New Readers/Offline

From Meta, a Wikimedia project coordination wiki
Jump to navigation Jump to search

Other languages:

The New Readers team conducted a series of research efforts over early 2016. From our findings, we're hoping to improve the way that readers in our target countries experience Wikipedia through changes to user interface, awareness, partnerships, and community efforts.

This is our active work space as we're brainstorming and developing. We'd love your input.

Prototypes: give your feedback

We've built 3 prototypes to learn more about what will work best for readers who want to have access to content they find online later, when they might not be connected to the internet. Check them out below, including instructions for how to use them. You can leave feedback about each of the ideas on the talk page.

The prototypes are currently built for Wikipeda content, but if we do push any of these live, they will be usable on any project if the project is interested.

We're also getting feedback from readers who aren't as familiar with Wikimedia projects through design research. We will summarize that feedback and share it later. If you know anyone who might be interested in participating, let us know by emailing Anne Gomez <>.

Because these are prototypes, they are not fully functional. Please note a few things:

  1. All prototypes are designed and built for mobile.
  2. Some functionality isn't included in the prototypes, including Search.
  3. Some functionality might work slowly, as it's running on Tool Labs. This will be resolved in future versions.
  4. Some features may not work in some browsers/on some devices. For example, they likely won't work in proxy browsers like UC or Opera mini.

Prototype #1: "Saved Pages"

A potential new feature that leverages "web workers" to create a progressive web app that allows readers to save articles in their browser for reading offline. Pages saved using this will be available offline, and you can add the Saved Pages screen to your home screen for easier access later.

Phabricator task: T148365

Testing instructions:

  1. On your phone, open this link:
  2. Browse around and look for some articles you want to save. Save them.
  3. Disconnect from your network and navigate back to your saved pages.
  4. Add "saved pages" to your homescreen to find them later.
  5. Think about your experience and give us feedback on the talk page.

Prototype #2: "Wiki-later"

This prototype lets you download an entire article's content as a PDF that is specifically designed for mobile. Save it and view it later, when you don't have internet access or don't want to spend your data.

Phabricator task: T148359

Testing instructions:

  1. On your phone, open this link:
  2. Browse around and download a full article as a PDF. [NOTE: this is slow to generate]
  3. Save the PDF to your device and read it later. [NOTE: the prototype doesn't have current print styles. You can see what this would look like at this link.]
  4. Think about your experience and give us feedback on the talk page.

Prototype #3: "Quickfact"

Phabricator task: T148364

Quickfact is a short version of an article that is easily savable/sharable by social networks. It includes the lead paragraph, image, and some wikidata from the article.

This prototype is in progress, and the file will be a JPG soon (instead of a PDF). These files are very small and easy to share.

Testing instructions:

  1. On your phone, open this link:
  2. Browse around and download a Quickfact [NOTE: this is slow to generate]
  3. Save the file to your device and read it later. [NOTE: the prototype doesn't have current print styles. You can see what this would look like at this link.]
  4. Think about your experience and give us feedback on the talk page.

Design research

In parallel with the feedback we're hoping to get from the above section, we are also conducting interviews with people who are not editors. We will update findings here when we have them, but in the meanwhile here is our process.

Interviews will be conducted from November 28 - December 9, 2016.


The following message (or something similar) was sent to chapter lists, wikitech-l, individuals in the Movement, and others who have expressed interest.

Hello! Wikipedia needs your help to improve. We’re looking for people who have internet access at least sometimes and like to have access to content when they’re not connected to the internet. We would like to learn from you. If you would be willing to participate in a study to tell us more, please fill out this survey [1] before November 25. Testing will only be available in English for now.

Thank you!

[1] This survey is hosted on a third-party service. For information about data handling and privacy, please see the survey privacy statement:

We also posted on the Wikipedia Facebook page and reached out to people who have expressed interest in participating in design research for Wikipedia in the past, sending them to the qualifying survey.

Qualifying participants

We used the following criteria to qualify participants to schedule times for interviews:

  • they have access to equipment and internet to connect for an hour long video call
  • they self-report as finding content online to use and/or share when offline more than once per month
  • they're available during the time frame of the study and willing to participate
  • they have a mobile phone that they use to browse the internet

Interview process

Each interview will be 1 hour long and include a range of questions about internet usage and downloading/sharing content from the internet offline as well as hands-on usage of the 3 prototypes.

