Page MenuHomePhabricator

Decide on future of Android Wiktionary app
Closed, DeclinedPublic

Description

I would like to open a conversation to revitalize the Wiktionary app.

I can't install it on my phone anymore but the store page is still online. According to the Android store there are over 100K installs (https://play.google.com/store/apps/details?id=org.wiktionary). That is a pretty amazing number for an app that got its last update in 2013. I am not sure how trustworthy that number is but it could indicate that there is interest in reading the Wiktionary content on mobile phones.

At least 1 app seems to be trying to scam users with a similar name. Hopefully users are not installing it with the feeling that it comes from the foundation.

Another important simultaneous development is that we have a growing corpus of structured Lexeme data that could be used in the app. This lexeme show the great potential for viewing and editing of dictionary data: https://www.wikidata.org/wiki/Lexeme:L3054

Users of the app could have all of these things:

  • Lists of conjugations
  • Links to other dictionary databases
  • Listen to pronunciations
  • See pictures where the word is used (useful for e.g. public signs, where a word could often appear)
  • Browse back and forth in etymology

And users could be invited to:

  • Add data
  • Add their own recording of how they pronounce a word-form
  • Add a photo of where the word is used (e.g. an information sign)

Because Wiktionary is lemma bases and not lexeme based, there is not 1:1 mapping between Wiktionary pages and Wikidata lexemes. If the app gets picked up I think starting with the lexemes is better because they are more structured and offer a lot of quick wins. The Wiktionary communities could be convinced to switch to 1-lexeme-per-page model in order to integrate better with Wikidata and the app.

To conclude I want to say that I am in favor of revitalizing the app sooner rather than later. It seems like with structured lexeme data the time is ripe. Of course the Wikipedia app could also benefit from lexeme integration, but having an app focused on dictionary content seems like a fabulous idea.

Event Timeline

I would like to open a conversation to revitalize the Wiktionary app.

I'm not sure a task is the best place for that. :)

@Aklapper I was unsure where to post this. Do you have a suggestion. Definitely could be a talk or discussion at the next Hackathon.

jberkel renamed this task from Decide on future of Wiktionary app to Decide on future of Android Wiktionary app.Sep 29 2018, 1:37 PM

@Tobias1984 there's the mobile mailinglist, although there's not much activity (mostly release announcements).

For Android you could also use Kiwix. But Kiwix is a general purpose viewer and not a dictionary, so I find it of limited use. I started to work on a dictionary app (motî) which uses Wiktionary data, but it's iOS only.

I personally would like to see discussion about a new (official?) cross-platform Wiktionary app.

The Wiktionary communities could be convinced to switch to 1-lexeme-per-page model in order to integrate better with Wikidata and the app.

I am very curious about this statement. How do you think you can convinced 100+ communities to change the way data are structured?

About an app, I support this idea. I recently discovered a a free Android app by Livio based on Wiktionary content, downloaded 10 million times. I think it's remarkable. It works very well for French and Wiktionary is mentioned as a source in every pages.

A good dictionary app need to provide definitions on a glance and more information for people looking for more. Content like examples, pictures, quotations and thesauri are present on Wiktionaries but not in Wikidata and will not be. Quotations are very important to get a meaning and a proof of existence for rare words.

I am very curious about this statement. How do you think you can convinced 100+ communities to change the way data are structured?

I agree with this sentiment. Wikidata's development strategy seems to be very top-down, creating facts and then expecting other projects to adapt to it, instead of listening to what the existing communities need.

Heya, I just wanted to chime in here with some context on how this idea might move forward. I see the wishlist item was declined and I recently sent an email to another volunteer with related information about the status of the Wiktionary app.

The app has not been updated in a long time (over 6 years). The foundation does not have the resources to build such an app. Keeping it in the Android store in its current state has a maintenance cost and sets an exception that it is supported (bugs, feature requests, compatibility with new devices, etc), which has not been true for many years.. The best bet is a group of volunteers get together to create a new app. There are ways of doing this.

The old app runs on an ancient version of a "middleware" called PhoneGap that is now owned by Adobe. It is not recommended to use this old code, but in theory anyone could compile it and release it. The source code is on Github.

The second idea is to look for open-source frameworks (or start from scratch) and build a new app. It may be possible to get Grant funding through the WMF to help with this. The more work you have to show before applying for a grant is recommended. I don't know all the details, but there are folks on Meta who can help with this.

Our technical community is also open to new projects and opportunities to get involved. We have folks who know our APIs and MediaWiki that can help. Whomever wants to take on this work is not doing it alone.

I think setting up a project page somewhere (Meta/MediaWiki) and getting the word out through mailing lists, IRC, Facebook, whatever) is a great place to start. See if there is interest, discuss what would be a minimally-viable product and get cracking'!

There is precedent for this. The Commons Android app is a great example of a community-lead mobile app that has received some grant support. It's the official app by default and highly regarded.

I do agree with Andre, this phab task is far from the idea place to discuss this. I'll make you all a deal. :) Get a project page up on Meta and if you need help learning more about the resources I mentioned, mention me on the talk page. I'll point you in the right direction.

The second idea is to look for open-source frameworks (or start from scratch) and build a new app.

May be pushing the following forward might lead to a better alternative.

T165495: Generic MediaWiki Android app

I'm boldly declining this per last comments and in favor of T165495: Generic MediaWiki Android app.

Enervation subscribed.

Since T165495: Generic MediaWiki Android app has been declined, perhaps we should restart the discussion of the Wiktionary mobile app. T165495 was declined with the comment "Since the app is now wholly dependent on Wikipedia-specific RestBase services, we can no longer officially support a generic mediawiki flavor." How much work would it be to adapt the official Wikipedia app for Wiktionary? https://blog.go2tech.de/2016/02/how-i-customized-the-wikipedia-app-to-match-the-needs-of-droidwiki-de/ makes it sound like all it takes to make a Wiktionary app is some simple text replacement, but that was in 2016. If the Wikipedia API has since substantially diverged from the Wiktionary API, perhaps it would be a lot harder now.

Anyone is free to restart the discussion and that's no good reason to reopen this ticket... Declining to reflect reality.