Common Voice Android app: legal question about the logo and the name "Common Voice"

Hi, I’ve developed an unofficial Common Voice app for Android (https://github.com/Sav22999/common-voice-android/).
My question is: can I use the name “Common Voice” inside the app or it’s a TM name?
Of course I specify every time the app is not developed by Mozilla and that the Common Voice project is “by Mozilla”, and to use the app you need to accept Mozilla’s Terms. In the app I don’t use anymore the logo, so people cannot wrong.

The app is just to improve the Common Voice project; it is open-source and absolutely free, so I don’t earn anything and I don’t want to pay for a legal issue.

Sincerely, Sav.

4 Likes

Hi @Sav22999, thanks for reaching out and for the interest in making Common Voice more android focused. We actually do not officially permit the use of the Common Voice name outside of the official platform. You are encouraged to mention on your adaptation that it is based on the Common Voice project and our platform.

We want to take this opportunity to let you know that the ability to contribute to the Common Voice dataset via an unofficial app will be removed in the near term. The ability to do this was never an intention of the project. The Common Voice dataset is required to be collected on the official platform to ensure terms of service, privacy and data security are maintained. Regrettably any data that has been collected via your app will not be used for the official Common Voice dataset. You are encouraged to fork the project and create a platform for the purpose of collecting a separate dataset.

We appreciate you reaching out to clarify branding. Please let us know if you have any other questions! cc @nukeador @Christos_Bacharakis1

Ok, I’ve decided because of “recordings and validations collected via the app will not be used for dataset”, to discontinue the app.

The main goal of the app was just to improve the project and give possibilty to users to use a native app on their device(s).

I’ll announce this decision in the next update of the app, the last, where I will invite users to use website instead of the app and I’ll link also to this discussion.

This are the app statistics (no worries! personal data are not been collected and users could also disable those anonymous statistics):
https://www.saveriomorelli.com/commonvoice/statistics/

This are the Google Play Store users:

I just want to share some messages users sent to me (I’ve asked to they if I could use these screenshots and they authorised me):






I just want to say that many users used the app and they found out the project also thanks to F-Droid and Google Play Store, so I want to advise strongly to insert an official app in your plans.

Sincerely, Sav.

9 Likes

This is very sad, a lot of users are using this app because let them to save bandwidth or use low-budget phone (also with the new mode offline), so basically the CV team is closing the project to a specific segment of users (that is against the Mozilla manifesto).
I can understand that is difficult to follow the quality or the review because after all is a different client developed differently from the official roadmap, also that the team behind CV is busy in a lot of things.

As I said I understand the issue, that was the same at the time of the Telegram bot to contribute to CV. I think that if the code is open source, publicly available and doesn’t have any server to share the information (as this one) except to CV official server, maybe require just a legal overview (already has the ToS as I can remember).

So probably this mean remove that user’s opportunity, maybe for months, until an official strategy is developed and other months to be implemented but the official roadmap is already defined, so I guess that the plan is for 2021/2022.

Looking at the people that share (it is opt-in) their installation of the app (https://www.saveriomorelli.com/commonvoice/statistics/) we are talking of an average of 60 people daily that is the half of the people today at 9AM was contributing to CV (120~ looking at the stats box in the home page).

The plan was to help CV community members to have another way to use the portal that don’t have 500/40X errors or with other issues to contribute to the original CV dataset. Saverio share the statistics of the app publicly, so I think that analysis of the merits of the app it is important to consider.
The Android audience is different from the iOS one, also from kind of users and age. So I think that the Android app until the website: is a PWA supported, more performance friendly and similar features I don’t think that the project should be blocked.

I am saying in this way, honestly my pure opinion, because I met the rest of the CV team in All Hands and know some people of the team since a lot. So I know that the team is working hard in a lot of things/side and the priorities are others but also let more people contributing is very important.

10 Likes

How sad!!!

The website is really difficult to use on my device, the app was so much better.

Please consider forking the app and turning them into an official (perhaps still volunteer-driven?) Mozilla app that can even be distributed via F-Droid.

And all my contributions are now lost? How doubly sad!

Especially as I had to agree to the TOS anyway, for making an account.

There wasn’t a way to not agree to them, so they are valid, just as those I made from the website on my desktop computer.

Also, the offline capability the app added recently is huge! An incredible amount of thought and consideration has gone into the app, and Mozilla would do well on taking the improvements to user experience into consideration for their own site (or, app).

Thank you, Saverio, for all your work. It was fun to contribute in this easy way.

6 Likes

Thanks for your feedback everyone, we will take take all of this into consideration as we move forward. We hear you on offline contribution capabilities especially for low-bandwidth users, and it’s something top of mind for us as well. One of our technical concerns here is also API security best practices, but rest assured that we will not turn off API without officially letting you all know first.

I also wanted to clarify - any data that is already part of our canonical database is part of the dataset. We do not strip out any clips from the dataset release except those that don’t have valid audio data.

