The Wikimedia Foundation Community Resources team, best known for operating the Wikimedia Foundation grants program, has announced major changes in the sort of projects which it will fund and also published its 2018–19 annual report. In a July email to the Wikimedia-l email list, the WMF representative announced that the rapid grant process for proposals of up to US$2,000 would no longer sponsor travel support, equipment purchases, or Wikimedia meetups. Wiki editing parties, photowalks, promotion campaigns, video campaigns, and "other" are still fundable causes. All submissions must happen between days 1–15 of a month. Lastly, the process reserves certain months only for submissions of certain campaigns, as listed:
Besides the reforms, the grants team presented the newly published Wikimedia Grants annual spending report for fiscal year 2018–19. The WMF sent US$7.4 million to 86 countries in 394 grant proposals. To meet Wikimedia community report demands the WMF also has sorted the various proposals in categories of contemporary interest, as follows: 9% went to individual applicants rather than organizations; 33% of the money went to projects in the Global South (emerging communities, less wealthy economies), and 4% went to projects which address the Wikimedia gender gap. 64% of the money went to Wikimedia affiliate organizations who received it as part of their annual plan grant overseen by the Funds Dissemination Committee or the simple annual plan grants process.
To anyone who evaluates this report, consider and speak out about what information the Wikimedia community needs to encourage good understanding of itself and planning for the future. Examples of data which would be currently hard to access include reports of funding by region (each of North America, South Asia, Europe, South America for example); evaluation of the extent to which projects delivered promised outcomes; relative impact of funding administration versus programming versus infrastructure; and counts of Wikimedia users participating in various projects. The Wikimedia platform is digitally native and produces huge amounts of metrics to interpret how grant-funded interventions change it. Anyone with ideas for what to do with such reports has the standing Wikimedia community invitation to discuss and suggest in all relevant community forums.
On 14 July 2019 the Pune Mirror reported news that the Affiliations Committee (AffCom) would revoke official recognition and WMF grant funding eligibility of Wikimedia India (WM India) as a Wikimedia movement affiliate. WM India members have objected to the decision and posted "Wikimedia India's Demand For A Fair And Transparent Hearing" in Wikimedia-l, as well as continued conversation in the India focused wiki mailing list. Right now, there is no particular communication process in place for derecognition of Wikimedia community groups and various social pressures prevent WM India, WMF, and AffCom from speaking openly in the wiki. These transparency challenges led one Wikimedia community member to remark being "surprised that the community has to learn about the de-recognition from the press".
While AffCom has derecognized chapters in the past and each case is somewhat different, this case is more unusual because of Wikimedia India's special place among Wikimedia affiliates. Typical Wikimedia chapters design their own programs at the community level and present them in their region. In India, the WMF has invested much more money operating programs of foreign origin within India, with the role of WM India being to represent the local community in overseeing and approving these projects. WMF programs in India include operating the first Wikimedia Foundation satellite office there from 2010–2012, piloting the Wikipedia:India Education Program there from 2012–2014, operating Wikipedia Zero in India from 2012–2016, for its television video campaign for Wikipedia awareness in Hindi language in 2017, and for its unique sponsorship relationship from 2013–present with the Centre for Internet and Society as an expert organization outside of the Wikimedia community. In these and many other collaboration attempts, culture clashes have been routine. While difficult to summarize, one commonly voiced criticism is that that WMF challenges the autonomy of the regional community by investing more money in WMF-led programs than local-community led programs. There is no organized repository of journalism, documentation, and accounting about WMF investment in India, and perhaps the narrative in the memory of the community there is not published anywhere to be found.
The revocation of WM India's status recognition coincides with several other major Wikimedia projects in India. One is the Wikimedia Foundation's July 2019 India fundraising announcement in which the organization describes plans to seek donations and sponsorship in India. Another is Google's sponsorship of the Wikimedia Foundation to advance the "Supporting Indian Language Wikipedias Program". In remarks to The Signpost a WMF representative shared that the fundraising team and AffCom had taken action on relations with India independently of each other. Also, the collaboration with Google has been ongoing for some years, and the new development is a progression of years of past engagement in the region as reported in various third party media sources.
The Signpost welcomes the English Wikipedia's newest administrators, Johnuniq, Valereee and Kosack.
It took me a while to realize that I shouldn't edit articles about politicians during an election year unless I was prepared for a knock-down, drag-out fight. After my third edit on a political article I finally figured out "Oh, they're probably all going to be like this." With the already vicious tone of the U.S. Presidential campaign, we may be in for yet tougher times. This month we take a tour of how the media is reporting the difficulties of editing political articles on Wikipedia. - S
Many Wikipedians may be too busy building our encyclopedia, or dealing with our usual squabbles, to see the wide range of topics involving Wikipedia that are covered by the media. The odd bits this month include a book review, a country rapper in a promotional video, a Commons photographer accused of "predatory" copyright lawsuits, Gaelic Football statistics, Indian police forces, the British schools curriculum, and our inclusion in a lunar library.
In June, when the Fram drama was at its peak, three editors with advanced permissions reverted office actions as a result of community discussions concluding that the Fram ban as it was carried out by the Foundation was improper. The first revert was by Floquenbeam, who unblocked Fram. The WMFOffice role account then reblocked Fram and desysopped Floquenbeam, and posted a message stating that both actions were office actions. Following this, Bishonen unblocked Fram again and bureaucrat WJBScribe restored Fram's sysop flag. (Neither unblock technically affected the ban; even though they had been unblocked, if Fram were to edit, they would have been globally locked. After Fram accidentally made an edit on the English Wikipedia, which they quickly self-reverted, they requested a reblock on EnWP in order to prevent a global lock.) The WMF did not revert either of these actions. Not too long after, WJBScribe resigned the crat tools, and WJBScribe and Floquenbeam requested to be desysopped. These actions resulted in an RfC asking whether WJBScribe and Floquenbeam's actions would be considered "under a cloud". (If this were the case, a new RfA/RfB would be required if they wanted to regain their tools; otherwise, they could just request them back at WP:BN.) While the RfC has not yet been closed, it is likely to conclude that neither action prior to resignation was "under a cloud". However, Floquenbeam has voluntarily run a reconfirmation request for adminship. Voting on the RfA ended on July 29 with 73.6% support; a bureaucrat chat is currently underway.
