From MediaWiki.org
Jump to: navigation, search
For help on editing pages, see Help:Editing

The Editing department of the Wikimedia Foundation's Product group works on extending and improving the editing tools used at Wikimedia – VisualEditor and other content tools, Collaboration tools, Multimedia support, Language tools for translation, content and communication, and Parsing and other infrastructure. We exist to support new and existing editors alike across desktop, tablet and phone platforms.

Progress on annual plan, quarterly goals and other work[edit]

Below we provide brief narratives of our status on each of Editing's quarterly goals, and alongside them the department's teams do a lot of other work to support the annual plan, from support work for other departments to preparations for future quarterly goals.



  • We're working to consolidate our various data sources into one place. We've worked with Analytics Engineering on the historic "ee" dashboards, and have migrated the aggregate numbers to a new dashboard. Product-specific numbers will follow later.
  • Our team members continue to spend a lot of time working to support other teams in Editing and more widely. We're currently working with various teams to help them specify good team-level metrics for their quarterly goals.



  • We're working on the recent changes filters beta feature, which is now available in beta labs and testwiki for early testing/QA. Since it's not at all ready yet, it cannot be enabled by regular users at this stage. We hope to make it available as a beta feature in production some time in February.



  • Our quarterly goal is to provide and apply a style guide which all teams in Editing and more widely follow, as part of "Improve editing experience by reducing technical and product debt". We have implemented most of the changes into OOUI, focussing especially on accessibility-related improvements, and are now gathering feedback and considering documentation.
  • Our team members continue to spend most of their time embedded with other teams.



  • Our quarterly goal, under "Invest in new types of content creation", is to replace the contenteditable editing component of Content Translation to VisualEditor and preserve the current functionalities of Content Translation. We have started the preparatory work for this eventual change with refactoring the existing code. At present the CX classes are being reorganized to be used with OOJS. (T152586))
  • Our work on deploying Compact Language Links out of beta (T66793) is now paused and being reviewed; it is done for all Wikivoyages except Russian, and all Wikibookses, Wikiversities, Wikinewses, Wikiquotes, and Wiktionaries, and for all Wikipedias except English, Swedish, German, Dutch, and French, for which new schedules will be drawn up soon. (T136677)



  • Our quarterly goal, as part of "Improve editing experience by reducing technical and product debt", is to provide 3D model file support. This will involve picking up the work on 3D file support, getting the 3d2png service and 3D extension into production, and working with Reading on MediaViewer integrations.
  • We're gearing up for the work with colleagues on Structured data for Commons. More on this soon.
  • Work on providing rich media file annotations is slower than we had hoped. We are going to provide design options to the Commons community to decide which they prefer, ideally this month or early next.
  • We've paused our work on the non-destructive image editing tool (Standalone demo, MediaWiki demo). because it relies on the Thumbor service that Performance are bringing to production. They believe it will be a small amount of work to modify the service so that we can use it, and should be available soon, when we hope to release this to users.



  • Our quarterly goal within "2: Improve editing experience by reducing technical and product debt" is to replace the Tidy dependency from MediaWiki output. We're planning to provide an alternative to Tidy (disabled by default) and the tools for individual editors to assess the impact of switching on each page. We postponed this to Q3 since we might deploy a PHP version instead which is fairly close to being ready.
  • We're working with Reading and others on support for multi-script wikis (like Chinese), initially just for reading but later for editing too. This work is now awaiting review and related additional fixes before we can declare it completed. We made a bunch of fixes to the core parser to fix some edge cases.
  • We're working on several improvements for better support of media in Parsoid. Our initial work on image galleries is now complete and was deployed in Dec 2016. We updated our HTML spec for review by the Services team. The extension is also being reviewed by the VisualEditor team and might require additional fixes before being enabled for use inside VE. For video and audio files, the specification has been agreed upon and we'll pick up this project again in January 2017.
  • We have started working to finish up and provide to editors the output of the GSoC "linter" project to spot wikitext mistakes and issues for editors to fix. This should help improve the quality of the wikitext corpus and reduce errors for users. We have deployed this to the Beta cluster and expect to roll this out in production in a phased manner starting out with testwiki and smaller wikis.
  • Our work on implementing into MediaWiki generic shadow namespace support continues, albeit slowly, with a series of wide-ranging changes to MediaWiki planned.
  • We've paused our work on providing a system to balance templates, due to competing concerns.
  • We've paused our work on providing a system to isolate styles of templates, due to competing concerns.
  • We've paused work from earlier this year on reducing the bandwidth demands of read-mode pages by considering how to move data-mw out of inline HTML. After discussions between us, Services, VisualEditor, we collectively decided that there is no urgency to deploy this since Reading has its own version to strip this in production, and VE prefers the inlined data-mw in HTML. This will be revisited when we are closer to adopting Parsoid HTML for read views.



  • Our quarterly goal, as part of "Maintain and incrementally improve current content creation and curation interfaces", is to give users visual diffs inside the visual editor. This will let editors see the changes they and others have made without needing to know the intricacies of wikitext, augmenting the current two-column wikitext diff interface. We have an early version of the diffing system, and we're working on the interaction design to support the richness of wikitext edits. We hope to release this for users in the next month or so. Eventually, we will provide it outside the editor too as an option when looking at diffs in any context.
  • We're also working to provide a unified, consistent editing interface between mobile and desktop Web, and between visual and wikitext editors. The initial version of the new wikitext editing mode was released as a desktop Beta Feature for opt-in testing in December. Work on further features for this is underway.
  • Our work to release the visual editor to all Wikipedias is now almost complete. There are now outstanding only the eight multi-script Wikipedias (including Chinese) which are technologically blocked, on which we're working with Parsing. There are also three other Wikipedias in incomplete deployment states (English, Spanish and Dutch), for which work is paused pending the new wikitext editor release, and won't resume until some point, probably in mid-2017.
  • Our work to release the visual editor to sister projects is now paused. Work on the blocker bugs to Wikisource working had stalled as we'd put resources elsewhere, but has recently resumed following the Developer Summit 2017. We hope to get this ready early in 2017.
  • We paused work on the single edit tab integration on other wikis in June 2016, pending the new wikitext editor release, and so won't resume until some point later in 2017.


Discussion about major things happens on the ​wikitech-l mailing list. For specific IRC channels, see the team subpages linked above.

See also[edit]

Other documents[edit]