1 Like

When ?

It sounds generic, sorry for my bad english.

2 Likes

I’m really sorry to hear about this. The website on mobile is pretty unusable to put it frankly (at least on my low-end phone) and having the app installed helped me contribute more to the project, both in terms of recordings and validations. The app also has the advantage of offline recordings, which can be really useful if you have limited mobile bandwidth.
I also speculate that recordings gotten through the app are vastly more accurate, since you are forced to listen to your own recordings before you are allowed to upload them, unlike on the website.

I honestly don’t understand why Mozilla would take a similar stance, and I’m hoping they will change their minds. Anyhow, I would still like to read more about this decision that is apparently against the “open web” and “open knowledge” paradigm Mozilla advocates for.

7 Likes

Hi!

I’m really confused about your decision to block the unofficial Android App. I think it is clear to everyone that the official website is a bit buggy and doesn’t really work well on low-end devices yet. To give another example: On my Android-7-mobile with Firefox mobile I only see the white loading screen for minutes when I open voice.mozilla.org :sweat_smile:. This circumstance is of course annoying for me, but it is absolutely understandable when the developers are currently busy with other important issues.

But against this background I am really surprised that the unofficial Android App is going to be blocked. I think we should be thankful to @Sav22999 that he has put effort in the development of an Android App. This allows me and many others (see the app’s statistics) to contribute to Common Voice in a much easier way, while reducing the pressure on the official development team of the website.

I understand Mozilla’s position so you have legal concerns regarding Sav’s app:

It’s not really obvious to me what you’re concerned about. The android app asks the user to agree to Mozilla’s terms of use directly in the beginning. Furthermore the whole source code is available so that you could check through it. I am aware that this does mean work. However, Mozilla has committed itself to its manifesto, which explicitly demands decentralized and worldwide participation (Principle 6). (Of course this means work.)

In the event that the Android app should indeed be blocked due to legal concerns, then it would be fair to us and @Sav22999 if you would explain them to us (principle 8: ”Transparent community-based processes“). I am sure, in the case that you have reasonable legal concerns, we would show understanding and continue to be proud of being a part of Mozilla’s mission :slight_smile: :slight_smile: .

8 Likes

Hi Megan,

that’s a sad decision. In my modest opinion, that I’m just a supporter, it’s a wrong path, sorry.

In part I understand your points, but in, Common Voice project in general, I see too much rejections to contributors (developers in this case) that try to support the community and the beautiful CV project.

I decided to comment here not just because the specific Saverio’s app story (to which goes my absolute solidarity, I’m sure about his positive and proactive intent and realization), but because some minor facts I experienced myself last fall.

By example in past I raised in this forum some perplexities regarding:

gender bias (statistics are impressive with just 19% of women contribution)
age bias (people with age < 19 years old not allowed)
bad words censorship

I didn’t see pretty any feedback in community forum ( a part the remind of mandatory respect of USA regulations). Right, but isn’t CV an international project in the general/long vision terms?

Are the topics above mentioned minor points for the CV project ? I don’t think so if, since many years, we talk so much about data feed (in this case spoken languages) bias in current machine learning tech.

Another minor fact: I commented an issue on Github, just agreeing on the request of ready to use DeepSpeech foreign (italian in my case) language models. My request has considered out of scope and has been defined ‘insinuating’ and the issue has been closed abruptly. Is this the way to reply to such a request?

Don’t you want to spread DeeSpeech creating developers engagement? Is really your suggested path to use DeepSpeech with mandatory dataset retraining?
Is that your meaning of engagement?

Again, see what happened with the third party developer telegram bot (Daniele reminded) and the lack of Common Voice official API (another mystery/shutdown).

Sincerely I’m really perplexed and I do not understand Mozilla CV long term strategy. It seems to me very authoritative and “top down” approach. Something one could expect from Amazon, not from Mozilla.

Don’t Common Voice want to be an opensource / opendata REAL alternative to big players wallet gardens?

I don’t understand these continue “closures”.
Very sad.
Giorgio

4 Likes

So this invalidates what Megan wrote about the invalidation of all the recordings gathered with the app?

The point is that they are deprecating also the iOS app as written in the roadmap, so they don’t want to split their internal resources to follow different projects but just focus on the portal that needs a lot of work.

I suggest to keep this thread about the Android app topic and move on this discussion in the right one like Common Voice Quarter 1 2020 Community Update about the roadmap as example.

I think that we need to track all the issues that the website has more clearly like with https://webcompat.com/ so there will be bugs that are not just comments, so they need to check those.

I have just one thing to add to my other comment: the CV team tried the Android app? Because I have doubts about it like as example of the T&C part that was added an year ago I think.

1 Like

I hope it has been a misundersting.

I renew my question about: “we will take take all of this into consideration as we move forward”

When ? I fear that silence means “sine die”.

2 Likes

“So this invalidates what Megan wrote about the invalidation of all the recordings gathered with the app?”