Following the Wikimedia Foundation's ban on Fram but before the release of the statement from the board, the message box at the top of Wikipedia's page on office actions was changed from "policy" to "information page". A talk page discussion followed about whether it should continue to be listed as an English Wikipedia policy, have some sort of box indicating that it's not a local policy but it is a global one, or redirect it to the global policy page on Meta. Around this time, it was pointed out that in 2017, the Wikimedia Foundation had overwritten the page to more closely resemble the Meta version of the page. One month and several infopage box variants later, consensus was reached that the page should be demoted to an information page. Following the close, the page was converted into a summary of the global policy, featuring historical background and a link to the full policy on Meta.
Wikipedians have continued to discuss changes to Wikipedia's policy on restoration of administrator privileges (resysops), following contentious resysop requests by users who had been largely inactive for the last several years (one admin who requested a resysop hadn't edited in the last two years and had made fewer than 30 edits in the last decade). TonyBallioni started the first phase of the RfC, which asks for the community's general opinion on whether the resysop policy should be stricter, looser, or stay more or less the same. A later RfC may be set up to examine specific proposals, depending on how this one goes.
Last issue, we reported on a proposal to desysop all the admins on the Azerbaijani Wikipedia. On July 3, the RfC on Meta was closed by steward Mardetanha with the following statement (typos have been corrected):
After some discussion with fellow stewards, I am closing this RFC. There is mild support to desysop all admins on azwiki, but from what we see it is not going to fix the problem, as some of the admins have been chosen recently, and one has been reelected during this RFC. Desysopping all of them is not going to do any good to the community, but we have to ask all azwiki admins to be more careful and make sure that they are following rules and guidelines and our norms. As a result of the RFC, one problematic admin [Cekli] has been desysopped, and if he wants to gain his access back, he has to go through another RFA, but to fix the problem and make azwiki better, I will be traveling to Azerbaijan this month (with WMF support) to help and discuss the issues in more detail, and we will try help azwiki admins to not repeat mistakes from the past. I am really hopeful azwiki admins have heard the concerns and will do their best to address them.
This close led to accusations of supervoting from several English Wikipedia editors, with several issues being raised with the close:
Proposals have been considered to overturn this close, but none have been officially made as of press time.
After the Fram case was declined, Arbitration Committee decided to review reversion of office actions generally, with 121 comments by members of the community. The committee decided by motion to "note without comment" the events surrounding last month's office actions and subsequent flurry of ENWP blocks and unblocks. The committee in their motion has advised the community "that administrators and bureaucrats are normally expected not to act when they know they do not have all of the relevant facts, and that this is especially important with regard to office actions where those facts may be highly sensitive." The committee further warned that future "wheel warring may be grounds for removal of administrative rights by the committee".
The second case accepted, titled "Antisemitism in Poland", concerns articles like History of the Jews in Poland, but arbitrator AGK said in the opening phase "even we are not sure yet what the dispute entails" other than a clear clash between (primarily) two editors. The workshop phase closed 30 June, and the committee is now deliberating offline on a decision. As of writing deadline, there have been no principles, findings of fact, nor remedies for the case posted.
Heinrich Berann's panoramic maps of four National Parks are available in high resolution from the U.S. National Park Service and now from Commons. The maps – works of art might be a better term – were originally published between 1987 and 1994. Berann was an Austrian who became known to Americans through his work for National Geographic magazine. Click on the images for a much larger view.
There are many opportunities to discuss bad news, problems, and concerns in the Wikiverse, and I think that having candid discussions about these issues is often important. Many days I spend more time thinking about problems than about what is going well. However, also I think that acknowledging the good side and taking a moment to be appreciative can be valuable.
I encourage you to add your comments about what's making you happy this month to the talk page of this Signpost piece.
At the moment I am not particularly happy, but I am grateful. I feel that the English Wikiquote of the day for 30 June 2019 is appropriate to share for contemplation in the context of current tensions.
"What I'm saying here is not, I agree, poetry, as poems should be written rarely and reluctantly, under unbearable duress and only with the hope that good spirits, not evil ones, choose us for their instrument."
— Czesław Miłosz, Polish writer, diplomat, and recipient of a Nobel Prize in Literature
I am grateful for hundreds of colleagues who have participated generously, courageously, and in good faith to advocate for what they believe are the best interests of the Wikiverse projects. (I request that in this Signpost article and its talk page we avoid advocating for particular outcomes so that we remain on topic.)
This past weekend I took time offline to go bicycling, including bicycling on part of the Burke-Gilman Trail. There are photos of the trail and surrounding scenery on Commons, and there is also a Wikidata item for the trail.
I think that time offline for reflection was good. I had in-person conversations that helped me to process recent events on English Wikipedia.
I would like to highlight the recognition of the Dehalvi Wikimedia Community User Group and Wikimedia Community User Group Math.
I thought that this thread on Wikitech-l was interesting: "Difference between #goal and #epic".
Last week there were at least two notable anniversaries in space exploration.
I enjoy looking at English Wikipedia featured pictures, including these subcategories of space featured pictures:
My guess is that many of us have looked up at the sky at night and thought about the Moon, the stars, and the planets. Maybe you have seen shooting stars, comets like Hale–Bopp, or the Milky Way. If you have the opportunity to try astronomy or night photography, I encourage you to do so.
I would like to close this section with two quotes.
First, the famous quote by astronaut Neil Armstrong as he first set foot on the Moon: "That's one small step for [a] man, one giant leap for mankind."
Second, a quote by amateur astronomer Thomas Bopp: " I--I'll never give up looking at the skies. There's so many beautiful things there. And I love to share that with people."
I’ve always appreciated my fellow Wikimedians. But when I do my normal editing, in my spare time, I respect them as my peers. I’m constantly grateful for the work Tegel does to keep Swedish Wikipedia free from spam and vandalism. In awe of Yger’s ambitious projects related to Swedish geography and administrative districts. Impressed by ArildV’s illustrations. Still miss cooperating with Åsa L on the Swedish science fiction articles. But in my work for the Foundation, there’s a shift – a feeling of being privileged to be part of this project full time.
