Grants:IEG/Improve 'Upload to Commons' Android App/Final

This project is funded by a Project Grant

proposal

final report

Welcome to this project's final report! This report shares the outcomes, impact and learnings from the Individual Engagement Grantee's 6-month project.

Part 1: The Project

Summary

In a few short sentences, give the main highlights of what happened with your project. Please include a few key outcomes or learnings from your project in bullet points, for readers who may not make it all the way through your report.

• The current version of the app is much more stable, and has many improvements that make life easier for the user, such as category suggestions based on the title of the upload, and a notification if the about-to-be-uploaded picture matches one that is already in the Commons database. It also has a new tutorial to educate new users on what types of photos should or should not be uploaded.
• We implemented a list and map of nearby places that need photos based on Wikidata, which turned out to be hugely popular among our users. For instance, Wikimedia Czech Republic has mentioned that this feature helps them with their media acquisition support program, where one of their goals is to take pictures of all unphotographed yet geo-located items on Wikidata.
• At the end of the grant, the app had 2565 active installs and had attracted 700+ new Commons contributors. 20554 new files were uploaded via the app during the grant period with an average deletion rate of 15.74%, and 3485 images that were uploaded via the app were used in Wikimedia articles.
• 8 new volunteer developers joined our community on GitHub, and a diverse group of developers came together to work on the app during the Wikimedia Hackathon 2017.

Methods and activities

What did you do in project?

Please list and describe the activities you've undertaken during this grant. Since you already told us about the setup and first 3 months of activities in your midpoint report, feel free to link back to those sections to give your readers the background, rather than repeating yourself here, and mostly focus on what's happened since your midpoint report in this section.

Improvements and new features in the app

Below I have listed the improvements and new features that were implemented in the app during the second half of the grant. The improvements for the first half of the grant can be found in the midpoint report.

• Added page to tutorial to educate new users on what types of images are useful. We added new pages to the tutorial with examples of ‘acceptable uploads’ and ‘bad uploads’, as well as an example of how to fill in the form fields for an image. The new tutorial is based on a draft of the Commons section of Pine’s educational video script, and modified to better accommodate issues specific to mobile app uploading. It was released under the Apache 2.0 license (same as the app), and the new pictures used in the tutorial were taken by myself and Nicolas Raoul, and released under the Public Domain license.

• 
  Tutorial page 1
• 
  Tutorial page 2
• 
  Tutorial page 3
• 
  Tutorial page 4
• 
  Tutorial page 5

• Added option to copy title and description from previous upload. A button was added to allow users to use the same title and description as the previous upload, to make it more convenient for people who upload several images of the same topic consecutively.

• Category suggestions based on entered title. In addition to suggestions for recent categories and nearby categories, we have added a new type of suggestion: title-based categories. A search is performed for categories matching the title that the user has entered for the image, via asynchronous queries to the Commons MediaWiki API. This is the same API that is used for the as-you-type category search.

• Filtered out year categories. Categories such as”X in 2006″ are very unlikely to be appropriate, so we have filtered out categories containing 4-digit numbers starting with ’19’ and ’20’ from all category suggestions. An exception is made for categories containing the current year or previous year. E.g. if a user searches for “Burning Man”, he should see a suggestion for ‘Burning Man 2015’ but not ‘Burning Man 2005′. The rationale for this design choice is that people are unlikely to store more than 2 years’ worth of pictures in their phone. So if someone has pictures of Burning Man on their phone, it is likely to be for the 2015 or 2016 event, not for 2005 etc.

• Prevented uploading of multiple copies of exact same image. As requested on the Commons village pump, we implemented a check to see if the exact same image already exists in the Commons database, in order to prevent unintentional duplicate uploads. This was done by checking for similar SHA1 hashcodes using this API

• Updated licenses and added option to select them directly from upload screen. CC-BY-4.0 and CC-BY-SA-4.0 have been added to the available licenses to choose from. There is now also a drop-down menu underneath the title and description fields where users can select the license that they want their image to be uploaded under

• 
  "Copy title/desc from previous upload" button
• 
  Category suggestions based on title
• 
  Warning that image already exists in Commons
• 
  New licenses and license selection

• Improved UI as per Material design guidelines. The UI has received an update, and is now in line with Material design guidelines. Also, a light theme has been added, which is more suited for daytime use.

