There will be no Tech News issue next week. The next issue of Tech News will be sent out on 2 September 2019.
Some abuse filters stopped working because of a code change. Only variables for the current action will work. Variables defined inside a branch may not work outside of that branch. You can read more to see how to fix the filters.
Only six accounts can be created from one IP address per day. Between 12 August and August 15 this was two accounts per day. This was because of a security issue. It is now six accounts per day again. 
Changes later this week
Only a limited number of accounts can be created from one IP address. An IP address can be whitelisted so that it can create as many accounts as needed. This is useful at events where many new persons learn to edit. IP addresses that are whitelisted for this reason will also not show CAPTCHAs when you create accounts. This will happen on Wednesday. 
The new version of MediaWiki will be on test wikis and MediaWiki.org from 20 August. It will be on non-Wikipedia wikis and some Wikipedias from 21 August. It will be on all wikis from 22 August (calendar).
You can join the technical advice meeting on IRC. During the meeting, volunteer developers can ask for advice. The meeting will be on 21 August at 15:00 (UTC). See how to join.
There is an RFC about creating a new global user group with the right to edit abuse filters. This will be used to fix broken filters and make sure all filters will still work when software changes happen. You can read more and comment.
Special:Contributions/newbies will no longer be working. This is because of performance reasons. It showed edits by new accounts. You can see this in the recent changes feed instead. 
Wikimedia Commons holds over fifty million freely-licensed media files. These millions of images, sounds, video, documents, three-dimensional files and more contain a vast amount of information related to the contents of the file and the the context for the world around them. As Commons has collected files over the years, the volunteers who curate and maintain the site have developed a system to contain and present this information to the world, using MediaWiki, wikitext, and templates.
A description template is the first and primary way information about a file is show to users. These templates can be a powerful tool for displaying information about files; descriptions provide meaningful context and information about the work presented. Descriptions can be as long as the user would like, providing wikitext markup and links for others to find out more. Description templates can also hold translations by adding language fields. However, the Structured Data team saw some areas that a feature like captions could improve upon from descriptions templates.
Multilingual captions help share the burden of descriptions by providing a space to describe a file in a way that is standard across all files, easy to translate, and easy to use. Captions do not support wikitext so there is no knowledge needed of how to links work in this space — links can still be provided in the more expansive file description. Captions are added during the upload process using the UploadWizard, or they can be added directly on any file page on Commons. The translation feature for captions is a simple interface that requires only a few steps to create and share a caption translation.
The “multilingual” in “multilingual captions” highlights a primary focus of Structured Data features: opening up access to Commons to as many languages as possible beyond its present capabilities. This is enormously beneficial to the Wikimedia movement and Wikimedia Foundations’ mission of sharing knowledge with the world. In addition to captions, future features planned provide supporting adding “statements” from Wikidata to files, effectively describing them in an organized way that can be accessed by programs and bots to present media. These statements can be multilingual as Wikidata supports translations, which will make statements searchable in any language that has a translation provided.
Content Translation achieved a new milestone, supporting already the creation of 500,000 Wikipedia articles. The Language team has been working during the last year to make the tool more solid, and has plans to expand the use of translation to help more communities to grow.
Wikipedia users can learn about many topics. However, the exact number of topics they can access is very different depending on the language they speak. While English speaking users can access more than 5 million articles, Bengali speakers have access to 75 thousand articles.
Translating articles into new languages is a practice that can help content to propagate more fluently across languages, and reduce this language gap. To facilitate this process, we here at the Wikimedia Foundation developed a content translation tool that helps Wikipedia editors to easily translate articles. Content Translation simplifies translating Wikipedia articles into different languages by automating many of the boring steps of the manual translation process.
In early August, Content Translation reached a new milestone: more than half a million articles were created since the tool was released four years ago, making this a good time to reflect on the impact of the tool and discuss future plans.
A more reliable tool
During the past year, the Language team worked on a new version of the tool. Based on user research and feedback, the plan was to create a more solid version of Content Translation to increase the tool adoption and use.
For the new version we replaced the default editing surface provided by the browser with Visual Editor, which supports rich wiki content in a way that is much more reliable. This required a rewrite most of the translation tools, and we wanted to take this opportunity to review them and provide better guidance for newcomers.