Take Tech News, for example. Most weeks I write Tech News, a weekly technical newsletter, in an attempt to make it possible to get a quick overview of important technical changes that will affect Wikimedia contributors. Every week it is translated and distributed in a number of languages, typically 15–20 and every week I try to use the thanks function to unobtrusively show my appreciation. The act becomes routine: open the history of every language, thank the person, go to the next translation, thank, thank, thank, thank, thank. But every week that action is fuelled by a profound feeling of gratitude. Years into writing the newsletter I still count myself fortunate to get to work with a group of people who, in many cases, week after week turn up to help Wikimedians be aware of how the tools they use keep being developed. There’s no simple way to tell a pseudonymous user – potentially someone you’ve hardly interacted with except through the thanks button – the deep affection you have for the usernames who turn up to make the translations for the various Wikimedia communities, how much you care about them. But as a collective, they remain one of my favourite group of persons in the world. Every week, I use the same words to thank them. It’s a copy and paste. I hardly read the words. But every week, I mean them. If someone goes on hiatus for a few months and comes back, it makes me genuinely happy, like seeing an old friend again.
I despair at times, of course, like almost everyone else who spends a fair chunk of their life thinking about the wikis. I sometimes think we fail to live up to our principles of civility and assuming good faith. I sometimes think we’re making it far too difficult to edit. I sometimes fear we’re not preparing for the world of tomorrow, given that we still edit in a desktop-first mindset when most of our readers access our information in other ways. It’s no fun being met with a lack of assumption of good faith when working on a project that’s crept into almost every aspect of my life and been part of me since my teens. But most of these things are normal parts of being a very small cog in a big machine. There are thousands and thousands and thousands of us, trying to make the wikis a better place. I’m just one of them.
It should become mundane, I suppose. I’ve been editing since 2004, been working for the Foundation since 2015. I’ve seen the good and bad parts of the communities, the beautiful and the ugly. The magic should fade. But it doesn’t, because Wikimedians don’t stop doing wonderful things in the name of creating a repository of knowledge, there to be used by anyone with internet access and even for some without.
And I continue to be grateful. Not always for specific details of my work, because the job is like any other job full of things one gets paid to do because one wouldn’t want to do it for free. But for being part of this. For working with the Wikimedia communities.
Skillful translations of the sentence "What's making you happy this week?" would be very much appreciated. If you see any inaccuracies in the translations in this article then please {{ping}} User:Pine in the discussion section of this page, or boldly make the correction to the text of the article. Thank you to everyone who has helped with translations so far.
What's making you happy this month? You are welcome to write a comment on the talk page of this Signpost piece.
Videowiki is a Wikimedia project that allows groups of editors to collaboratively write a script for a video followed by the automatic generation of the video based on the script. It works similarly to the way Wikipedia editors write articles together. To produce a video, editors link to Wikimedia Commons photos and videos right within the script. VideoWiki volunteers have produced 24 videos in English since the project's start in 2018. The existing videos are all on medical topics.
The WikiProject addresses issues including the difficulty of collaboratively editing videos, updating videos based on consensus, attribution, and translation.
Wikipedia's community of medical editors and various academic sources report that Wikipedia is the single most popular source of health information.[1] The VideoWiki team believes that increasing the accessibility of Wikipedia's content will better inform the large audience which seeks information from Wikipedia. Adding video to Wikipedia might appear to be a cosmetic change to a reader in the developed world, but for many in developing nations, video is an essential part of accessibility. In India, for instance, illiteracy runs at 25% and video has a greater presence as an information source. Taking the maxim that "the medium is the message", Wikimedia projects demonstrate respect by responding to audience demand with content in the popular format those demographics prefer. To make Wikipedia widely available in India — to truly let every single human being freely share in the sum of all knowledge — including information through video is required.
YouTube is often the first stop for those seeking information, but it struggles with fringe content. YouTube itself uses Wikipedia as part of its quality control process, which is one way that high quality Wikipedia content increases the quality of other sources. Wikipedia in general is more transparent and has a better reputation for reliability. Video on Wikipedia is an opportunity to provide an alternative which is both reliable and balanced.
The same populations which have challenges accessing Wikipedia are also disproportionately affected by diseases that can be prevented through education.[2] They also make greater use of informal healthcare providers, who are in equal need of guidance. These providers may not have formal education or be able to critically appraise health care information.[3] Neonatal hypothermia,[4] diarrheal diseases, and pneumonia[5] all cause childhood death, which could be prevented by the distribution of high quality information.[2] One sobering example is that only 1 in 5 children under 5 years of age in India receives oral rehydration therapy when affected by diarrhea.[6] Wikimedia Medicine is scheduled at Wikimania 2019 to present the Wikipedia's efforts to advance the United Nations Sustainable Development Goals (SDGs) by promoting education to reduce childhood deaths by 30 million and maternal mortality by 66%.[7] Although quality education is relevant to all 17 SDGs, SDGs 3 (Good Health and Well-being) and 4 (Quality Education) are most connected to VideoWiki's goal to save lives.
At WikiProject Videowiki, contributors match scripted text with media from Commons including video, GIFs, and static images. The product is a Videowiki script built within the Wikipedia namespace. The script has sections defined by headers after the model of a Wikipedia article, which means that just as in Wikipedia, editors can edit a sentence, a subtopic, or the entire script as they choose. The VideoWiki platform reads the text and shows the media listed within the section. At every new header the video transitions to the next subtopic. A text-to-speech engine (TTS) dubs in narration by default, but anyone may also overdub a video by uploading an audio file to Commons. The TTS service currently supports 5 languages but has modules to potentially support 21 languages. Once edited to the satisfaction of the community, the video can be compiled and uploaded to Commons as a WEBM file, where article editors can choose whether or not to add it to a written article. The all-on-Wikipedia process leaves VideoWiki videos subject to the same merciless editing as any other content on the Wikimedia platform. Just as written articles become more nuanced and readable with time, so should videos when scripts and visuals are supported by the pillars of Wikipedia.
All of the problems we see on English language video websites, including creating information in error, passing on misinformation, advertising, and charlatanism can be worse on non-English sites. A reliable voice in a transparent medium can improve quality of information and trust in the source. VideoWiki's design allows rapid translation in the same way English language propagates articles which the community at WikiProject Medicine develops. Translation task force members, or anyone else for that matter, can translate and move scripts to non-English language pages for TTS engines to recreate as videos. Since Wikipedia supports 301 languages (far more than TTS engines), overdubbing can also be done for non-TTS supported scripts.
Wikipedia has been proven to be a useful complement to traditional textbooks for knowledge transfer to medical students.[8] Since 2007 we have grown Wikipedia's medical content to over 220 thousand articles, in 281 languages and supported by 2.3 million references. The next major underserved demographic to serve with greater accessibility is the 50% of the world's population with limited Internet access and the 15% who have challenges reading. Getting life saving information into the last mile in a format and language people can understand is critical if Wikipedia is to best support the UN SDGs.