• Implemented beta testing. With over 2000 active installs, we decided that we should migrate to using a beta version for testing new releases rather than pushing directly to production, to minimize the negative impact of potential bugs on users. The beta version is open to all, and users can choose to opt in here.

• Switched to using a Wikidata query for "nearby places that need pictures". After receiving feedback from users who tried our initial "Nearby Places" prototype, we switched to querying Wikidata instead of Wiki-Needs-Pictures for generating the list of nearby places that need pictures. Wikidata has several advantages: the request is much faster and takes about 2 seconds (compared to half a minute), it takes less data to download the list, and it works for all Wikipedias regardless of language.

• 
  Original dark theme
• 
  Settings screen where theme can be chosen
• 
  New light theme

Improvements made during the Prague pre-hackathon and Vienna hackathon

Several of us came together to work on the app during the Wikimedia Hackathon Vienna (19th to 21st May) and also during the pre-hackathon in Prague (organized by Wikimedia Czech Republic from the 12th to 16th of May). Various improvements that were made during these events include:

• Added a map of nearby places that need pictures. In addition to the list of nearby places, we wanted to add a map for a few reasons: some people prefer using a map to a list, it is easier for people to plan photo trips with a map, and there is potential for light gamification in the future (similar to Pokemon Go). Every pin on the map is a Wikidata item that needs pictures; users can scroll around and browse, or they can select an item to get more information about it or directions to it. The map was implemented using the Mapbox API.

• Implemented a navigation drawer. The app was previously using an overflow menu in the action bar to access the Nearby Places feature, Settings, etc. Adding a navigation menu makes it much more easy to navigate around the app, and also looks more professional.

• Added more information in media details pane. In addition to the license, uploaded date, categories, etc., users can now view the coordinates of their uploaded image, and the date of upload.

• Fixed issues that users with memory-limited phones are having with the main screen. The previous version of the app triggered "Out of memory" crashes at the main (My Uploads) screen - this was found to be due to the app "guessing" thumbnail URLs and loading the full resolution image if the guess fails. This was fixed by retrieving proper thumbnail URLs for images from the Commons API, so full resolution images are never loaded.

• Added a logout option.

• Gave users the option of choosing how many of their previous uploads they would like to view on the main screen. Users can now choose to only view a smaller subset of their contributions in their main screen, if they want to make it easier to find a particular upload or to conserve memory and bandwidth.

• 
  Map of nearby places that need pictures
• 
  Navigation drawer
• 
  Media details pane with coordinates and upload date
• 
  Prehackathon participants in Wikimedia Czech Republic's Prague office

Publicity work

• Updated Google Play listing. The app's Google Play listing was updated with screenshots from the latest version, and the store description was modified to provide more information on what the app is used for and how to use it.

• Created a landing page for the app. The app now has a website with links for downloading the app from Google Play or F-Droid (courtesy of Tobias)

• Wikimedia blog post. We wrote a blog post for the official Wikimedia blog with an update on the new features in the app.

• Handed out flyers during the Wikimedia Hackathon. To raise awareness among Wikimedia contributors about the existence of the app, we printed flyers to hand out or put up on notice boards during the recent Wikimedia Hackathon.

• Spread word about the app among Wikimedia communities. We posted on various mailing lists, the Commons Village Pump, as well as on the social media networks of various Wikimedia communities about the new app updates.

• Created a video demonstrating the use and features of the app. A video has been made to demonstrate the usage and features of the app (courtesy of John Lubbock).

• 
  Google Play store listing
• 
  Landing page
• 
  Flyer for promoting the app during the Wikimedia Hackathon

Outcomes and impact

Outcomes

What are the results of your project?

Please discuss the outcomes of your experiments or pilot, telling us what you created or changed (organized, built, grew, etc) as a result of your project.

Number of active installs on Google Play

The number of active installs (as displayed on the Google Play Developer console) has grown from 217 active installs at the start of the grant period (20 June 2016), to 2565 active installs on 18th June 2017. This is a 1082% increase and, in my opinion, reflects the demand for such an app in the community.