As the new version became more complete it was gradually exposed more prominently during the year, and finally replaced the previous version completely without major regressions. During the year more than 149.000 translations were created, a 23% increase compared to the previous year.
We started conversations with different communities to identify the main blockers before the tool could be provided by default and exposed to more users.
Better collaboration between humans and machines
In addition to the number of articles created, we focused on the quality of the content. The new version improved the guidance provided to newcomers. In particular, a new system was created to encourage users to review and edit the initial machine translation, and approaches based on Artificial Intelligence were explored to improve some automatic steps.
Content Translation provides machine translation as initial content for editors to review and improve. The machine translation is provided as a starting point, and translators are highly encouraged to rewrite the content, in order to eliminate errors and make the translation sound more natural.
The new version incorporates new quality control mechanisms for machine translation. Now the tool encourages translators to review the initial automatic translations on a paragraph basis, keeps in a tracking category those translations published with unmodified content for editors to review, and prevents publishing those which exceed the limits defined. The limits to prevent publishing become more strict for users with previous deleted translations, users ignoring the warnings, and cases where several paragraphs contain unmodified contents. In this way, the limits adapt to reduce potential recurrent misuse of the tool.
In general, our measurements suggest that translations are less likely to be deleted than the articles started from scratch. The survival rate for translations even when those are created by newcomers seems quite good. A recent study shows that a significant percentage of the translations created with the tool survive the community review. Although the survival rate is better for experienced users, it is still very good for newcomers (users that created their account during the last 6 months). For example, only 7.5% of translations created by newcomers in last june were deleted after a month.
In addition, Artificial Intelligence is becoming more present in the tool to make the initial translations better:
We believe that automation with adequate quality control mechanisms makes it easy for translators to create higher quality translations more easily.
Translation has helped already many communities to create new content. However, there are still communities with potential to grow by using translation that have not been using the tool as much.
Content Translation’s Boost initiative is aimed at expanding the use of translation to help more communities grow. By enabling new and more visible ways to contribute by using translation, we expect communities to attract new editors, and expand the knowledge available in their languages.
We identified potential for expanding its use to more contexts that can benefit from translation:
Translation can be used by more wikis. The adoption of Content Translation varies significantly from wiki to wiki, and there are wikis with potential to benefit from using translation more.
Translation can be used in more ways. Currently, Content Translation focuses on creating new articles on desktop. Supporting new kinds of contribution such as expanding existing articles with new sections, or mobile translation enable more opportunities to contribute.
During the next months we will focus on wikis with potential to grow by translation. As a representative set of those wikis we have initially selected Malayalam, Bengali, Tagalog, Javanese, and Mongolian. We’ll be contacting these communities to gauge the interest in the project, and learn about their particular needs to support them better. We expect these and similar communities to benefit as a result. Our specific plans will be heavily influenced by research in the selected communities and their feedback. Please, provide any feedback about this initiative in the discussion page. We are interested in hearing your ideas on how to help communities grow by using translation.
Editors using the mobile website on Wikipedia can opt-in to new advanced features via your settings page. This will give access to more interface links, special pages, and tools. Feedback on the discussion page is appreciated. 
Due to the absence of volunteer maintenance of Cologne Blue skin, the link to activate it will be hidden. The skin will still work, but editors using it are encouraged to switch to another skin. 
Changes later this week
Due to Wikimania, there is no deployment this week. 
You can join the technical advice meeting on IRC. During the meeting, volunteer developers can ask for advice. The meeting will be on 13 August at 15:00 (UTC). See how to join.
The “Wikidata item” link will be moved from “Tools” to “In other projects” section on all Wikimedia projects, starting on August 21st. Full announcement, Phabricator task.
The Punjabi-language Wikisource is the fastest-growing Wikimedia project in the world. Rupika Sharma, a volunteer Wikimedia editor and community member, writes about one of the initiatives that has helped made this a reality.
Imagine a world where you grew up in a world where the greatest literary works in history never existed.
For many of the world’s language speakers, this can be their functional reality. Titles like these have either never been translated, or were translated decades ago and now hide in ancient paper-bound texts on dusty library shelves.