The primary focus of VideoWiki over 2018–2019 has been creating a collaborative, all-on-Wikipedia video editing platform. With the platform now designed and supported by the pillars of Wikipedia, the next need is to improve the quality of videos. To that end, the VideoWiki team hopes to grow the community that will rally around the application of wiki values to video production to:
At Wikimania 2019, where the theme is "Wikimedia and advancing the Sustainable Development Goals", let us focus our efforts on reaching the world's most vulnerable people by bringing Wikipedia's content into the last mile.
The past few years have seen an explosion of journalism, scholarship, and advocacy around the topic of ethical AI. This attention reflects a growing recognition that technology companies often fail to put the needs of the people who use machine learning (or "AI") technology, and of society as a whole, ahead of their business goals.
Much of the public conversation on the topic of ethical AI has revolved around general principles like fairness, transparency, and accountability. Articulating the principles that underlie ethical AI is an important step. But technology companies also need practical guidance on how to apply those principles when they develop products based on AI, so that they can identify major risks and make informed decisions.
What would a minimum viable process (MVP) for ethical AI product development look like at Wikimedia, given our strengths, weaknesses, mission, and values? How do we use AI to support knowledge equity, ensure the knowledge integrity, and help our movement thrive without undermining our values?
The Wikimedia Foundation's Research team has begun to tackle these questions in a new white paper. Ethical & Human centered AI at Wikimedia takes the 2030 strategic direction as a starting point, building from the observation that "Developing and harnessing technology in socially equitable and constructive ways—and preventing unintended negative consequences—requires thoughtful leadership and technical vigilance." The white paper was developed through an extensive literature review and consultation with subject matter experts, and builds off of other recent work by the Foundation’s Research and Audiences teams.
The white paper has two main components. First, it presents a set of risk scenarios—short vignettes that describe the release of a hypothetical AI-powered product, and some plausible consequences of that release on Wikimedia’s content, contributors, or readers. Second, it proposes a set of improvements we can make to the process we follow when we develop AI-powered products, and to the design of the products themselves, that will help us avoid the negative consequences described in the scenarios.
The risk scenarios are intended to spur discussion among AI product stakeholders—product teams, research scientists, organizational decision-makers, and volunteer communities. Scenarios like these can be used in discussions around product planning, development, and evaluation to raise important questions. They can help us uncover assumptions that might otherwise be left unstated, and highlight tensions between immediate goals and foundational values. The goal is to help people grapple with these trade-offs and identify alternative approaches that minimize the risk of unintended consequences.
Each of the six risk scenarios address a complex ethical issue that AI products can make worse—like the risk of reinforcing systemic bias, discouraging diversity, and creating inequity in access to information. They also help uncover subtler issues—like the risk of disrupting community workflows or subverting editorial judgement when we automate processes that are currently performed by people. While the negative outcomes described in the risk scenarios are hypothetical, each one is based on a realistic Wikimedia-specific AI product use case.
The eight process improvement proposals described in the white paper lay out courses of action that Wikimedia can take when developing AI products. Following these recommendations can help researchers and product teams identify risks and prevent negative impacts, and ensure that we continue to get better building AI products over time.
Some of the proposals focus on improving our software development process for AI products. They describe steps we should take when we develop machine learning algorithms, assess potential product applications for those algorithms, deploy those products on Wikimedia websites, and evaluate success and failure.
Other proposals focus on the design of the AI technologies themselves, and the tools and user interfaces we build around them. They describe ethical design patterns intended to allow the readers and contributors who use our AI products to understand how the algorithms work, provide feedback, and take control of their experience.
The technological, social, and regulatory landscape around AI is changing rapidly. The technology industry as a whole has only recently begun to acknowledge that the ethos of "move fast and break things" is neither an effective nor an ethical way to build complex and powerful products capable of having unexpected, disruptive, and often devastating impacts on individuals, communities, and social institutions. As a non-profit, mission-driven organization with a global reach, the Wikimedia Foundation must hold itself to a higher standard. We can't afford to build first and ask ethical questions later.
In many ways, the Wikimedia movement is ahead of the game here. We are a geographically and culturally diverse group of people united by a common cause. We already practice the kind of transparency, values-driven design, and consensus-based decision-making that are necessary to leverage the opportunities presented by AI technology while avoiding the damage it can cause. Because the Wikimedia Foundation serves as a steward for the code, content, and communities within the Movement, it is important that we consider the kinds of risks outlined in this white paper, and adopting solutions to anticipate and address them.
You can read more about this project on Meta-Wiki and view the white paper on Wikimedia Commons.
A monthly overview of recent academic research about Wikipedia and other Wikimedia projects, also published as the Wikimedia Research Newsletter.
In the recent research paper, "The Most Influential Medical Journals According to Wikipedia: Quantitative Analysis"[1], the authors sought to determine the ranking of the most cited medical journals in English Wikipedia by evaluating the number of days between article publication and their citation. They analyzed 11,325 medical articles in Wikipedia that included citations from 137,889 articles from over 15,000 journals. They found that the top five journals cited, in order, include The Cochrane Database of Systematic Reviews, The New England Journal of Medicine, PLOS One, The BMJ, and JAMA: The Journal of the American Medical Association. This ranking, along with the next 25 journals they found in the study, was related to the highest ranked journals based on Journal Citation Reports and the Scientific Journal Ranking, yet due to the Wikipedia focus around reviews and meta-analysis, there were some clear differences. While evidence of recentism was identified, all journals that appeared in this study are directly related to medicine. The researchers suggested that similar studies be applied to other disciplinary areas, especially as "Wikipedia editing increases information literacy,"(p. 9) while also being more widely used by academics.
