare English Wikipedia. Meta is the main site for stewards, the Foundation’s grantmaking activities, WMF elections, and the archives of chapter reports. The boundaries are not entirely distinct between Meta and the Foundation’s official website, and between Meta and Outreach. Despite the increasingly global reach of Meta, the site remains almost entirely in English, technocratic, and with a certain in-house mentality; editors who are new to the fray and who lack fluency in English face significant hurdles in participating.
Are candidates happy with the accessibility of Meta to members of the editing communities?
Some candidates point to the elephant in the room—that "most Wikipedia editors don't know exactly what Meta is all about" (Michel, Leigh). Milos says: "Meta is used by all Wikimedia communities, which creates a specific kind of mess. Although as a steward I'm a Meta 'native', it's often hard for me to navigate there. I don't have a particular idea of how to solve it, except to call [for] better design." Phoebe comments: "not many folks on the projects know about Meta, and they probably find it confusing to navigate, as the site has suffered from a lack of concerted maintenance over time. The best thing we could do ... is to continue to consolidate key information and projects like outreach and event planning there, and put in some design work to make it easier to use."
Samuel believes that "both small tools to support cross-wiki and cross-language discussions, and better coordination of announcements" would help. John says that although improvements have been made in translation tools, "Meta has been left to languish and be dominated by 'insiders' because the WMF and its Board have chosen to make many decisions without adequate consultation of the community, [with a few] notable exceptions." Tom emphasises the need for active links from other projects: "we lack enough hyperlinks to Meta, which means people approach it in a very haphazard manner. ... The meta community probably needs to broaden and diversify [with] more strategic collaboration there to build links between projects."
The big ballot starts Saturday
The inter-language issue is a prevalent theme. Liam points out that "because Meta like our global mailing-lists defaults to English, many people are excluded. This is a symptom of a global community and there's no perfect solution for it—either technical or cultural." Jerome-Yu thinks that "English proficiency is sometimes a big barrier for involvement in Meta and WMF business". For María, that this is especially true "when discussions get technical", and that even talk-page discussions and voting can be confusing to editors from non-English projects—for example, "not all Wikipedias have the request for comment
system so familiar to English Wikipedia contributors." She wants more bridges created between Meta and the projects, so that important movement-wide topics and decisions are easier to find and understand. John echoes this: "Movement-wide RFCs and surveys on meta have often been launched before translation into major languages and without notices sent to village pumps and mailing lists, disenfranchising large segments of our community."
To what extent should two-way communication between Meta participants be assisted by funded translation services, and would this be viable and practical?
The responses were generally cautious and suggest that there's no clear strategy for improving inter-linguistic communication on Meta. Leigh says: "if money were no object, of course funded translation would be a necessity. However, with 270+ languages, it's not feasible." Liam is against paying people to translate, pointing to the huge amount of text at issue: "Are we also willing to pay to translate fundraising info, monthly reports, important software messages, and election information? What translation is paid for and what is not would need consensus." On this issue, Michel refers to the goals of increasing participation and reach in the Strategic Plan. Volunteer translation is the default, he says, but for important matters the Foundation should consider paying for translation—at least for the major languages and the major pages. "That shouldn't cost an arm and a leg and it would lower the threshold for [participation by] non-English speaking editors." María believes "some crucial questions or issues for the movement should have translations assured to the main languages at the very least." Her preference would be "to empower our translation volunteers".
In Phoebe's opinion, "the Board should put in more effort to make sure critical Board information is widely translated. Paying for translation is a complex proposition: it varies hugely in quality and must be reviewed by someone knowledgeable about the projects; however, I'd support it as an aid to our volunteer translator community for some documents." Tom says: "Perhaps. But a proper cost–benefit analysis would be needed. The first step would be to actively try and create a volunteer translation service, perhaps as a thematic organisation. Another option would be to look at hiring some of our more prolific volunteers to conduct translation, similar to how the Foundation has hired other community members."
Samuel set out a specific plan: "We have a body of talented multilingual editors and translators. We should commit ourselves to supporting two-way communication across a set of five or six major languages, and invest in expanding and supporting our translator community. This may include funding translation as a temporary stopgap, but should focus on honoring and recruiting [volunteer] translators, and providing them with training, tools, and a visible portfolio. This has been done with other communities, such as GlobalVoices, which handle higher volumes of text than the daily throughput of Meta."