As an example of this problem, let’s take a look at the Punjabi language. Separated as part of the 1947 partition of British India, the language is today spoken by 120 million people in regions of Pakistan and India. I’m one of them. I grew up in northwest India and can still remember hearing about Chambe Diyan Kaliyan, a short story collection by Leo Tolstoy that was adapted into the Punjabi by Abhai Singh. That particular book is frequently cited in the history of Punjabi literature as one of the first collections of short stories to be published in the language.
You’ll note, though, that I didn’t say I can remember reading it—I’ve never been able to track down one of the published books to read it for myself, nor have I been able to find anything but a bunch of pop-culture songs with similar titles when I search for it online in Punjabi. All of which is to say that when I was growing up, reading and learning from Tolstoy’s story was functionally impossible for Punjabi speakers.
Thankfully, times are changing. While there are still many barriers to surmount, the advent of the internet has made the fundamental problem of publishing and distributing of translations far easier. The Wikimedia community has an entire project devoted to this sort of thing: Wikisource.
Bringing the lost literature of long-forgotten times into the modern era for interested users, Wikisource is a free e-library that provides freely licensed or public domain books free of cost, in different formats, and able to be used for any purpose. It is one of thirteen collaborative knowledge projects operated by the Wikimedia Foundation, the largest of which is Wikipedia, and Wikisource is available in nearly seventy languages.
The Punjabi-language Wikisource was and is small compared to other language Wikisources, and to grow this resource, I formed a partnership with a government library in the Indian city of Patiala to digitize public domain books. By making rare literature books accessible in languages that have little to no presence online, Wikisource serves the common people, allowing them to freely browse these resources.
As a titled Wikimedian-in-residence at the library, I helped their staff scan a selection of important books. The collaboration brought forty-two public domain Punjabi-language works online—including a reprint of Chambe Diyan Kaliyan, the Tolstoy short story collection. But just making the scanned images available online isn’t enough; they are not easy to read and often rank low in search engines. Wikisource plays a crucial middleman role: they host the images and pair them with searchable text versions, created and vetted by volunteers. They’re helped in this process by Jay Prakash’s IndicOCR, a new tool that helps to easily transcribe any Indic language to Wikisource. (It replaced an older Linux-based tool that could not be used on many devices.) In addition, Wikisource makes everything available in different file formats so that readers can download whatever works best on their device, whether it’s a computer, tablet, phone, or otherwise.
Finally, Wikisource also allows anyone to contribute, and so I helped organize an online contest, held from December 2018 to January 2019. Prize offerings and in-person trainings brought around three dozen new volunteers to the project, including twenty-four who made more than fifty edits. Kuljit Singh Khuddi, a new volunteer who joined Punjabi Wikisource during the contest, says that “I am proud to be able to contribute to my mother tongue on Wikisource. Such contests help make my language known worldwide.”
The results were stark—the contest made the Punjabi Wikisource the fastest-growing Wikimedia project in the entire world in both content and editors. As of October of last year, the Punjabi Wikisource contained a bit over 1,200 pages. By January of this year, it had over 6,770 belonging to 200 different books. Moreover, over 6,000 of these pages had been proofread by volunteers.
The growth of the Punjabi Wikisource through the contest and other volunteer work is just a beginning. There are a number of opportunities for supporting the project with technical contributions and GLAM partnerships with different government organizations and institutions.
Moreover, they’re just one of several expanding Wikisources in the region. The Wikisources for the Indic languages of Marathi, Kannada, and Assamese each more than doubled in size in the last year, and with every edit, they’re bringing the sum of all knowledge into their own mother tongues.
The first time Helena* — a scientist and published author — edited Wikipedia, her edit was immediately reverted: “It was not only reverted,” she recalled, “it was reverted with a ‘You don’t know your ass from a hole in the ground’ kind of a note.”
But she persisted. She created a new account, read Wikipedia’s policies, and continued to contribute. When another user blanked Helena’s Wikipedia user page — deleting content she had written and leaving her a death threat — administrators refused to act. At this point, Helena decided it was time for a break: “I took a hiatus. I told everybody to just basically go shove it. ‘I may never come back to Wikipedia’ is what the message says; ‘I’ll think about it.’”
