Program Goals and Status for FY18/19

Goal Owner: Leila Zia
Program Goals for FY18/19: We help Wikimedia editors identify gaps in the content, readership, and contributors of Wikimedia projects, as well as means for reducing such gaps. We enable Wikimedia developers to build products and services that can surface and help reduce Wikimedia knowledge gaps. We research and develop end-to-end technologies and systems that automatically identify such gaps, prioritize them, and recommend actions or frameworks for reducing them.
Annual Plan: TEC9: Address Knowledge Gaps
- Primary Goal is Knowledge Equity: grow new contributors and content
- Tech Goal: Supporting our Community of contributors

Q1 Goals

Outcome 1 / Output 0

One or more of the following: Interested editors will be able to use recommendation services that will allow them to have relevant information about the articles they want to edit immediately at their repository. Newcomers can more easily contribute to Wikimedia projects. Editathon organizers benefit from automatically generated templates and recommendations that can help them in onboarding new or less experienced editors. More readers will have access to content in their local languages. Developers will be able to surface more diverse set of recommendations for article expansion through their tools.

A unit test to measure the bias of recommendation algorithms

Goal(s)

A report of the state of the art on bias detection and algorithm audibility in the context of recommendation systems (based on the review of the literature and industry interviews)

Status

Note: July 2018

To do

Note: August 21, 2018

This is still in a

To do status, this effort will be toward developing a procedural framework and this will be the first step forward.

Note: September 13, 2018

Done Literature review at task T203049

Outcome 1 / Output 2

One or more of the following: Interested editors will be able to use recommendation services that will allow them to have relevant information about the articles they want to edit immediately at their repository. Newcomers can more easily contribute to Wikimedia projects. Editathon organizers benefit from automatically generated templates and recommendations that can help them in onboarding new or less experienced editors. More readers will have access to content in their local languages. Developers will be able to surface more diverse set of recommendations for article expansion through their tools.

Section recommendation algorithm in many languages.

Goal(s)

Build a section recommender system based on the section mapping algorithm

Status

Note: July 2018

To do

Note: August 21, 2018

To do

Note: September 12, 2018

Postponed Moved to next quarter as we needed to spend some more time to improve section mapping algorithm before starting section recommendations. Section mapping is going slowly primarily due to the fact that building a training set relies on experienced bilingual editors to help us label mappings.

Outcome 1 / Output 4

One or more of the following: Interested editors will be able to use recommendation services that will allow them to have relevant information about the articles they want to edit immediately at their repository. Newcomers can more easily contribute to Wikimedia projects. Editathon organizers benefit from automatically generated templates and recommendations that can help them in onboarding new or less experienced editors. More readers will have access to content in their local languages. Developers will be able to surface more diverse set of recommendations for article expansion through their tools.

Public test (vs. production) APIs corresponding to algorithms designed and tested in other outputs

Goal(s)

Build a test API for the section recommendation algorithm in Output 2

Status

Note: July 2018

To do

Note: August 21, 2018

To do This is waiting for Output 2 to be finished first before it can be started in earnest.

Note: September 12, 2018

To do This step could only start if the previous goal would be completed. Given that that goal is pushed to Q2 this one should move to Q2 as well.

Outcome 2 / Output 1

Interested editors, developers, and partners can identify more types of gaps in content

An improved task recommendation gadget or API

Goal(s)

Improve article recommendation API to completion (of the second stage improvements)

Status

Note: July 2018

In progress

Note: August 21, 2018

In progress The database blockers have been removed and this should be in progress.

Note: September 13, 2018

Done The algorithm for recommendation articles to be translated is improved such that instead of ranking based on pageviews in the source language, it now ranks missing articles based on the the probability that the article should exist in the destination language considering sitelink counts as well as pageviews in the top 50 WP languages that have the article. We will continue the improvements in Q2, this task has met the intended goal for Q1.

Outcome 2 / Output 2

Interested editors, developers, and partners can identify more types of gaps in content

A framework for understanding and measuring the knowledge gaps and inequality of access to knowledge that includes reader representation by demographics and characterizes readers who come to Wikipedia based on their readership characteristics as well as demographics.

Dependency: Formal collaborators

Goal(s)

Explore the interlanguage navigation patterns as a first approach to understand knowledge gaps in specific languages. Done
Characterize Wikipedia readers across languages based on survey responses, request, article, and session activity. Done

Status

Note: July 2018

In progress

Note: August 21, 2018

Partially done The second portion of this goal is done and will be captured in outcome 4.

