Talk:Reading/Strategy/Kickoff

Please respond by September 8, 2015.

If you are instead looking to provide feedback on the general strategy process, please visit the main Reading strategy talk page. Otherwise, we would appreciate your answers to the following questions (see kickoff content page for further context) -

Horizon for achieving winning aspiration: 2–5 years

1. Please help the [foundation:Staff_and_contractors#Reading WMF Reading] Department complete the following sentence: We aspire for a movement where in 2–5 years readers ...
2. Where should the WMF Reading department focus to fulfill aspiration #1?
   • In which demographics and user segments?
   • On which distribution channel(s) and technological platforms?
   • What types of product categories and services?
   • Across which geographies?
3. What approach should WMF Reading pursue to achieve success in the focus areas from #2, to deliver on its mission in a way that distinguishes the user experience in a meaningful way?

If we knew how people read / used our content it would help the editor community determine how it should be structured.

As part of the strategy process I believe one of the most important (if not the most) questions we are looking to answer from our readers is "Why are you reading this article?" so that we can learn why they are consuming the content. Is it because of checking a quick fact? Is it to quickly learn about something unknown? (definition & pictures) Is it to deeply research a topic? Is it general reading?

It is very important that we start to understand better our readers, to inform and create ways for editors to display the content and create better reading experiences.

This is a great point.

we had reader feedback, but the community was dismissive.

given the lag time for producing content, we might want to think about writing where the readers will be in the future. quickly responding to trending "news" is reactive.

Our work isn't jsut to let people find what they want to find. We must show readers things that they get interested in reading, even if they didn't try to find them.

We had an interesting piece that most people do not read beyond the lead. ~~~~

Here's the report I remember: https://meta.wikimedia.org/wiki/Research:Which_parts_of_an_article_do_readers_read

Very insightful, IMO

For instance, taking the example from below: "if you aspire for access to everyone", then you think we should focus solutions that help provide equal experience for everyone regardless of their internet speed? Or we should provide better cross platform support? Or we should focus on certain georgaphies that have least access, currently.

Focus areas for: We aspire for a movement where in 2-5 years readers have access to excellent reading experiences wherever they are, independently of which country they live in.

- User research should be a primal focus to inform state of reading experiences.

- Design and User experience evangelization on editor communities.

- Excellent content distribution story with lower connectivity. (Better dumps story and better offline reading experiences).

- Understanding different parts of the world and how the content could reach the readers.

- Targeted reader experiences from the same content that adapt and show the content optimized for the readers of a specific segment.

The WMF Reading Department should develop tools that help Wikimedia editors to improve the reading experience of projects.

Thanks @NaBUru38 for your reply. Would you please elaborate examples for such tools?

For example, wide tables look awful in small screens. There should be a better way of displaying them. For example there should be optional breaks in lists, so the wider the screen, the more columns appear.

Also, I've already mentioned Wikipedia portals . We should have better tools to build them, like generating summaries of similar articles automatically (I do that manually).

You are already doing interesting things with the mobile apps, and I hope that continues - Wikimedia data should be available at your fingertips in any situation you need it.

However, I would also like to see a much greater emphasis on the quality of the knowledge presented to the reader. Wikimedia has been reluctant to meddle in the community editing processes, and for good reasons - any sense of top-down control scares contributors away. But maybe your department could come with ways to strengthen the incentives to contribute reliable and neutral information, and work to gain support from the editor community to implement them.

For instance, how about a simple setting in the mobile app that make it only search in "featured" and/or "good" quality articles - this would make it easier for the reader search just the good stuff, while also incentivizing the editors to produce good stuff, since editors contribute because they want their content to be read.

1. Please help the WMF Reading department complete the following sentence: We aspire for a movement where, in 2 to 5 years, readers... who need information on matters that seriously impact their quality of life can use Wikipedia to find answers they can trust, in a format they can easily understand.
2. Where should WMF Reading department focus in order to fulfill the aspiration from #1?
   • In which demographics and user segments?
     Segment users by their needs. There is a large edutainment / infotainment segment who want to read interesting things on Wikipedia, or use it to understand their leisure activities better. Providing for that is a valuable thing but not the main thrust of our mission; features that make the site more flashy or entertaining or try get people to browse more random stuff should not be a focus.
     There are various other segments, people in situations where the availability of a wide, deep, reliable and accessible information source can have life-altering consequences, and this should be the main focus of the movement. Health and education are two prime examples; so is, in a less direct way, information on topics related to major policy questions or social change. People belonging in these groups care a lot about reliability (or maybe not but will still be strongly impacted by trusting unreliable information). For some of these groups, accessibility is important (e.g. laypeople looking for medical information, people in poor living conditions looking for sustainable technologies, people in areas with poor internet connections), for some not so much (e.g. students will usually have the time and skill to figure out difficult or badly written text, and will have high-quality internet access).
     These segments probably don't correlate strongly with demographics, although students and less affluent groups might be more often in serious need of information.
   • On which distribution channel(s) and technological platforms?
     We have two large distribution channels: search engines and social media. The first is mainly useful for people looking for specific information; the second is less targeted and makes it possible for people to encounter information that is important for them but they weren't aware they needed it. Both of those are important and should be examined, although the current culture and content formats of the Wikimedia movement are better aligned with search engines.
     We should research alternate channels that can reach demographies not available via these channels (typically areas with poor or no internet connection), such as offline content bundles.
     As for platforms, focus resources instead of dispersing them, and focus on the platforms where most of the users are: the desktop and the mobile web (which would ideally converge over time), and provide great APIs so that external players can cover other platforms.
   • Across which geographies?
     All of them; the Global South has a larger need for information but also much more severe accessibility problems so how much effort goes there should depend on how much ability we realistically have to bridge those.
3. What approach should WMF Reading pursue to achieve success in the focus areas from #2, in order to deliver on its mission in a way that distinguishes the user experience in a meaningful way?
   • Invest into user research and better metrics and consult with groups with similar challenges (journalists, educators, activists) to develop an ability to measure accessibility and perceived reliability and usefulness of content.
   • Similarly, develop an ability to measure demand; something that - unlike page views - is weighted towards serious uses of our content.
   • Facilitate expert reviews to measure actual reliability of content samples; research techniques to predict reliability over a wider range of content based on that.
   • Expose reliability information to readers (e.g. somthing like WikiTrust); provide demand/reliability/accessibility information to editors so they can prioritize work (give the WP 1.0 idea some software support).
   • Invest into research and consultation with groups experienced in the area to identify the main accessibility barriers (in the wide sense of accessibility; could include anything from no internet connection to inability to reading complicated English prose to lack of time or motivation to read long-form text) and work on removing them.
   • Collaborate with reusers who are focused on reliability and accessibility (Google Knowledge Graph being the obvious example here); provide great APIs for fetching content (including sub-article-level content).