Today, Helena has been editing Wikipedia for more than 14 years. When asked to reflect on her earliest experiences, she replied, “There’s a wonderful phrase. I culled it out the other day. I put it in a little file folder to share with you. ‘We throw brand new potential editors directly into shark-infested waters, then yell at them for splashing at the sharks.’”
Note: Our paper is a qualitative study and, thus, is not intended to be generalizable. Also, some Wikipedia editors have criticized our study for containing “outdated” data, but qualitative work is time-consuming, and academic research requires a lengthy peer review process that does not allow for immediate publication.
Wikipedia is written primarily by men. This unequal division of labor in social systems is known as the gender gap.
Women who edit Wikipedia have different perspectives about the gender gap, but many have witnessed or experienced harassment. 
Experienced women Wikipedians have developed sophisticated tactics to participate in the community even when they feel unsafe.
Based on our conversations with experienced women Wikipedians, we share three provocations for designing safer spaces: (1) when aiming for inclusivity, consider safety a design requirement; (2) recognize your own assumptions and biases about safety; (3) provide users with tools for creating their own safe spaces.
We interviewed 25 women who are experienced Wikipedia editors.
There’s this one guy who is part of the chapter here that was, for a while, posting date invitations on my talk page, that would say how much he wanted to spend time with me. Then it became a thing at edit‑a‑thons that he would attend too, where I felt like he was harassing me.
Mia, editing for 5 years
Wikipedia is more than a website. It’s a community stretching across the globe and made up of a multiplicity of online and offline spaces, many of which are porous. Each of these spaces — article talk pages, internet relay chat (IRC), edit-a-thons, meetups, conferences — has its own character, which is shaped by design affordances as well as community norms and values. Interactions in these different parts of Wikipedia, as Mia notes, often bleed across these sociotechnical boundaries — often without consent or control.
I’m not an administrator and don’t want to be. Editors who work in topics which are perennially under assault […] often burn out for periods, sometimes permanently. I have tremendous admiration for those who have taken on the Sisyphean task […] I couldn’t do that work.
Oona, editing for 12 years
Women Wikipedians have developed sophisticated tactics to sustain their participation as community members even when they feel unsafe among their peers. For women like Oona, choosing what work to avoid is one way to protect themselves. For other women, choosing what to edit (for example, avoiding controversial topics) is another way to manage their participation so that they can “avoid drama” and its associated harassment risks.
We don’t feel safe on Wiki. Not all of us, but a lot of us don’t, so why keep doing this on Wiki when we can take it another place […] ?
Jenn, editing for 12 years
Because Wikipedia does not allow users to create exclusive online spaces (e.g., women-only), women like Jenn have created or joined private groups on Facebook to cultivate and promote safe peer engagement. These spaces allow participants to share their personal experiences as well as their ideas about editing and other ways of participating bravely in the Wikipedia community. We take from these examples the need for designers to consider safety as a design requirement when aiming for inclusivity. This approach, in turn, means recognizing one’s own assumptions and biases when imagining the design of idealized scenarios, but also sometimes translates into the realization that providing users with tools to create their own safe spaces is the best means to ensure the safety of all.
* We’ve used pseudonyms to protect our participants’ identities.
This article summarizes a paper authored by Amanda Menking, Ingrid Erickson, and Wanda Pratt. This paper was presented at CHI 2019, a conference of Human-Computer Interaction.This post was originally published via Medium.
Since returning from my first Wikimedia event, the ESEAP Region Strategy Summit in Bangkok a few weeks ago, I have been asked many times by family, friends and workmates “so, what was it like?”. And every time I (figuratively) scratch my head, wondering how to describe the unique experience of two days spent with 30 Wikimedians representing a remarkable 14 countries – that is, one or two Wikimedians from each of the ESEAP member communities (ESEAP stands for East, Southeast Asia and the Pacific).
I could tell people about the goal of the weekend: for participants to exchange stories from their Wikimedia communities and provide the movement strategy Working Groups with a response on movement strategy discussions. I could also describe the activities we engaged in: sharing stories and mapping them to thematic areas such as capacity building or roles and responsibilities on Saturday, and generating recommendations for working groups on Sunday. These discussions led us to explore a wide range of innovative ideas related to all aspects of the 2030 strategy.