John says: "I do not support the Foundation's outsourcing of translation to US for-profit organisations. ... In my experience, volunteers are happy to do translations if they have sufficient time." He wants "a small grants program" to facilitate discussion and dissemination of upcoming changes for editors in a particular language, and to coordinate volunteer translation, or when necessary do the translation themselves or outsource it to a local translator. He believes the WMF should fund chapters to help volunteers without English-language skills who wish to participate in an English-dominated Meta discussions.
Two second-language speakers were concerned about the quality of paid translations they have seen. Milos says that outside the key documents, translations are often "not of substantial value. English is global lingua franca and a person wanting to participate in global community has to have functional knowledge of English." Jerome-Yu has experience with the translation service hired by WMF for English to Chinese: "I still need to fix a lot, there's unfamiliarity with the terms, and ... the service is not really viable." In some cases, he says, fixing up paid translations "is actually more difficult and tedious than doing our own all over again."
The chapters (blue) are being allocated increasing proportions of donors' funds, year by year.
The chapters are independent organisations founded by interested individuals to support and promote Wikimedia projects. While projects such as Wikipedia are globally structured by language, most chapters are set up as nation-state-based organizations. The chapters have grown significantly over the past five years, now numbering 39 and accounting for almost a third of the movement's expenditure of donors' funds. The Foundation's financial reforms last year put chapter funding and scrutiny under more centralised control by insisting on the publication of regular activity reports on Meta. The Foundation has introduced two new types of entities that are not tied to any geographical region—user groups and thematic organisations—although few of these have been launched at this early stage.
Is the substantial funding allocated to chapters yielding value for money in terms of the quality and scope of WMF sites related to developing-world languages?
Despite the specific terms of the question, candidates responded to a wide spectrum of self-selected responses—some focusing on chapters, others on languages. Leigh says: "No. There is too little oversight to how the money is being spent or even if a chapter is following Wikimedia rules and values. There are far too many stories of people building little "kingdoms" of themselves and even marginalizing those they do not like."
Milos doesn't think we could create something "substantially better than chapters are", although he says it is difficult to predict how chapters, thematic organisations, and user groups will play out against each other in the coming years. Liam is upbeat about chapters: "I have always championed the raising of the bar in terms of professionalism and programmatic-reporting by chapters—and this is what we have seen in the last year with the introduction of the FDC. But this should be matched by programs to support the different stages of development of an organisation." He believes that "only when we have different metrics for different levels of organisational capacity can we make judgements on whether a chapter (or any project for that matter) is 'yielding value for money'."
María has confidence in the movement's grantmaking ability: "There are controls in place that make it possible to deny anyone, be it the developing world or elsewhere, continued access to WMF funding if they fail to report and fail to demonstrate impact." Phoebe is "really pleased with the general direction of the new Affiliations Committee and funding models (of grants and the FDC); I strongly supported more community review for all Wikimedia groups and activities, and we are slowly getting there." Jerome-Yu is generally supportive of chapters, but we should "also empower other entities, and even individuals, to fill the gap the chapters may have missed".
Samuel says: "We need to allocate more funding to chapters in the developing world in order to better advance the quality and scope of the projects in local languages. The funding allocated to chapters in the developing world has supported many of the largest outreach, writing, and content partnerships in local languages. ... Chapters in the developed world receive more than 90% of all chapter funding. They have supported some excellent projects to improve offline access to knowledge, he says, but local language quality and scope is not their primary focus."
On support for the developing-world languages on WMF sites, Tom says: "There are some great success stories within the developing world ... [like the] Kiwix and the Africa projects of European chapters. [While] a lot of funding goes into Western chapters, now we need to find ways to fund on-the-ground outreach in the developing world. The chapter model does seem to work, so why not continue to use it." Phoebe says: "we have a long way to go", but that we need to balance this against the "need to retain our editor base for our more established projects, which many chapters have been focusing on. ... For many parts of the developing world, there's a lot of potential for supporting the creation of user groups—more lightweight structures that can support particular initiatives."