The authors in this study, "Detecting pages to protect in Wikipedia across multiple languages,"[2] wanted to understand aspects of page protection due to concerns related to vandalism, libel, and edit wars, and determine if tools could help automate this process. The researchers studied two data-sets: the 0.2% of pages which were protected in April 2016, and a similarly-sized random selection of unprotected pages. Their system performed well in predicting candidates for protection and has been developed to work across languages. The researchers hope to test this tool in live Wikipedias as their next step in automated page protection tests. (See also an earlier paper by some of the same authors: "DePP: A System for Detecting Pages to Protect in Wikipedia")
In "XLORE2: Large-scale Cross-lingual Knowledge Graph Construction and Application,"[3] from the inaugural issue (Winter 2019) of Data Intelligence—a joint venture between MIT and the Chinese Academy of Sciences—the authors explore better methods of mapping concepts between Chinese and English in XLORE2, whose taxonomy "is derived from the Wikipedia category system." Fewer than 5% of the over 100,000 infobox attributes in English Wikipedia are matched in Chinese Wikipedia. The authors discuss methods for improving the quality of typological relationships derived from English Wikipedia. Besides English and Chinese Wikipedia, their knowledge base also uses data from Baidu Baike and Hudong Baike.
In "Automated Detection of Online Abuse"[4] researchers applied machine learning to analyze the user behavior of blocked accounts on English Wikipedia. This analysis identified activity patterns of misconduct and modeled an WP:automated moderation system which could monitor unblocked user accounts to detect patterns of misconduct before human patrol identifies them. The research team's Wikimedia project page includes supplementary materials, including essays on ethical considerations for technological development in this direction.
See the research events page on Meta-wiki for upcoming conferences and events, including submission deadlines, and the page of the monthly Wikimedia Research Showcase for videos and slides of past presentations.
Other recent publications that could not be covered in time for this issue include the items listed below. Contributions, whether reviewing or summarizing newly published research, are always welcome.
From the abstract:[5]
"This paper will look at how cultural heritage organizations [GLAMs] can work with Wikidata, positioning themselves to become a more useful and accessible knowledge resource to the world."
This paper[6] provides a unified framework to extract topic specific subgraphs from Wikidata. These datasets can help develop new methods of knowledge graph processing and relational learning.
From the abstract:[7]
"...we propose a novel method for estimating socioeconomic indicators using open-source, geolocated textual information from Wikipedia articles. We demonstrate that modern NLP techniques can be used to predict community-level asset wealth and education outcomes using nearby geolocated Wikipedia articles. When paired with nightlights satellite imagery, our method outperforms all previously published benchmarks for this prediction task, indicating the potential of Wikipedia to inform both research in the social sciences and future policy decisions."
This paper[8] offers an effective method to build a comprehensive knowledge base by extracting information from Wikipedia infoboxes.
From the abstract:[9]
"The relationship between editors, derived from their sequence of editing activity, results in a directed network structure called the revision network, that potentially holds valuable insights into editing activity. In this paper we create revision networks to assess differences between controversial and non-controversial articles, as labelled by Wikipedia. Originating from complex networks, we apply motif analysis, which determines the under or over-representation of induced sub-structures, in this case triads of editors. We analyse 21,631 Wikipedia articles in this way, and use principal component analysis to consider the relationship between their motif subgraph ratio profiles. Results show that a small number of induced triads play an important role in characterising relationships between editors, with controversial articles having a tendency to cluster. This provides useful insight into editing behaviour [... and also] a potentially useful feature for future prediction of controversial Wikipedia articles."
From the abstract:[10]
[Wikidata's] data quality is managed and monitored by its community using several quality control mechanisms, recently including formal schemas in the Shape Expressions language. However, larger schemas can be tedious to write, making automatic inference of schemas from a set of exemplary Items an attractive prospect. This thesis investigates this option by updating and adapting the RDF2Graph program to infer schemas from a set of Wikidata Items, and providing a web-based tool which makes this process available to the Wikidata community."
From the abstract:[11]
"..."
"This talk explores the role Wikidata [...] might play in the task of assembling biodiversity information into a single, richly annotated and cross linked structure known as the biodiversity knowledge graph [...] Much of the content of Wikispecies is being automatically added to Wikidata [...] Wikidata is a candidate for the 'bibliography of life' [...], a database of all taxonomic literature."
This paper[12] presents a system for detecting newsworthy events occurring at different locations of the world from a wide range of categories using Twitter and Wikipedia.
Described as "the first comparative analysis of deep-learning models to assess Wikipedia article quality", this paper[13] observes that the most important quality indicators (model features) appear to be the following: That the article has been reviewed by "quality" editors, the edit count of the contributors, and the number of its translations.
From the English abstract:[14]
"In this empirical study, we examine a facet of [Wikipedia's] governance: the modalities of construction of two rules related to the citation of sources. We show that these rules are discussed and written by a minority of contributors who are particularly involved. Thus, in Wikipedia, there is no “political class” cut off from the ground. The modalities for the elaboration of the two rules are studied and discussed using Ostrom’s theory of the Commons."
From the abstract:[15]
We show on both Romanian and Danish wikis that using only the edit and their distribution over time to feed clustering techniques, allows to build [editor's] profiles with good accuracy and stability. This suggests that light monitoring of newcomers may be sufficient to adapt the interaction with them and to increase the retention rate."
Coming as no surprise to those who have been following the remarkably linear decrease in the total number of administrators, a new record was recently set: fewer than 500 active admins.[a]
The large graphic above plots the total number of administrators (+) each month since 2011. That year was chosen as the beginning of the "inactive admin suspension rule", under which the privilege is removed for user accounts that don't make any admin actions for a period of time. The points are fitted to a linear trend line. The trend is a consistently decreasing admin count since 2011. Widefox, who has been maintaining trend graphs at User:Widefox/editors since 2017, shared this analysis: "A linear decline trend in the number of admins is a good fit with r2 = 0.994 . The WP:FRAMBAN loss of admins is a significant drop in the last two years, but similar to fluctuations before 2017."
A second chart to the right shows the active editor fraction: between 40 and 50 percent of all administrator accounts are active at a given time. If you look at the official tally (raw data compiled by bots), the active editor count peaked at just over 1000 in 2008, declined over the next few years to around 550, and has oscillated steadily between 510 and 570 since early 2017. Every tally during 2019 has been in the 43–46% range.
The Signpost followed conversations between various editors who were curious whether the aftermath of WMF's June 2019 office action had an effect on the admin corps headcount, and how that interacted with the overall trend noted above. According to Iridescent, "On Frambanday, 10 June 2019, the figure drops for the first time in recorded history below 510 (we only started counting in 2007), and drops steadily from then on before dropping below 500 for the first time on 28 July."[b]
Prior discussions have not resulted in consensus on how problematic a loss of administrators is; some take the view that bots and other efficiencies can take up the slack. Others like Iridescent express concern: "[T]he issue isn't so much the decline in admin activity per se, but the fact that if admin numbers continue to drop while editing continues, the admin/page ratio increases exponentially and we eventually reach the point where problematic pages need to be locked because there are insufficient admins to perform the routine maintenance." Whichever view is true, English Wikipedia now has 5.896 million articles and 48.247 million pages[c] versus 496 active administrators – or nearly 100,000 pages per active administrator – and the ratio is growing for the foreseeable future. Whether that is a problem, or how a problem would manifest, are questions still to be answered.