Note: Unfortunately with the recent Developer console update, I cannot display the active installs from before September 2016 in this graph. However, you can view the earlier stats from the graph posted in my midpoint report.

Number of files uploaded vs files deleted per quarter

The Commons app stats tool (created by Yusuke) allows us to visualize the number of uploads made using our app, as well as the number of files uploaded via our app that were deleted, on a weekly and quarterly basis. The graph below was taken on the 20th of June 2017. The app was first reintroduced to the Google Play store as a community-maintained app from 2015Q3, and this grant ran from 2016Q3 to 2017Q2. Uploads prior to 2015Q3 were made with the WMF-maintained version of the app that was pulled from the store in 2014Q3 due to lack of resources for maintenance.

Based on the data from this tool, there were 20554 new files uploaded via the app during the grant period, and 3237 of them were deleted. This averages out to a 15.74% deletion rate across all 4 quarters. This is a significant percentage (which means there is room for improvement), but it means that 84.26% of files uploaded via the app are consistently of acceptable quality (even before the tutorial was made). This should hopefully put to rest fears that a few people have of the "selfie-pocalypse", where files that need to be deleted outnumber files that are usable.

Number of images used in Wikimedia projects

We used GLAMorous to obtain the total number of images uploaded via the Commons app that were used in Wikimedia articles from July 2016 to June 2017.

• Total image usages: 5778
• Distinct images used: 3485 (17.32% of all images of "Uploaded_with_Mobile/Android_(Jul_2016_-_Jun_2017)" category)

The five highest image usages:

Site	Images used
de.wikipedia	1344
en.wikipedia	1173
wikidata.wikipedia	448
es.wikipedia	337
zh.wikipedia	272

New code contributors on GitHub

There are 8 new contributors who have joined our developer community on GitHub during the period of the grant:

• Technical definition of "new contributors": Volunteer contributors who have made at least 5 commits that were merged into the 'master' branch of our GitHub code repository during the grant period, and who did not contribute any commits prior to the start of the grant period
• GitHub usernames of new contributors, ordered by number of commits: maskaravivek, veyndan, neslihanturan, tobias47n9e, sandarumk, addshore, Deskana, jayvdb

The graph below demonstrates the activity of our code repository on GitHub:

Progress towards stated goals

Please use the below table to:

1. List each of your original measures of success (your targets) from your project plan.
2. List the actual outcome that was achieved.
3. Explain how your outcome compares with the original target. Did you reach your targets? Why or why not?

Planned measure of success (include numeric target, if applicable)	Actual result	Explanation
All tasks completed	Completed except for #76.	I encountered significant technical difficulties with #76 and eventually found that the amount of time being spent on that task was exceeding the potential benefits, and thus decided to move on. All of the other tasks were completed, including the tutorial. There are also many additional improvements to the app that were not listed in the proposal.
50% increase in active installs between the start and end of the grant period. There were 217 active installs at the start of the grant period, so the goal was 326 active installs at the end of the grant period.	Completed many times over. Active installs at the end of the grant period (18 June 2017): 2565.	The outcome was a 1082% increase from the start of the grant period and 2239 more than the goal.
80+ pictures being uploaded daily through the app that are of acceptable quality (not reverted after 3 days)	Not completed. An average of 65.5 pictures were uploaded daily from 12 June 2017 to 18 June 2017 and not reverted by 22 June 2017	This figure was obtained by running this query on PetScan and then dividing by the number of days. There was an issue with our initial daily estimate in the proposal, where we forgot to select "only pages created during the above time window", so we overestimated what we should be aiming for. More realistically, running the same query for these dates in 2016, the number of pictures uploaded daily before the grant started was 24.5, so in my opinion 65.5 per day is a reasonable improvement
10+ new contributors who have never contributed to Commons before	Completed. There were 700+ new users whose first upload was made via mobile app	This query pulls up a list of users whose first upload was via mobile app and made on or after July 2016. An exact number of users cannot be obtained, as the list contains uploads from the iOS app as well, but considering that the iOS app was pulled from the store before the grant started, it provides a decent estimate with a large margin (700+). A more in-depth look at the files uploaded by the first 10 users listed in that query (the oldest ones, i.e. during the time when uploads via the deprecated iOS app were most likely) reveals that all 10 were uploaded via the Android app, hence the estimate appears to be sound.
Less than 10% of pictures uploaded by new users are unusable (selfies, copyvios, etc). This will be measured over a period of 2 weeks at the end of the grant.	Not completed. 11.7% of pictures that were uploaded from 5-18 June 2017 were deleted	Similar to the previous metric, we use 'deletions' as a quantitative measure of unusability - if a picture hasn't been deleted after >3 days, we consider it usable, and vice versa. New uploads from 5-18 June 2017: 862. Deleted uploads: 101. 11.7% of new uploads from 5-18 June were deleted, hence we missed our mark by a bit. We intend to continue working on improving this ratio in the proposed renewal.