Note: September 12, 2018

Done The documentation for the first goal and the second goal.

Outcome 3 / Output 1

More minority voices and diverse newcomers in Wikimedia projects can stay longer on the projects to contribute.

An improved socio-technical framework to remove the barriers for contribution by populations that are currently considered minorities on our projects.

Dependencies: Formal collaborators & Legal

Goal(s)

Run an experiment to test the effectiveness of the first design of the framework
Provide an early analysis of the experiment and iterate if needed

Status

Note: July 2018

In progress &

To do

Note: August 21, 2018

In progress &

To do

Note: September 13, 2018

In progress Details: Given that the second stage of the first goal in Outcome 3 / Output 2 failed (see below for details), we will need to revise our strategy for this part of the study as well. We expect to take part of this goal to Q2 and the remainder to Q3.

Outcome 3 / Output 2

More minority voices and diverse newcomers in Wikimedia projects can stay longer on the projects to contribute.

An algorithm to address Wikipedia's cold-start problem when it comes to learning user interests when they join the project.

Dependencies: Formal collaborators & Legal

Goal(s)

Run the experiment to test the quality of the algorithm to elicit user interests
Analyze the result of the experiment. Devise next steps.

Status

Note: July 2018

In progress &

To do

Note: August 21, 2018

In progress &

To do this is on track for a September deadline.

Note: September 13, 2018

Done Details: Both bullet points are done. However, the second stage of the experiment failed in that we did not manage to gather enough responses from users that had participated in the first stage of the experiment. We will introduce two new goals based on the result of our learnings for Q2.

Outcome 4 / Output 1

More decision makers can make more informed decisions about the audiences to target, the gaps to prioritize, and other research findings. More researchers can build on top of the knowledge generated through this research.

Citable knowledge about the state of gaps in Wikimedia projects, the needs of Wikimedia users by demographics, and beyond.

Goal(s)

Submit the research for characterizing Wikipedia readers across languages for publications (stretch)

Status

Note: July 2018

In progress

Note: August 21, 2018

Done

Q2 Goals

Outcome 1 / Output 0

One or more of the following: Interested editors will be able to use recommendation services that will allow them to have relevant information about the articles they want to edit immediately at their repository. Newcomers can more easily contribute to Wikimedia projects. Editathon organizers benefit from automatically generated templates and recommendations that can help them in onboarding new or less experienced editors. More readers will have access to content in their local languages. Developers will be able to surface more diverse set of recommendations for article expansion through their tools.

A unit test to measure the bias of recommendation algorithms (task T203048)

Goal(s)

Iterate on and improve the report of the state of the art on bias detection and algorithm audibility in the context of recommendation systems through internal subject matter expert and stakeholder input (task T206051)

Status

Note: October 18, 2018

In progress. Reviewing a set of 6 process frameworks developed by governments, Tech Orgs and NGOs for addressing harmful bias in algorithmically-driven tools has begun, which will help to synthesize locally-applicable "best practices". This is in preparation to start to work in gathering input from internal subject matter expert and stakeholders.

Note: December 14, 2018

This goal might be delayed ...

Outcome 1 / Output 2

One or more of the following: Interested editors will be able to use recommendation services that will allow them to have relevant information about the articles they want to edit immediately at their repository. Newcomers can more easily contribute to Wikimedia projects. Editathon organizers benefit from automatically generated templates and recommendations that can help them in onboarding new or less experienced editors. More readers will have access to content in their local languages. Developers will be able to surface more diverse set of recommendations for article expansion through their tools.

Section recommendation algorithm in many languages. (task T203044)

Goal(s)

Build a section recommender system based on the section mapping algorithm

Status

Note: October 2018

In progress. We are investigating ways to address section dependence (task T183038), an important step prior to start fusing the results from translation detection and synonym detection.

To do November 2018

In progress. We are making progress on task T183038 and we expect to have the results of that work ready by the end of November. This will allow us to start building the section recommendation system.

Note: December 14, 2018

This is

Partially done and will be finished up by next week.

Outcome 1 / Output 4

One or more of the following: Interested editors will be able to use recommendation services that will allow them to have relevant information about the articles they want to edit immediately at their repository. Newcomers can more easily contribute to Wikimedia projects. Editathon organizers benefit from automatically generated templates and recommendations that can help them in onboarding new or less experienced editors. More readers will have access to content in their local languages. Developers will be able to surface more diverse set of recommendations for article expansion through their tools.