For previous Signpost coverage, see my "Wandering in the RfA desert of 2018", published almost a year and a half ago, and other related articles in the sidebar.
Also see
← Last week's report | Next week's report →
It's been a week with many entries driven by women, such as the women's football World Cup (#4, #8) and women politicians running for president of the US (#5, #7, and #18). Otherwise, it's the usual offenders: movies (#1, #2, #6, #10), sports (#9, #13), and TV (#3).
For the week of June 23 to 29, 2019, the 25 most popular articles on Wikipedia, as determined from the WP:5000 report were:
Rank | Article | Class | Views | Image | Notes |
---|---|---|---|---|---|
1 | Kabir Singh | 1,470,837 | In another one of those articles that makes our readers seek what does crore mean, Shahid Kapoor is an alcoholic surgeon who goes on a self-destructive path once his girlfriend (played by Kiara Advani, in the picture with Shahid) is forced to marry someone else. Reviews were mixed, with claims that Kabir Singh glamorises misogyny and toxic masculinity, but it is the third highest grossing Bollywood movie so far this year. | ||
2 | Toy Story 4 | 1,123,417 | Pixar started the 2010s with Toy Story 3, and is closing the decade (which also featured four other sequels and a prequel) by returning to Sheriff Woody and his pals, thankfully in a movie that provides as much of an emotional and entertaining closure as its predecessor. Unsurprisingly, Toy Story 4 got glowing reviews and already amassed half a billion dollars worldwide. | ||
3 | Chernobyl disaster | 1,072,936 | After quite a long hold atop the list, the 1986 disaster dramatized in an HBO miniseries falls but without leaving the podium. | ||
4 | Megan Rapinoe | 1,021,012 | Colin Kaepernick, move over for political protests within real football. The U.S. team that is the big favorite in the World Cup (#8) is captained by someone who is refusing to sing along with the anthem and said she'll not go to the White House if the team wins (which it seems they will, to the chagrin of this writer and everyone who'd like to see someone else with the title rather than a squad which already won 3 out of 7 tournaments), given her fierce opposition to the man currently living there. | ||
5 | Kamala Harris | 977,593 | Speaking of He Who We Prefer Not To Name, given he ran against Hillary Clinton in the last election, the Democractic Party is all full of women who want to follow her footsteps. One of them is this Californian Senator. | ||
6 | Avengers: Endgame | 848,358 | The final confrontation with Thanos never left the top 10 here, but after dropping out from the box office rankings for a week, returned there as the film was reissued with extra scenes (outgrossing Avatar is just around the corner!). Meanwhile, Marvel fans get ready for the first follow-up (#10), coming out this week. | ||
7 | Marianne Williamson | 811,846 | Another woman pushing to be the Democrat candidate, known for her books, plus involvement with peace and AIDS relief causes. | ||
8 | 2019 FIFA Women's World Cup | 773,035 | Everyone who likes this competition having some emotion instead of the certainty that the U.S. team will win certainly mourned that both the host team and the Lionesses (pictured) couldn't make it through. How I wish I could root for the | ||
9 | 2019 Cricket World Cup | 771,533 | The other tournament currently played on European grass fields. | ||
10 | Spider-Man: Far From Home | 769,347 | "Spider-Man! Where are you coming from, Spider-Man? Nobody knows who you are!" (Bark, bark, bark, bark!) |
O beautiful 'momgst factual sites,
For endless waves of FA's,
For series, movies, tragedies,
A catch-all knowledge place,
Wikipedia, Wikipedia,
How trivial your readers have been,
Why are we here, if these should be,
The most read articles of the past week.
For the week of June 30 to July 6, 2019, the 25 most popular articles on Wikipedia, as determined from the WP:5000 report were:
Rank | Article | Class | Views | Image | Notes |
---|---|---|---|---|---|
1 | Spider-Man: Far From Home | 1,908,257 | The man who can do whatever a spider can came swinging back onto cinema screens for the first time in four days, as the web-slinging teenager takes a trip to my side of the pond. The film is highly enjoyable, especially after they drop the villainous illusion, and acts as a worthy bookend, at least in my opinion, to the bombast and spectacle of #8. If you think that this is the last entry on the report from the Spider-Verse, your tingle may be broken. | ||
2 | Stranger Things (season 3) | 1,348,233 | Competing for the attention of our eyeballs with #1 is the return of Netflix's (arguably biggest) original series, which sees the return of nosebleed Jean Grey, Toothless the (Dungeon and) Dragon, and beep beep Richie, as they battle both monstrous entities from the Upside Down and their hormones. If you were concerned that the new season would abandon the near-intoxicating nostalgia of its predecessors, the fact that much of the season involves the characters going to the mall should alleviate your fears. | ||
3 | Cori Gauff | 1,200,886 | Over on the grass courts of the All England Club, Wimbledon is underway, meaning that the world gets to pretend it cares about tennis for a fortnight. A story of actual intrigue, alongside the Williams-Murray partnership and the continued resurgence of Rafa and Roger, was that of Cori Gauff, the youngest player in the history of the tournament at only 15 years of age, who drew attention for her nonchalance as she felled Venus Williams in the first round, going on to reach the final 16. | ||
4 | Stranger Things | 1,066,290 | As an author of the Top 25 Report, I am often very bored at having to come up with novel commentary for the ceaseless plethora of Netflix Originals, many of which (I'm (not) looking at you, Sandy) are not very good, to be diplomatic about affairs. However, I am a big fan of most of Stranger Things, with its superb theme, likable cast, and a highly bingeable story evocative of the work of two Stephens. Gauging from what I have seen of the new season (#2), released on America's birthday, the Duffer brothers have kept up the good work. | ||