John is critical of some aspects of the centralised approach: "Most Wikimedia funding for languages in the developing world remains under the operational budget and control of the US Wikimedia Foundation. Some chapters have funded programs to help the languages of the developing world and have had good return on investment, but most funding given to chapters (especially with the new FDC model) tend to be spent in the first world, or on program activity that benefits the English Wikipedia and other European-language projects." Jerome-Yu would rather there were less administrative burden for entities to get funding, whether from the WMF or locally, citing issues with charity status and the volunteer effort necessary to gain WMF funding. He wants to see chapters "be professionalized to free up volunteers to deal with the quality and scope things."
Is the nation-based model for chapters working, given that our editing communities define themselves through languages that transcend national borders?
Existing chapters, both founded (dark blue
) and approved (dark turquoise
), planned chapters (green
), and chapters in discussion (light blue
) as of 13 March 2012.
Here again, the responses were strikingly different. Milos says, "mostly not, but that's highly politicized question without possibility to be solved in the near future." For Leigh, the nation-based model has worked best in Europe, and is useful to an extent; but "the Foundation has started to realize some of those limitations". Michel's answer is a "pragmatic" one, based on history: "Chapters owe at least part of their existence to the idea that one could physically meet up. Nationhood is a 19th-century invention that we cannot yet do without; it's a reality we all live with."
Samuel regards the geographically defined chapters to be "one of many models for movement entities that we need to integrate our work with the rest of the world. Social infrastructure—from volunteer communities and media to pools of content or funding—has developed along geographic lines, and is often drawn up along them." But other models that transcend geography are also urgently needed, he says, "which is why I pushed to expand the set of models we recognize for Wikimedia affiliation."
Liam highlights the advantages of having chapters correspond to legal jurisdictions with which so much of the freeing of cultural products is involved. "We often need a non-profit organisation to have a bank account, or to talk with the local government, or to form partnerships with national cultural institutions." Phoebe's view resonated with this: "the national model ... works with the governmental and legal constraints. ... But there's a big gap—most language editions of our projects are edited by a diaspora community of people located around the world, and I'm not sure that we've managed to tap into that effectively with current outreach efforts. I think the new thematic organizations/user group model, working with chapters, is a very promising one to address cross-border interest groups and language issues." María also sees the value in the nation-state model: "Most organizations, particularly public ones, are country-based. And most of the work chapters do is offwiki outreach, so in that context it can't just be understood from a language point of view."
While broadly supportive of "the current model for chapters and other affiliates", John believes it is "far from optimal". He sets out a possible process by which entities should be nurtured and funded once approved by the Affiliations Committee, "with the WMF providing support in finding suitable candidates if necessary for expert positions, and the candidates also being identified to the WMF as is required for stewards."
Might the new entities—thematic organisations and user groups—become more relevant than chapters to the Board's medium-term strategies?
The prevailing mood among candidates is that it's too early to tell, and that shouldn't be a competition between different entities. For Francis, what matters is "the integration of accountability" among these various layers. Leigh cites "the community of medical Wikipedians and the spinning off of US/Canada education activities from the Wikipedia Education Program" as showing promise already for thematic organisations. "However, there have already been conflicts with geographical chapters who feel threatened by these new group types." Milos predicts that chapters will be no less relevant than thematic organizations during this decade. Liam's view is that at this early stage "it's hard to tell how effective or popular [thematic organisations and user groups] are going to be. More important ... is to give every type of organisation every possible chance to succeed. ... Either way it's not going to a one-size-fits-all solution." María believes that "ideally, affiliates will complement each other and work together when needed."
Tom points out that "chapters and thematic groups both have strong relevance to the ongoing strategic plans. It's not an either/or situation". He raises an interesting question: "How do [these new groups] properly interact with other language projects? Do we end up with multiple thematic groups for different languages?" For John, "chapters and thematic organisations are complementary [to chapters.] The WMF should always be a stakeholder in any large program to provide oversight and fill any gaps in expertise". Jerome-Yu believes all types of entities should be treated equally, like chapters, but that clearer guidelines are necessary to attract more applications for thematic organisations and user groups.
Michel, again, steps back to take a big-picture view: "I think we'll find that geography is overrated. That's not to say it's not important: non-virtual human interaction is great, and I've benefited greatly from meeting people at Wikimania and the Education Summit. But I think that the kind of professionalization that will, for instance, increase article quality (one of the strategic plans) will have to come from thematic organizations."
Editors are invited to ask questions of the candidates on Meta.
- This article has been updated to reflect the new election timeline, which has been delayed for one week.