Public test (vs. production) APIs corresponding to algorithms designed and tested in other outputs (task T203046

Goal(s)

Build a test API for the section recommendation algorithm in Output 2

Note that the completion of this goal is dependent on the goal in Outcome 1 / Output 2.

Status

Note: October 2018

To do

To do November 2018

To do This is still at todo stage since first the section mapping system should be completed.

Note: December 14, 2018

This is still

Postponed until Q3

Outcome 2 / Output 1

Interested editors, developers, and partners can identify more types of gaps in content

An improved task recommendation gadget or API (task T203041)

Goal(s)

Improve article recommendation API to completion (of the second stage improvements)

Status

Note: October 2018

In progress. With the removal of hard-coded Interlanguage Links in 2015, we need to revisit the problem of how to identify articles that are not linked via Wikidata but are still (almost) the same. task T207406

Note: November 2018

In progress We have made good progress on this front and expect the work that was planned for this quarter to finish by the end of the quarter. The updated API will be shared by then.

Note: December 14, 2018

This goal is

Partially done and will be completed by end of next week.

Outcome 2 / Output 2

Interested editors, developers, and partners can identify more types of gaps in content

A framework for understanding and measuring the knowledge gaps and inequality of access to knowledge that includes reader representation by demographics and characterizes readers who come to Wikipedia based on their readership characteristics as well as demographics. (task T203042)

Dependencies on: Formal collaborators, Legal, and community discussions.

Goal(s)

Expand the taxonomy of Wikipedia readers to include questions about demographics.
Prepare the infrastructure (both technical and legal) for conducting the survey(s)
(stretch) Run the survey in one or more Wikipedia languages.

Status

Note: October 2018

In progress. Literature review and brainstorming in progress.
To do. dependent on the completion of the previous step.
To do. same as above.

To do November 2018

In progress. Literature review and brainstorming in progress.
In progress
N Postponed. We expect that the running of the survey to be moved to Q3 as the start of the English campaign is near and we are too close to the holiday season now and that can introduce seasonality into the resutls.

To do December 2018

Expand the taxonomy of Wikipedia readers is

Done, preparing the infrastructure is still

In progress and should be

Partially done by the end of this year. The stretch goal has been

Postponed until Q3.

Outcome 3 / Output 1

More minority voices and diverse newcomers in Wikimedia projects can stay longer on the projects to contribute.

An improved socio-technical framework to remove the barriers for contribution by populations that are currently considered minorities on our projects. (task T201705)

Dependencies: Formal collaborators & Legal

Goal(s)

Devise the framework for matching newcomers to improve the first design of the framework
(stretch) Run an experiment to test the effectiveness of the first design of the framework

Status

Note: October 2018

In progress. Ongoing discussions and brainstorming between researchers and Growth team.
To do. Dependent on the completion of the previous step.

Note: December 14, 2018

The first portion of this goal is still

In progress and will go into Q3 and the stretch goal will also be finished up in Q3.

Outcome 3 / Output 2

More minority voices and diverse newcomers in Wikimedia projects can stay longer on the projects to contribute.

An algorithm to address Wikipedia's cold-start problem when it comes to learning user interests when they join the project. (task T201706)

Dependencies: Formal collaborators & Legal

Goal(s)

Develop a new experiment plan for testing the quality of the algorithm to elicit user interests
Conduct the experiment
(stretch) Analyze the result of the experiment. Devise next steps.

Status

Note: October 2018

In progress. Ongoing discussions and brainstorming between researchers and Growth team.
To do. Dependent on the completion of the previous step.
To do. Dependent on the completion of the previous step.

Note: December 14, 2018

All goals are now

Done

Outcome 3 / Output 3

More minority voices and diverse newcomers in Wikimedia projects can stay longer on the projects to contribute.

A series of baseline statistics on contributor diversity in one or more Wikimedia projects. (task T201707)

Dependencies: Formal collaborators & Legal

Goal(s)

TBD: Note that there are ongoing discussions in the phabricator task linked above to see if this output should be done in Q2 (vs. Q3 as the work in Q2 is becoming quite massive;)

Status

Note: October 2018

We will need to wait for the result of the first bullet point out of Outcome 1, Output 2 before we can pick this up. As noted in the Goals, this may be delayed to next quarter.

Note: December 14, 2018

This goal was not defined for Q2 and will be captured in the following quarters.