5 | 2019 Cricket World Cup | 1,038,856 | It's coming home, it's coming home, it's coming... Yes, it would appear that neither Finch nor Kohli (to the doubtless despair of much of our readership) will be able to prevent Ireland's greatest cricketer from lifting the World Cup on home(?) soil, with only New Zealand left standing in his way. | ||
6 | Midsommar (film) | 1,019,055 | The newest horror film from Ari Aster, director of the critically acclaimed and audience maligned Hereditary was released during the week, courting the lucrative audience segment of arachnophobes who are OK with brutalistic pagan rituals. It is an admittedly niche market. As someone who vehemently dislikes the jump-scare riddled tedium of contemporary horror, I appreciate Aster's efforts, but as someone who found much of Hereditary to be more hilariously absurdist than abjectly horrifying, I think I will give this one a miss. | ||
7 | Tyler Skaggs | 975,817 | A sadder entry now, as an Angel ascended at the age of 29. Skaggs, a pitcher for the non-dodging LA outfit, was preparing to face some walkers when he was found unresponsive in his hotel, leading to an outpouring of grief and morbid interest in the player from fans of the national pastime. | ||
8 | Avengers: Endgame | 875,684 | You knew this would be here, with the re-release hyping up and subsequently underwhelming MCU fans worldwide, and with the gap between the top two entries on the list of highest-grossing films drawing tantalisingly closer - this film, this box office juggernaut so large that the industry is still reeling, just has to be on the report. It is inevitable. | ||
9 | Kabir Singh | 795,093 | With their enjoyment and anticipation for #5 tarnished by kiwis, our Indian contingent went back to frantically researching the newest Bollywood hit. This time, it is Sandeep Vanga's remake of his own Telugu film that has caught their attention. The massive hit is produced by T-Series, and has been critically maligned despite its status as number one. Sound familiar? | ||
10 | Alex Morgan | 762,853 | Many historians cite the spilling of tea as a moment where Anglo-American relations seriously deteriorated. In Parc Olympique Lyonnais last week, star-spangled soccer starlet, having slotted the ball beyond the English keeper, popped out her pinky and triggered another breakdown between the two allies, leading to some stunningly bad takes (including from another Morgan). I'll go to the hilt for the World Cup winner, because our freedom to mock Britain must be preserved, now more than ever. |
More than a year and a half later, again it's time for a Report heavy on Stranger Things, comprising seven entries. The only other subject just as present are sports, all of them on grass in Western Europe: football in France (#3, #8) and tennis (#7) in England. And yet the list is topped by one shocking death, of an actor who couldn't even make it to his 21st birthday and a perverted criminal. Thankfully, Spider-Man is here (#6) to help this intro end on a positive note.
For the week of July 7 to 14, 2019, the 25 most popular articles on Wikipedia, as determined from the WP:5000 report were:
Rank | Article | Class | Views | Image | About |
---|---|---|---|---|---|
1 | Cameron Boyce | 6,641,458 | Few things bring in impressive view counts like a death. And even if actor Cameron Boyce was hardly known for those unfamiliar with Disney Channel productions, the sheer fact that he left us at just 20 due to an epileptic seizure certainly intrigued\shocked enough people to bring in millions of visits to his page. | ||
2 | Jeffrey Epstein | 2,793,733 | A convicted sex offender. We're starting off in some dark notes! Convicted for procuring an underaged girl for prostitution back in 2006, this financier is back in the clutches of the law, accused of sex trafficking of minors in Florida and New York. | ||
3 | Megan Rapinoe | 1,960,356 | She's a lesbian, she's outspoken against Trump, and she steamrolled in the World Cup as captain of the U.S. team who won it all, winning the Golden Boot (top scorer) and Ball (best player) along the way. | ||
4 | Stranger Things (season 3) | 1,615,595 | After taking last year off, the "1980s Steven Spielberg adventure meets 1980s Stephen King horror" returned to Netflix, now with Soviet Russia making a small Indiana town become an interdimensional hell again. | ||
5 | Stranger Things | 1,412,749 | |||
6 | Spider-Man: Far From Home | 1,381,177 | The most popular and suffering Marvel Comics superhero is back on the big screen, and his bad luck ensures his European road trip is crashed by monsters who are seemingly connected to some Mysterio. Entertaining and with some cool surprises (even if like its predecessor, it has way too much comic relief), Far From Home is certainly giving what audiences wanted after Avengers: Endgame, and is nearing $900 million after just two weeks in theaters. | ||
7 | Roger Federer | 1,213,282 | At the age of 37, the biggest winner of tennis Grand Slams still has it even in defeat, only losing the Wimbledon final to Novak Djokovic after more than four hours of play. | ||
8 | Alex Morgan | 1,114,345 | Megan Rapinoe's (#3) teammate with the most luck in striking some goals, having scored the same 6 bouts of the captain on the way to winning her second World Cup title. | ||
9 | Freddie Mercury | 1,075,248 | Well, look who is back here. Along with the HBO premiere of Bohemian Rhapsody, the other reason for the iconic Queen frontman to get a spike in views is actor John Glover claiming he had sex with Mercury back in the 1980s. | ||
10 | Millie Bobby Brown | 1,063,970 | She entered the top ten, but the most adequate rank for this British actress (the breakout star of #5's child\teen cast, who earlier this year lent a hand to Godzilla) would be Eleven. |
What lies beneath Earth always intrigued man, even if actual people who walked outside this planet were only those involved with exorbitant expenditures to put a man on the Moon (#5, #7). The prospect of life in other worlds moves our top entry, regarding plans to walk into an air base that supposedly housed alien corpses, and is present with the extraterrestrials in Marvel movies (#8, #10) and the beasts from the Upside Down in Stranger Things (#6). But Earth and humans is all we have so far, with all the sports (#2, #3) and movies (#4) possible available to make us forget of bad things such as people dying due to an arson attack (#9).