This is also what I’m wondering. Is Mozilla going to delete contributions made through this app retroactively (before the moment the notice was posted above, 4 days ago) or not?

1 Like

I apologize for my perhaps slightly-irritated tone, but this is really enraging me. I really wish Mozilla would reconsider this, and put the reversal very high up on the importance scale.

I only learned of the CV project when I saw this app on F-Droid. It seemed great to me – people often say I have a “weird accent” in several languages, speech-to-text hates me, etc. Being able to contribute and improve a free voice database (and benefit myself and others down the line) would be pleasing to me.

The mobile site didn’t work well and I’m not often at a laptop/desktop, so the lovely, FOSS app, made out of love for the purpose, and which is clearly labelled as unofficial all over, seemed like a no-brainer.

And you want to have him discontinue it? It really seems like a case of looking the gift horse in the mouth. And cutting off a stream of contributors is just a shame.

If you insist on discontinuing it, I and other contributors would absolutely require a major overhaul of the mobile site to work on devices that cost less than $700… or else contributions would be hardly possible! :frowning:

8 Likes

This decision made me, and not only me, very sad. I found the application more immediate and more user friendly than officially site, especially for smartphones.
I think the app might help you to rapidly spread your project among people.

Please, reconsider your decision.

Best regards.

Alessio

6 Likes

It’s sad to read the work of @Sav22999 would be discontinued since he made a very good job.

Most of my contributions (more than 2500) are coming from the mobile app.
The app improved a lot to a point where it’s a real pleasure to contribute, without a lot of small frustrations we could have with the website.

I definitely hope Mozilla will consider again this question. There are other solutions than just blocking every app to use the API. Especially when an app like the one @Sav22999 made exists and helps a lot of users to contribute to this huge project.

8 Likes

This news is incredibly disappointing for me to hear. I’ll outline a few reasons that I liked Sav’s unofficial Android app better than the official website. I really hope that Mozilla can allow the app, since I don’t think I’ll contribute any more without it.

  1. The interface flow is better. After pressing the button to stop recording, the app allows me to immediately listen again, and then I can choose to either send the recording or re-record the clip. This is so much better than the website, especially since I often stumble over my words while speaking, and have to re-record. On the website, recordings are done in batches of five, so I have to remember that I misspoke, then try to figure out which of the 5 I misspoke in, and then use the buttons to re-record that clip only, and then finally send the 5. It’s so much more annoying and it adds a lot of friction to me trying to speak.

  2. It’s more friendly for my disabilities. I have autism and I find animations on the screen very distracting; they stop me focusing on the task that I’m trying to perform. For example, this made Discord actually unusable for me when they made messages change colour when hovered. On the common voice website, everything is animated with no option to turn it off, which steals my focus away from trying to read and speak the sentences. In the unofficial app, the interface is much plainer, and I have the option to turn off animations entirely, which I use and really appreciate.

  3. Having a home screen icon encourages me to take the time to contribute when I’m not doing anything else. For example, if I’m at the bus stop waiting for the bus, I’ll often swipe through the home screens of my phone, waiting for an app icon to catch my attention and keep me occupied until the bus comes. Seeing the common voice home screen icon reminds me that it exists. I open the app, I press one button, and I can instantly start speaking. The website is a website, it’s not a home screen icon, it can’t even be installed as a PWA. Not remembering that common voice exists when I have time to kill completely stops me from contributing.

Could these problems be solved on the website? Perhaps. They could design and implement a new interface flow that makes more sense, they could add an option to disable animations, and they could make it installable as a PWA. But will they? I have no idea. It would take quite some time, and I bet the common voice team has other things to focus on. This is why I like the app — it has these features right now. Due to these reasons, if the app doesn’t work, then I simply won’t use common voice.

6 Likes

BLEAH!
I am disgusted with these decisions!
Sav22999 was wrong not to ask for permission first, but you from Mozilla are wrong in everything else.
The work he has done is excellent, the app he created has open source code, his system works better than your site, he has no ulterior motives and you answer that his work is not valid?
I am not surprised that you are losing users with Firefox and that you will soon lose developers with these behaviors …

3 Likes

+1 to the issue of the recording flow, I’d never thought about it in these terms but I think it’s a great summary of the advantage the app has over the website.

1 Like

Hi all, and thank you so much for sharing your thoughts and concerns.

I want to begin by quoting Jenny’s comment from a few days ago.

The reasons behind considering blocking access to 3rd party applications have only to do with security and complying with legal requirements around data collection.

We hear you, the Common Voice mobile experience is not the best, and we want to fix it as much as you do.

We will spend the last two weeks in July, planning our work for the second half of the year. During that time, we will review community feedback, all open issues, our planning so far, and we will prioritize our work for Half 2 2020 accordingly.

After that, I will personally come back to you with news about the future of the CV Android/mobile experience, how we will handle data from 3rd party apps, and with an updated and more detailed Product Community Roadmap.

4 Likes