Outcome 4 / Output 1

More decision makers can make more informed decisions about the audiences to target, the gaps to prioritize, and other research findings. More researchers can build on top of the knowledge generated through this research.

Citable knowledge about the state of gaps in Wikimedia projects, the needs of Wikimedia users by demographics, and beyond. (task T201699)

Goal(s)

Finalize the documentation for the research on characterizing Wikipedia readers
A series of presentations to teams in WMF as well as language communities about the results on characterizing Wikipedia readers.
(stretch) Submit the research for aligning article sections across languages

Status

Note: October 2018

In progress: check task T201702 for more details.
In progress: 5 given so far. details at task T207096, task T203990, task T206266, task T206362, task T206782 all under the umbrella task T201699.
In progress: we are working hard against the upcoming abstract deadline on October 29 and full paper deadline on November 5. task T201715

To do November 2018

In progress: The bulk of the work is done and the long tail of finalizing details is left. We expect to be able to finalize this by the end of Q2.
Done
N Postponed We decided to move this goal to Q3 and target a different conference (SIGIR 2019) as we needed some time to improve the results of the research.

Note: December 14, 2018

All goals will be

Done by end of next week, but the stretch goal will be

[postponed until Q3.

Q3 Goals

Outcome 1 / Output 0

One or more of the following: Interested editors will be able to use recommendation services that will allow them to have relevant information about the articles they want to edit immediately at their repository. Newcomers can more easily contribute to Wikimedia projects. Editathon organizers benefit from automatically generated templates and recommendations that can help them in onboarding new or less experienced editors. More readers will have access to content in their local languages. Developers will be able to surface more diverse set of recommendations for article expansion through their tools.

A unit test to measure the bias of recommendation algorithms task T203048

Goal(s)

Share process proposals with a broader internal audience and elicit feedback task T212438 Done
Identify, prioritize, and document open questions and key considerations for implementing specific proposals at WMF task T212440 Done
Incorporate best practices and feedback from external subject matter experts task T212443 Done
Finalize and publish report v1.0 task T212445 Done

Status

Note: January 17, 2019

A variety of meetings (during All-Hands and otherwise) are set up with external and internal people to achieve the first 3 goals. The last goal can be picked up once all other ones are completed.

Note: February 22, 2019

The first two goals are

Done. The other two goals are

In progress. We are working with Policy, Scoring Platform, and Berkman Klein center to organize a workshop in April or May 2019 and that can help with gathering more input from external subject matter experts. The third goal corresponding to gathering external expert input may as a result be extended to Q4. The last goal is

In progress and will be published by the end of March (current quarter).

Note: April 17, 2019

The report is finalized and soon will be published on Figshare. I feel comfortable to call the last goal listed above as

Done based on that. The third goal listed needs some more iterations in Q4.

Outcome 1 / Output 1

One or more of the following: Interested editors will be able to use recommendation services that will allow them to have relevant information about the articles they want to edit immediately at their repository. Newcomers can more easily contribute to Wikimedia projects. Editathon organizers benefit from automatically generated templates and recommendations that can help them in onboarding new or less experienced editors. More readers will have access to content in their local languages. Developers will be able to surface more diverse set of recommendations for article expansion through their tools.

Improved section recommendation algorithm with user-feedback

Goal(s)

Elicit feedback from both experienced and new editors about the section recommendation system to identify feedback types task T225136 In progress
Update the section recommendation model to include user feedback for improvements task T215347 Done

Status

Note: January 17, 2019

We are working on improving the section recommendation system and once that's finalized, we will start the work on these two items.

Note: February 22, 2019

Both goals are still

In progress. No major blockers exist.

Note: April 17, 2019

We ran one set of feedback collection and we received two lines of feedback. While we could call the goals as

Partially done based on those, we see a need for some more deeper feedback. As a result, for now we keep the status as

In progress and continue working on it in Q4 until we see enough feedback and results to help us iterate on the model.

Outcome 1 / Output 2

One or more of the following: Interested editors will be able to use recommendation services that will allow them to have relevant information about the articles they want to edit immediately at their repository. Newcomers can more easily contribute to Wikimedia projects. Editathon organizers benefit from automatically generated templates and recommendations that can help them in onboarding new or less experienced editors. More readers will have access to content in their local languages. Developers will be able to surface more diverse set of recommendations for article expansion through their tools.

Section recommendation algorithm in many languages. task T203044

Goal(s)

Improve the section recommender system task T215348 Done

Status

Note: January 17, 2019

We are working on this actively.