For the week of July 14 to 20, 2019, the 25 most popular articles on Wikipedia, as determined from the WP:5000 report were:
Rank | Article | Class | Views | Image | About |
---|---|---|---|---|---|
1 | Area 51 | 2,826,689 | On June 27, a funny fellow named Matty Roberts posted a gag post on Facebook: "Storm Area 51, They Can't Stop All of Us." He proposed that a flashmob of thousands of alien-hunters simultanously storm the walls of Area 51, the infamous Air Force base in Nevada that supposedly housed extraterrestrials, on 20 September. He meant it as a joke, but three million people have signaled their intent to take him up on it. The government already warned that whoever trespasses into armed forces territory is subject to arrest and prosecution. | ||
2 | Novak Djokovic | 1,832,087 | Wimbledon's final was between two tennis legends, whose confrontations warrant their own article. After nearly five hours of play, with Federer nearly winning twice, his Serbian adversary won, marking his fifth title in the British grass. | ||
3 | Roger Federer | 1,707,049 | |||
4 | The Lion King (2019 film) | 1,301,407 | Out of Disney's tendency to remake its animated classics, The Lion King made the least sense: with no human characters, it effectively would just trade the 2D animation for the latest computer graphics. And with barely any changes to the plot, only adding short amounts of new content for the sake of padding the runtime to two hours, it is certainly the same movie again, only less inspired (the musical numbers suffered the most - what the hell did this movie do to "Be Prepared"?!), unlike what director Jon Favreau had previously done in The Jungle Book. Not that audiences cared, given this new Lion King made half a billion dollars in just a weekend. | ||
5 | Apollo 11 | 1,075,876 | "If you believed they put a man on the Moon, man on the Moon..." Well, some stupid people claim otherwise, but on July 20, 1969, Apollo 11 managed to land on Earth's satellite and fly back. 50 years later, this milestone of spaceflight is celebrated as returns to the Moon are planned, along with going to Mars. | ||
6 | Stranger Things | 881,105 | In-between two entries about humans leaving Earth in the 1960s, interdimensional monsters entering Earth in the 1980s. | ||
7 | Neil Armstrong | 857,633 | "One small step for man, one giant leap for mankind", said Armstrong as he left the Apollo 11's (#5) Lunar Module - "The Eagle has landed"- and became the first man on the Moon. Armstrong, who died in 2012, was described as a "reluctant American hero", given he was a fairly reserved person, and continued to keep a low profile after his return to Earth. | ||
8 | Spider-Man: Far From Home | 788,599 | Peter Parker's ruined European vacation has gotten the approval of both reviewers and audiences, and managed to beat Spider-Man 3 as the highest-grossing solo movie of the Webhead. | ||
9 | Kyoto Animation | 756,402 | An arsonist set ablaze the studio responsible for animes such as The Melancholy of Haruhi Suzumiya, killing at least 34 people, injuring about the same amount of people, and destroying computers and materials. | ||
10 | Avengers: Endgame | 753,872 | In February, the Academy somehow passed over the Visual Effects of Avengers: Infinity War to award the story of our #7. Follow-up Endgame has since become the highest-grossing movie ever, so the Oscar voters certainly have a chance at fixing this snub next year. |
The number of active editors on the German-language Wikipedia (de:WP) has long been declining, dropping almost two-thirds since its peak in 2007. The French-language Wikipedia (fr:WP) has almost caught up in the number of articles, and by some measures, such as total monthly editors, is overtaking de:WP. This article examines this phenomenon using data from Wikistats 2, which was introduced this year, data from the earlier Wikistats 1, and other data collected by Erik Zachte over the years. Wikistats 2 includes some wonderful graphics, but, because I've combined that data with other data, my own graphs are presented.
The English-language Wikipedia started on January 15, 2001. Less than three months later both de:WP and fr:WP versions began. Both German and French chapters, Wikimedia Deutschland (WMDE) and Wikimedia France, were founded in 2004.
From this similar start fr:WP developed like a "little sister" to de:WP's "big brother." Its growth lagged de:WP and it never appeared to be completely catching up, until now. With impressive growth in the early years, de:WP had a stormy start while fr:WP had more leisurely growth. This difference may be due to the population of native speakers in each language. In 2017, there were 105 million native German speakers against only 79 million French native speakers. There were 33% more German speakers. The number of German native speakers ranked 10th among all languages, French ranked 11th according to the Duden publishing house. By the way, we also find a similar percentage difference in Wikipedia page views from Wikistats 2 in June: 643 million views to 869 million views (a 35% difference).
At the end of June 2019, fr:WP with 2.1 million articles was still behind de:WP with 2.3 million articles. However, the average monthly increase in articles since December 2017 in fr:WP has been higher than the increase in de:WP, which may be explained by a greater use of bots at fr:WP. For the end of June, the following averages were available: 10,086 new articles for fr:WP, and 9,411 new articles for de:WP. Both have now reached about the same size and growth, which is likely due to the comparable number of active editors. With the monthly number of content edits in de:WP at around 500 thousand per month, fr:WP has almost caught up. In June 2019 it was 474 thousand to 442 thousand edits; for the past 12-month period the mean values are 512 thousand to 460 thousand edits.
Both projects currently have a total of about 55,000 editors monthly (mean values: 55.9 thousand for fr:WP; 55.5 thousand editors for de:WP). These include not only registered users, but also IP's and bots, each with at least one edit. Active editors, those with at least five edits per month, show the same trend. The number is around 8,650 active editors with a small lead for fr:WP. In June 2019 there were 55.9 thousand to 55.5 thousand (all editors) and 8,650 to 8,010 (at least 5 edits). For median values over the past year, the following relation applies to the active editors: 8,793 (fr:WP) to 8,483 (de:WP). The stable development of the active editing participants in fr:WP is countered by a long-lasting decline in editors in de:WP. For the past three years, from July 2016 to June 2019, the mean value in the fr:WP increased slightly by 1.7% (from 8,641 editors to 8,797); however, it steadily declined by 16% in the de:WP (from 10,130 editors to 8483).
Data on the registrations of new user accounts show a positive picture, but fr:WP with about 15,300 new accounts monthly, is well above the number of new users in de:WP with 9,700 new accounts monthly. However, caution is advised, since very few of the new registrants edit and, again, very few remain active. The order of magnitude is 1-5%, depending on the degree of activity and the time interval examined.
Until December 2018, Wikistats 1 recorded the number of newcomers who had reached their tenth edit. These numbers allow the following comparison: There were 704 newcomers in fr:WP, but only 453 in de:WP. For the median 12-month values (2018): 758 newcomers to 589, ie 29% more in fr:WP. Of the approximately 15,000 newcomers in fr:WP and 9,700 in de:WP, around 5% each reach at least 10 edits. The more than 50% higher level of account registrations in fr:WP shows a level of recruiting, which – unlike in the de:WP – obviously compensates for or even exceeds the number of departures to editors.
Considerable differences are also evident in the full-time staff of the Chapters. With only eight permanent employees in 2019 [1] Wikimedia France is well behind the chapters of the three D-A-CH countries (Germany, Austria and Switzerland) together. Above all, WMDE could perhaps look to Wikimedia France to see how leaner structures might produce better results in the interests of reinforcing editor communities.