The interviews will take place via video call that will be recorded for internal use only. In the calls, we ask participants to turn their laptop around and hold their smartphone in front of the screen so that they can show us how they navigate around their phone while also talking to us. In this way, we are able to not only hear people's responses, but also see how they interact with prototypes.

Next steps

After the 2 weeks of interviews, we will publish results here as well as a wrap-up of any comments we receive from community on the talk page.

We know that this process is limiting in that it only allows us to talk to people who have sufficient access to the internet to participate in an hour-long video conference, and so we plan to conduct in-country research in Q3FY1617 (January - March 2017) in order to get more targeted feedback.

What we're focused on

The New Readers team is currently in the process of mapping out everything around Offline in the Wikimedia Movement. While that is ongoing, we are also focusing product development on the following finding/use case:

People are increasingly getting information online, then consuming or sharing it offline. (Finding 20)

Users are frequently moving what's online to offline for repeated viewing, printing, or sharing. These behaviors are growing along with the tools that make them possible.


  • Nigeria: “Sideloading” and music/video sharing are common practices among the digitally savvy, helping users save on data costs (especially when sharing large files) and making technology and media discovery more social. As a result, file-sharing apps (e.g. Xender) are very popular.
  • India: Offline modes of retaining and exchanging information are gaining popularity. Most commonly cited exchange apps are Xender, SHAREit, and ShareApp. Downloading to print information is another form of offline transfer. Downloading online content, including videos and songs to watch or listen to later, and school assignment materials to print for use or submission, is a widespread behavior. However, saving Wikipedia articles for later was not observed beyond one instance.
  • Mexico: [to be added?]


Personas are used in product development to ground ideas and implementation in real goals, desires, and limitations that we saw from interviews in the field research conducted in early 2016. We'll be using these personas to drive our strategies from design, messaging, and partnerships.

We will be using the following personas:

  • Chris (Reader, Nigeria)
  • Kumari (Reader, India)
  • Wayne (Wikipedia admin, USA)


As with any project, we're collecting examples of how other people both within and outside of the Wikimedia movement are addressing this pattern of user behavior. Please add any examples you know of and/or tell us what you like/dislike about them.



  • Kiwix is an offline reader for web content. It's especially thought to make Wikipedia available offline. This is done by reading the content of the project stored in a file format ZIM, a high compressed open format with additional meta-data. Kiwix also gives you the freedom to copy, modify and distribute the data. It's available in a number of ways with different use cases. It's also open-source, so other people are able to download, modify, and distribute it as they like.
  • Kiwix app: you can get the Kiwix app for Android or iOS and download files containing all or some portion of Wikipedia within the app.
  • Kiwix by USB, Raspberry Pi, etc.: some people are also putting copies of Kiwix (and content) on their own devices for distribution.
  • Medical Wikipedia (Offline): Doc James and the WikiMed project have launched 10 language versions of offline medical app. These include medical, anatomy, medication, and sanitation information in a specific language that can be downloaded directly from Google Play. The medical apps have seen more than 60,000 downloads with about 75% of these being from low and middle income countries.
  • The Wikipedia Apps have a "save for later" feature that allows users to bookmark articles. Those articles are later available offline, but can't be shared.
  • Downloading content.
  • PDF
  • Wikibooks
  • Screenshots
  • Progressive web apps
  • Prototypes
  • RFC for an API driven frontend
  • Partnerships with
  • Internet in a box
  • Endless Mobile
  • EGranary / WiderNet
  • XOWA, offline Wikipedia for desktop

Other offline support


  • Sideloading
  • Xender allows device-to-device transfering of files (pictures, videos, music, documents, apps). It works 1:1 or 1:many. Available for Android, iOS, and Windows phones.
  • SHAREit (by Lenovo)
  • ShareApp
  • P2P
  • WebRTC based p2p frameworks WebCDN: browser based, researchy, needs coordinator server to establish contact (not fully offline)