Note: February 22, 2019

This work is

In progress and we're on track with it.

To do March 2019

Discussed...

Outcome 1 / Output 4

One or more of the following: Interested editors will be able to use recommendation services that will allow them to have relevant information about the articles they want to edit immediately at their repository. Newcomers can more easily contribute to Wikimedia projects. Editathon organizers benefit from automatically generated templates and recommendations that can help them in onboarding new or less experienced editors. More readers will have access to content in their local languages. Developers will be able to surface more diverse set of recommendations for article expansion through their tools.

Public test (vs. production) APIs corresponding to algorithms designed and tested in other outputs task T203046

Goal(s)

Build a test API for the section recommendation algorithm in Output 2 Done

Status

Note: January 17, 2019

You can test the current model in en, fr, es, ar, ru, ja via http://diegotest.wmflabs.org/en/Quilombo. We expect to improve the quality of the recommendations further in this quarter.

Note: February 22, 2019

This goal is

Done. We have built a simple UI for testing the API with community members as well. https://secrec.wmflabs.org/

Outcome 2 / Output 1

Interested editors, developers, and partners can identify more types of gaps in content

An improved article recommendation pipeline and API task T203041

Dependancies on: Analytics

Goal(s)

Set up the production pipeline to generate recommendations in Hadoop task T210844 Where applicable (we're still blocked on other parts)
Improve the current API to filter out newly created articles task T213866 Done
Improve the model for identifying similar Wikidata items task T210433 In progress

Status

Note: January 17, 2019

There are quite a few team dependencies for having this work completed. At the moment, we are making good progress.

Note: February 22, 2019

Some of the smaller tasks have been completed. We're still

Blocked on some of the items like task T213976, which will happen in Q4, it seems. We also switched focus to page-links-change event stream during this month. Now that it's in a good shape, we'll pick up this output during March again and work on tasks such as task T210844 that are not blocked on other teams.

Note: April 17, 2019

Parts of the production pipeline that don't depend on Analytics have been automated. Once we're unblocked on task T209655 and task T213976, we'll be able to fully automate the pipeline.

Outcome 2 / Output 2

Interested editors, developers, and partners can identify more types of gaps in content

A framework for understanding and measuring the knowledge gaps and inequality of access to knowledge that includes reader representation by demographics and characterizes readers who come to Wikipedia based on their readership characteristics as well as demographics.

Dependancies on: Formal Collaborators, Legal, Reading Web

Goal(s)

Finalize the choice of Wikipedia languages for running the demographics survey task T212442 Done
Work with Legal to develop a privacy statement for the new round of survey task T212446 Done
Run the survey in one or more Wikipedia languages task T212444 Done
~~Prepare the data for analysis in the following quarter task T212448~~

Status

Note: January 17, 2019

We are on track for this goal. Conversations with Legal about the type of demographic questions we can ask is ongoing and our understanding is that we can conclude them soon. We are going to start working with the community on 2019-01-17 to identify potential languages the surveys will be launched in. We expect a test (in 1-2 wikis) no later than mid-February and if the results of the test are fine, a larger scale survey no later than the end of February, or first week of March.

Note: February 22, 2019

The first goal is

In progress and we expect to finalize the list no later than 2 weeks from now. We have already made a call in wikimedia-l about it. The second goal is almost completed,

Partially done. The third goal: the first test survey is scheduled for 2019-02-27 and this goal is now {[in progress}}. We expect to be able to meet the last goal by the end of March.

Note: March 14, 2019

The choice of languages (first goal) is

Partially done and will be completed by 2018-03-21. The two goals after that are

Done. As part of the third goal, we ran a pilot test and we are seeing some unexpected distribution of demographics as well as unexpected quicksurvey results in mobile. We have put a hold on running the survey in many more languages until we address that. The fourth goal is dropped as a result.

Outcome 3 / Output 1

More minority voices and diverse newcomers in Wikimedia projects can stay longer on the projects to contribute.

An improved socio-technical framework to remove the barriers for contribution by populations that are currently considered minorities on our projects. task T201705

Dependancies on: Formal Collaborators, Growth Team (potentially)

Goal(s)

Improve the current design of the model (to be tested) for increasing the retention of women Blocked
(conditional on the result of the first goal) Prepare the infrastructure for running the experiment To do
(stretch and conditional on the result of the first goal) run the experiment To do

Status

Note: January 17, 2019

We are working on the first goal. Note that finding a model has proved itself to be very difficult as the matching of users and them collaborating has a lot of barriers at the moment. We did test this approach last quarter and could not make it work, and our hypothesis is that because of the complexity of communications between users in Wikipedia, users don't participate. We are iterating on the model now but we should note that tighter collaboration between Audiences and Tech may be needed to make this work.