Think back to your overall project goals. Do you feel you achieved your goals? Why or why not?

Goal	Achieved?	Why/why not?
Make this app more user-friendly and convenient to use for both newcomers and experienced users	Yes	Multiple improvements have been made to the user interface of the app, and we have received good feedback from users on the improvements and new features. Many bugs and crashes were also fixed.
Encourage submission of appropriately-categorized, unique and useful pictures to the Commons database	Yes	A significant number of images uploaded via the app have been used in various Wikipedia articles, and the ratio of deletions to uploaded files is relatively low
Grow the contributor base by promoting this app as an easy and non-intimidating way for newcomers to start contributing to WM and Commons	Unsure	Promoting the app to newcomers turned out to be much more difficult than I thought, as I had overestimated the likelihood of mainstream media and third party tech blogs wanting to write about it. This led to us switching tacks to focusing on publicizing the app among Wikimedia communities instead. However, given the large increase of active installs, as well as the significant number of users whose first upload was via mobile app, it's possible that this goal might have been mostly achieved regardless of the difficulties.

Global Metrics

We are trying to understand the overall outcomes of the work being funded across all grantees. In addition to the measures of success for your specific program (in above section), please use the table below to let us know how your project contributed to the "Global Metrics." We know that not all projects will have results for each type of metric, so feel free to put "0" as often as necessary.

1. Next to each metric, list the actual numerical outcome achieved through this project.
2. Where necessary, explain the context behind your outcome. For example, if you were funded for a research project which resulted in 0 new images, your explanation might be "This project focused solely on participation and articles written/improved, the goal was not to collect images."

For more information and a sample, see Global Metrics.

Metric	Achieved outcome	Explanation
1. Number of active editors involved	N/A	This is a long-term software development project for Wikimedia Commons (not Wikipedia), so I believe that the number of active editors involved is not applicable. At any rate we do not know how to measure this.
2. Number of new editors	761 new Commons contributors	This number was taken from this query, which obtains the number of users whose first contribution to Commons was via mobile app, from July 2016 onwards. Caveat: the list contains uploads from the iOS app as well, but considering that the iOS app was pulled from the store in early 2016 and in view of the sampling done (see measure of success #4), it provides a reasonable estimate.
3. Number of individuals involved	1671 individuals	The number obtained is a sum of 1653 uploaders (based on data from Uploaders in cat), 15 active code contributors (based on number of contributors in this GitHub graph with 5 or more commits), and 3 active contributors (not on GitHub - Vojtěch, Martin and John).
4. Number of new images/media added to Wikimedia articles/pages	5778 (total) or 3485 (distinct)	These stats are obtained from GLAMorous, where we query the "Uploaded_with_Mobile/Android_(Jul_2016_-_Jun_2017)" category. 17.32% of all images of that category were used in various Wikimedia articles.
5. Number of articles added or improved on Wikimedia projects	N/A	Does not apply to our project, I believe metric #4 is a better measure for us
6. Absolute value of bytes added to or deleted from Wikimedia projects	N/A	We do not think there is any value in measuring the megabytes of pictures uploaded via the app, since bigger pictures are not always better

Learning question

Did your work increase the motivation of contributors, and how do you know?