At times I must admit I began a session thinking “I have no idea what I could contribute here” but after listening to others and through the facilitators’ encouragement, I was generally able to find a meaningful experience to share or a suggestion to make. I was also impressed with the way people from such diverse backgrounds were able to respond to each other’s problems and experiences with meaningful suggestions. It wasn’t unusual, for example, for someone from one country to describe a need in their editing community which a Wikimedian from somewhere else had already experienced and could provide a possible solution to. For example, while discussing ways of recognising the efforts of Wikipedia editors, participants from Indonesia shared their method of providing a letter of acknowledgement from their chapter President detailing the impact of an editor’s work.
But these answers are not the whole picture. In addition to the organized activities and planned outcomes, there was much more to the event. There were new friendships, for example – I found I had much in common with the women from Taiwan, who have created a strong network of editors writing about women (their meet-ups are called “A Room of WikiWomen’s Own”), and also with the members from Australia, who wonder how to engage offline with editors in rural and remote parts of the country.
In addition, there was a generous sharing of expertise for people working on development projects. This was a particularly exciting aspect of the event for me, as in New Zealand the editing community is small, and largely unconnected, which means that there are great opportunities for community growth. I was able to spend time with Wikimedians from communities such as Indonesia and Thailand which had already been through similar growth patterns, and hear their thoughts and suggestions, and hear their responses to New Zealand’s plans. I also learned a great deal about the workings of these more developed communities and I was able to reflect on whether our New Zealand community would follow a similar development path or something slightly different.
All in all, it was a very worthwhile event which provided much inspiration and insight into the movement as a whole. I definitely feel much more aware of the “big picture” of Wikimedia’s goals and strategies and more connected to communities in the region. This event was so inspiring, that since my return home from the event, the New Zealand editing community has now finalized our plans for a User Group and made an application to the Affiliations Committee for recognition. We look forward to becoming a strong group making solid contributions to the 2030 strategy.
And in terms of answering that omnipresent question of what a Wikimedia event is like, I’ve now settled on a concise New Zealand response – “awesome”!
Tech News is a newsletter for reaching out with technical updates to the general Wikimedia editor communities, to make sure they can keep track of what’s happening.
It is a popular publication, typically distributed in 15–20 languages due to the fantastic help of volunteer translators. It reaches roughly 100 community pages (Village Pumps, etc.) and some 780 individual subscribers (add yourself!) weekly, in addition to those who read it on Meta, see it included in the Signpost or other community publications, or get it in their email inbox.
But do you know how is Tech News written?
You have built a new universal tool for the Wikimedia community? Or you plan to push a change into production that will change things for everyone? Please add it to the next issue of Tech News! The newsletter relies on additions or suggestions made by the technical community.
What to add to Tech News?
We welcome any kind of information that has a technical impact on the wikis. Typical items are new or upcoming features, or potential breaking changes. However, we have several cases where Tech News is not the right place. We have listed several inclusion criteria. If you have any doubt, please ask us!
But in short, any change that has a significant impact on the wikis should be added on Tech News (and the earlier, the better).
How to add something?
There are several ways to make the communities aware of a technical change.
The most common way is: please add your update yourself! Tech/News/Next will take you to the relevant issue. Remember to link to a relevant Phabricator task, wiki page or email.
To add your item to the next issue, the recipe for a good addition is very simple:
a couple of sentence per item, written in a simple English — making it too long puts an unreasonable burden on the translators.
a link to a Phabricator task, a wiki page or an email if they need more information.
And that’s all! Don’t worry about polish, we’ll take care of that. But keep an eye on it: we may need some more information from you!
Tech News is for non-technical readers, and for an international audience. It should be easy to translate as well as be written for people who’s native language is not English (en-1 and en-2 readers).
Concerning other options, there’s a “#user-notice” tag in Phabricator. Add it to the task together with a simple 1–3 sentence explanation of what this is and how it affects editors. And you can always write a message on Tech News talk page.
To keep things focused, keep in mind that Tech News is not a general Wikimedia newsletter, or a way to reach the Wikimedia technical community.
Rule of thumb: the earlier you keep Tech News editors aware of a possible changes, the better! This way, communities will have the information in advance.