Note: February 22, 2019

This goal is now

Blocked as a result of midterm planning and transitions in Research. We will get to it if possible during this quarter but we cannot promise this at this point.

Note: April 17, 2019

Leila discussed this set of goals with Erika Bjune/ Acting CTO, and we decided to

Postpone these goals for Q4. Over the past 9 months we have tried a few different ways for testing the framework that we have designed, however, the tests require front-end changes and we have not been able to mobilize resources within the Foundation to make this happen. This work remains to be of critical importance to the Foundation and we highly recommend coming back to it in the context of FY19-20 annual planning and figure out what structures we need in place to be able to do this kind of tests. We expect this line of testing to require deep collaborations between Audiences and Technology departments.

Outcome 4 / Output 1

More decision makers can make more informed decisions about the audiences to target, the gaps to prioritize, and other research findings. More researchers can build on top of the knowledge generated through this research.

Citable knowledge about the state of gaps in Wikimedia projects, the needs of Wikimedia users by demographics, and beyond. task T201699

Goal(s)

Submit the research on aligning article sections across languages to a peer-reviewed publication venue task T201715 Done
Submit the research on eliciting new editor interest models to a peer-reviewed publication venue task T201709 Done

Dependancies on: Formal collaborators

Status

Note: January 17, 2019

The first submission is completed. The second will happen on January 28.

Note: February 22, 2019

Both goals are

Done now.

Q4 Goals

Outcome 1 / Output 0

One or more of the following: Interested editors will be able to use recommendation services that will allow them to have relevant information about the articles they want to edit immediately at their repository. Newcomers can more easily contribute to Wikimedia projects. Editathon organizers benefit from automatically generated templates and recommendations that can help them in onboarding new or less experienced editors. More readers will have access to content in their local languages. Developers will be able to surface more diverse set of recommendations for article expansion through their tools.

A unit test to measure the bias of recommendation algorithms task T203048

Goal(s)

(carry-over from Q3) Incorporate best practices and feedback from external subject matter experts through feedback at Community Data Science Collective workshop and the Round-table in Georgetown Institute for Technology, Law, and Policy task T212443 Done

Status

To do April 2019

Discussed...

To do May 2019

Discussed...

To do June 2019

Discussed...

Outcome 1 / Output 1

One or more of the following: Interested editors will be able to use recommendation services that will allow them to have relevant information about the articles they want to edit immediately at their repository. Newcomers can more easily contribute to Wikimedia projects. Editathon organizers benefit from automatically generated templates and recommendations that can help them in onboarding new or less experienced editors. More readers will have access to content in their local languages. Developers will be able to surface more diverse set of recommendations for article expansion through their tools.

Improved section recommendation algorithm with user-feedback

Goal(s)

(carry-over) Elicit feedback from both experienced and new editors about the section recommendation system to identify feedback types task T215347 In progress
(carry-over) Update the section recommendation model to include user feedback for improvements task T215347 Done
Parameter alignment in templates (service to Languages team) task T221211 Done

Status

To do April 2019

Discussed...

To do May 2019

Discussed...

To do June 2019

The first goal is in-progress and we intend to meet it early enough that by July 15 the second goal can be achieved, too. The last goal was added in place of another goal to build a model to understand the context of the section recommendations. Instead, we decided to use similar approaches as developed for section alignment to help Languages team with parameter alignment within Templates across languages.

Outcome 1 / Output 5

One or more of the following: Interested editors will be able to use recommendation services that will allow them to have relevant information about the articles they want to edit immediately at their repository. Newcomers can more easily contribute to Wikimedia projects. Editathon organizers benefit from automatically generated templates and recommendations that can help them in onboarding new or less experienced editors. More readers will have access to content in their local languages. Developers will be able to surface more diverse set of recommendations for article expansion through their tools.

The first version of the algorithm that prioritizes missing sections

Goal(s)

Develop the first version of the model that can prioritize missing sections task T227651 Done

Status

To do April 2019

Discussed...

To do May 2019

Discussed...

To do June 2019

This goal is in-progress and will be completed by July 15.