Yes, I believe it did. We have received a lot of encouraging feedback from users, both online (in Google Play reviews, our forums, on GitHub and mailing lists, etc) and in person (at the Vienna Hackathon) who tell us that the app has made it easier and more fun for them to upload photos to Commons. I have reproduced some of our publicly-posted Google Play reviews below:

• "Must have app for Wikipedians. The handy app for photographers who wants to contribute to wikimedia ptojects. I use this when I visit nearby for uploading photo to commons instantly." - Nahid Hossain
• "An easy to use App. It's cool how it lets you contribute to Wikipedia via creative Commons licensed photos that you take this has been the easiest way for me to add to Wikipedia." - CJ Flynn
• "Uploading images made easy. I frequently work on Wikimedia Foundation projects including wikicommons for uploading imagea for the articles." - Asad Ali Jogi
• "Great. The app is now more stable and faster. Upload works well and choosing categories is easy. Searching for nearby missing images is a lot of fun." - Tobias Schönberg
• "The app is easy to upload photography on commons." - Janak Bhatta

Indicators of impact

Do you see any indication that your project has had impact towards Wikimedia's strategic priorities? We've provided 3 options below for the strategic priorities that IEG projects are mostly likely to impact. Select one or more that you think are relevant and share any measures of success you have that point to this impact. You might also consider any other kinds of impact you had not anticipated when you planned this project.

Option A: How did you increase participation in one or more Wikimedia projects?

Yes, I think this project has increased participation by encouraging people to upload photos to Wikimedia Commons. This can be seen via the number of usable (undeleted) photos uploaded through the app, the number of new contributors whose first upload was via our mobile app, and the number of photos uploaded through the app that were used in Wikipedia articles.

An interesting secondary (unplanned) impact is increased participation of volunteer developers. Some of the volunteers who joined our developer community during the period of the grant were relatively new to Wikimedia projects. I think having an active project with a relatively small codebase (such as this) appears to be beneficial in attracting new volunteer developers. Also of note, our team at the Wikimedia Hackathon and Prague pre-hackathon was one of the most diverse, both in terms of nationalities (7 different nationalities), as well as gender (3 women).

Project resources

Please provide links to all public, online documents and other artifacts that you created during the course of this project. Examples include: meeting notes, participant lists, photos or graphics uploaded to Wikimedia Commons, template messages sent to participants, wiki pages, social media (Facebook groups, Twitter accounts), datasets, surveys, questionnaires, code repositories... If possible, include a brief summary with each link.

• GitHub page - we use GitHub as our code repository and for collaboration
• Website - landing page for the app
• Google Play store listing - where Android users can download the app
• Stats tool - for visualizing uploads and deletes
• Pre-hackathon in Prague - an event where collaborators met and did some work on the app prior to the Vienna Hackathon
• My personal blog - contains regular updates on the app
• Wikimedia blog post - blog post that I wrote for the official Wikimedia blog about new features in the app
• Mailing list posts - 4 September 2016, 15 December 2016, June 9 2017

Learning

The best thing about trying something new is that you learn from it. We want to follow in your footsteps and learn along with you, and we want to know that you took enough risks in your project to have learned something really interesting! Think about what recommendations you have for others who may follow in your footsteps, and use the below sections to describe what worked and what didn’t.

What worked well

What did you try that was successful and you'd recommend others do? To help spread successful strategies so that they can be of use to others in the movement, rather than writing lots of text here, we'd like you to share your finding in the form of a link to a learning pattern.

• Grants:Learning_patterns/Keeping_documentation_of_discussions_with_team
• Grants:Learning_patterns/Number_of_images_or_media_used_in_Wikimedia_projects

What didn’t work

What did you try that you learned didn't work? What would you think about doing differently in the future? Please list these as short bullet points.

• Publicizing the app to non-Wikimedians. Initially, the main aim of publicity was to increase non-Wikimedian awareness of this app. After writing the Wikimedia blog post, I attempted my first round of press notifications, where I contacted tech blogs that had written about the app in 2013 to let them know that it was back up on the Play Store and receiving regular updates. I contacted many blogs, and only one of them responded. This led to having to reconsider our publicity plans. After some discussion, I realized that I might have been jumping the gun in trying to go directly to publicizing the app to non-Wikimedians. For one thing, given that there are many Wikimedians who are unaware that the app exists, it would make sense to tailor the first publicity attempt to them. For another, it is likely that if we wanted to effectively publicize the app to non-Wikimedians through general media, it would require a lot more time and manpower than was budgeted for the task, especially given our "unofficial" status (we cannot claim to be part of WMF or endorsed by WMF). Thus, we switched tacks to focusing on publicizing the app among Wikimedia communities instead.