Translations and distribution
When we have all the news, we assemble the newsletter. Translations are really important. This is why we have a tight calendar concerning inclusions:
Most important information for the week is gathered on a first draft on Thursday. This draft is sent to translators for an early translation round.
On Friday, the information is checked again. Some last minute items are added, no-relevant anymore ones are removed. Sometimes things are updated. Then, the newsletter is frozen for good: no more additions are allowed.
Over the weekend, volunteer translators work on translations.
The distribution is done on Monday afternoon/evening UTC.
As we rely on collaboration to create the newsletter weekly, we also rely on your opinion to improve Tech News.
Are there ways we could make Tech News better at spreading information about technical updates that are relevant for Wikimedia contributors? Something we do that’s unnecessary? Things we’re often missing? Things we fail to explain? Anything we’re particularly good at and should keep doing? Tell us – on Tech News’ talk page, privately, or, of course in the comments below.
The new version of MediaWiki will be on test wikis and MediaWiki.org from 6 August. It will be on non-Wikipedia wikis and some Wikipedias from 7 August. It will be on all wikis from 8 August (calendar).
You can join the technical advice meeting on IRC. During the meeting, volunteer developers can ask for advice. The meeting will be on 7 August at 15:00 (UTC). See how to join.
Today everyone can see IP addresses if someone edits without an account. In the future this could be more hidden. This is to protect unregistered editors so fewer can see their IP address. This would only happen after we make sure the tools for vandal fighting can still be effective. You can read more and comment.
Have you ever tried zooming in on an image on your computer? Did you notice how the quality steadily got worse and worse?
This is an unfortunate feature of “raster graphics,” the file formats that use a rectangular grid of pixels to make up an image, and present a particular problem for people making maps, for example—how else would you be able to take a regional map and zoom down into a particular town?
The solution is vector graphics. These allow for arbitrary scaling at high quality and give sharp high-resolution renderings. SVG, the vector file format used most often on Wikimedia projects, allows other editors to edit the files, something especially suitable for Wikimedia’s long-held ethos of allowing anyone to edit. But what SVG does not do well is allow anyone else to easily edit the files, as it requires specialized knowledge. As recently as last year, less than a handful of overtaxed volunteer Wikimedia editors working on Indic-language Wikimedia projects were equipped with the skills required to edit vector graphics, and there was little awareness of them among other Wikimedians. Things like map legends could be easily translated into hundreds of languages if only more people knew how to do it.
• • •
A training program set out to change that. In September 2018, Indian Wikimedians hosted the Wikigraphists Bootcamp. Named for the specialized community of editors who make maps and other vector graphics on Wikimedia Commons, the bootcamp included three online sessions during August and September 2018, and one three-day onsite workshop during the last week of September. Twenty-two Wikimedians representing nine Indic language Wikimedia communities, along with Indian Wikimedians working on the English Wikipedia and Wikimedia Commons, took part in the training.
Though this whole project was able to meet most of its goals and develop a community of Wikigraphists, at least with basic skills, a follow-up was extremely necessary—not only to help ensure the trained Wikimedians continued their efforts in the area, but also to spread the concept of vector graphics among the larger Indian community.
The answer to this came with “SVG Translation Campaign 2019 in India,” a thirty-eight-day-long campaign that started on 21 February 2019, chosen to align with UNESCO’s International Mother Language Day. Funded by CIS-A2K, the aim was to create SVG diagrams and maps with labels in the native languages of India. Nineteen Indic languages participated in the campaign, which was organized by four coordinators and twenty-six language organizers. Several online training sessions were conducted for language-organizers before the campaign began, and who in turn conducted brief sessions—some on-site—in their communities. This approach worked out well, and as a result, 194 users created more than 2,500 SVG files with labels in native Indian languages. The campaign was quite successful in creating mass awareness among Indian communities about vector graphics and their importance in Wikimedia projects.
The Wikimedians are currently working on setting up a dedicated Graphics Lab for Indic communities, where Wikimedians would be easily able to make requests for required graphical content on Indic-language Wikimedia projects. Several are now enthusiastic about learning more about advanced skills related to vector images, and further training needs are being explored.