Many of our readers want rich content and want to be able to easily access our current content in different ways. For example this study found that 82% of people in Kenya want spoken content http://translatorswithoutborders.org/sites/all/themes/twb/pdfs/TWB%20Impact%20Study%20Infograph.pdf

We need to partner with other organizations to increase the number of videos we have. There are a lot of great CC BY SA NC / ND videos out their. The organizations that produce them are happy for us to use them as we are NC. Is there a need for an exception for this content? Yes I know this is a bit of a push. But we allow fully copyright images under specific rules right now.

There' have been some experiments in the past with automated text to speech engines at hackathon events. The identified problems back then were mostly:

• Cost (commercial, or self hosting were both expensive)
• English only
• What is the workflow. How do people find it and how do they use it.
• We need a Free/Open audio format and player.

As for video... We need to be able to sustain video first. That part of our technology stack is nightmare'ish..

rich content including video, for just about every highly viewed article. (channeling Andrew Lih) they are working around at Schnittserver at Internet Archive for raw footage, and then uploading finished to commons; VIDEO PRODUCTION TOOLS AND COMMUNITY IN THE WIKIVERSE

this could be supported, and championed

Rich content is harder to develop than plain text. Of course it's richer for readers, but it's costlier for us.

Video/audio are also much harder problems for collaborative editing and ease of updates. For some pages/topics this may not be a problem (core STEM topics come to mind) but for anything that is in flux (living biography, current or near term events) this would be a major challenge.

it is a major challenge, and ugly process, and bandwidth hog, but the future. it will require much training of video makers in editing and production there is no technical reason we cannot have a video of every blp speaking, (upping the audio recording) a link to youtube is not good enough

For example: We aspire for a movement where in 2-5 years readers everywhere will have access to our projects. Based on your vision you can complete questions 2 and 3.

I'd aspire for a movement where in 2-5 years readers will have more control over how information is presented. Our ability to feed people what they want in my opinion would benefit from a personal profile, where someone can say. "Feed me math content on an elementary level (introduction only, no complex formulas, or even a new "explain like I'm 5" paragraph), but read me history pages with full narrative. This becomes even more relevant when users aren't in front of a UI, but ask for texts to be spoken out loud (e.g. via Amazon Echo), as spoken texts emphasize linear flow of the article more.

I aspire for a movement where in 2-5 years visually impaired readers can access the same material (text as well as photos) as able bodied readers, in multiple languages.

We aspire for a movement where in 2-5 years readers... can read good content about vital articles and trending articles of readers.

Contributing as a user or dev is as easy and common as Facebook or GitHub.

We aspire for a movement where in 2-5 years readers have access to excellent reading experiences wherever they are, independently of which country they live in.

... will find articles on relevant topics, written by real people, who made conscious decisions about focus and content of the articles. Do not go the way of bot-created-articles, based on data - be it from WikiData or other sources. Data can at best support knowledge, but never replace it. Knowledge is the connection between data, usable knowledge is curated by editors, who evaluate the relevancy of certain aspects with regard to the subject.

... can find interesting, relevant, accurate and accesible information.

exist

Please don't forget the Wikipedia portals. They are a great way to show interesting information to readers.

We have some very large infoboxes. They are collapsed within the app but on the mobile site come before any other content. We should consider collapsing on mobile. Or at least allowing parts of them to be collapsed.

I believe the Beta or Experimental mode of the mobile website is already experimenting with this.

@TheDJ In the alpha mode we were hiding the infobox and substituting it with a wikidata generated infobox but that experiment is being phased out. Right now I don't think we have anything like what the apps do.

@Doc James I've created T109703 to track this request. I would appreciate if you chimed in there and expanded your thoughts. Your input would be extremely valuable.

The term "user groups" has a technical meaning in the realm of affiliation, which isn't what you meant here, presumably. See m:Wikimedia User Groups.

You're right. I removed the term "user groups" and changed "segments" to "user segments".

I did that on the content page. I also realized I needed to do this on the "About this board" box as well, so did that too :)

The picture inside the app is the first picture within the article usually. However for the article metoprolol this does not occur? Wondering why?

If I'm correct, the logic for this is in Extension:PageImages, which job it is to filter and select an appropriate image from the page and provide that. In this case, i suspect that the image is every so slightly too small (probably in height) to be recognized as a good candidate. Alternatively it can be that it has trouble finding it inside this specific type of infobox.