• Time management. This isn't exactly something that "didn't work", but I struggled significantly with it. When I applied for the grant, I only estimated how much time I would spend actually developing the features and carrying out the tasks mentioned in the grant, without taking into consideration all the time that would need to be spent on basic maintenance (e.g. fixing new bugs) and project/product management. The latter took much more time than I anticipated, especially given that we grew a lot over the last year, both in terms of userbase and project contributors. This is certainly a "good problem to have", but as a result the grant took much longer than expected to finish. In the future I would likely budget some time each week for these tasks.

Next steps and opportunities

Are there opportunities for future growth of this project, or new areas you have uncovered in the course of this grant that could be fruitful for more exploration (either by yourself, or others)? What ideas or suggestions do you have for future projects based on the work you’ve completed? Please list these as short bullet points.

There are multiple areas in which the app could be improved, based on feedback received from users and the community. I intend to apply for a grant renewal focusing on these areas:

• Enhancements of the 'Nearby places that need pictures' feature. This is our most popular feature by far. We plan on allowing users to upload images directly for a particular item on the Nearby list or map, and providing suggestions for title, description and categories based on the information that is associated with that Wikidata item.
• Enabling two factor authentication and other quality-of-life improvements.
• UI overhaul that allows multiple uploads from the gallery, a more prominent placement for the upload button, and allowing users to view photos that other people have uploaded near them
• Making the app more interactive and improving user education
• Various technical improvements that would increase the stability of our app and make it easier for future code contributors to get on board

I will be drafting the renewal proposal soon.

Part 2: The Grant

Finances

Actual spending

Please copy and paste the completed table from your project finances page. Check that you’ve listed the actual expenditures compared with what was originally planned. If there are differences between the planned and actual use of funds, please use the column provided to explain them.

Expense	Approved amount	Actual funds spent	Difference
Developer salary for Phase 1	1200 USD	1200 USD	0
Developer salary for Phase 2	6000 USD (after budget change request)	6000 USD	0
Developer salary for Phase 3	1200 USD	1200 USD	0
Total	8400 USD	8400 USD	0

Remaining funds

Do you have any unspent funds from the grant?

Please answer yes or no. If yes, list the amount you did not use and explain why.

• No

If you have unspent funds, they must be returned to WMF. Please see the instructions for returning unspent funds and indicate here if this is still in progress, or if this is already completed:

• N/A

Documentation

Did you send documentation of all expenses paid with grant funds to grantsadminwikimedia.org, according to the guidelines here?

Please answer yes or no. If no, include an explanation.

• No. I was informed that receipts were not required for my grant, as the expense was salary for my own (developer) work.

Confirmation of project status

Did you comply with the requirements specified by WMF in the grant agreement?

Please answer yes or no.

• Yes

Is your project completed?

Please answer yes or no.

• Yes

Grantee reflection

We’d love to hear any thoughts you have on what this project has meant to you, or how the experience of being an IEGrantee has gone overall. Is there something that surprised you, or that you particularly enjoyed, or that you’ll do differently going forward as a result of the IEG experience? Please share it here!

I was amazed to see how the app userbase grew over the duration of the grant, and how many people were using the new features that were implemented. This work has been extremely meaningful to me, especially being able to see all the images being submitted via the app, and the Wikipedia articles that benefited from those images.

The community has been incredibly supportive towards this project, with many people helping in any way they can - code contributions, spreading word about the app, testing, providing feedback and input, reporting bugs and crashes, helping with designs/art, patrolling the images submitted via the app, creating a website or analytical tools for the app. From the bottom of my heart - thank you to all of you. The app would not be what it is today without you.

I also really enjoyed meeting up with the other contributors at the Wikimedia Hackathon 2017, and at the prehackathon that was organized by Wikimedia Czech Republic. This app started off with just Nicolas and myself, but now I feel like we are truly a team.