Flutter_facebook_login: ------ DEAD PROJECT ------

[feinstein]

I have no idea what happened to the original maintainer of this library @roughike, I hope he's in good health, but since there have been no bug fixes, upgrades, replies or PR accepted or denied, I think it's safe to assume that this library is dead and should be avoided from now on.

[Pebsie]

Such a shame as this functionality is essential for a lot of apps. I'm not sure Flutter is a viable choice without it.

[feinstein]

We can always fork it and update it ourselves, but I know it's not ideal. I hope the Firebase Auth team makes a replacement..... And hope flutter don't go the javascript way, with lots of abandoned highly-used projects.

[feinstein]

Good news, I could reach @roughike on Twitter, he appears to be fine, just not following the repo closely.

[cpboyd]

@feinstein IDK if @FirebaseExtended is the ideal place for this plugin, as they don't seem to have anything specifically targeting Facebook logins.

Ultimately, I think abandoned projects are inevitable if the project only has a single maintainer.

I definitely think there should potentially be another maintainer added to this repo in order to keep it up-to-date.

There's 11 pending PRs (granted some of them supersede others, since a few are related to SDK updates) so it's not like there's any lack of people willing to work on this project.

[feinstein]

As I said to some googlers, I think all major libraries should have someone from Google inside them as a fail-safe.

[cpboyd]

@feinstein I don't think it necessarily has to be someone from Google, but long-term libraries that need continual attention (like this one, since Facebook keeps updating their SDK to meet new iOS and Android requirements) should probably have more than one maintainer.

There are already other projects that ended up with a fork on Dart's Pub.dev.

Edit: I'll point to @react-native-community as an example of how React Native deals with important third-party libraries. There's 42 people in charge of those repos.

Perhaps something similar is necessary for Flutter?

[feinstein]

Yeah, sure there needs to have someone else... But how do you assure this will happen? Who finds volunteers? Who tries to convince maintainers that they need someone else into the project? I think Google is a good answer.

[cpboyd]

@feinstein On the other issue, someone mentioned @fluttercommunity:
https://github.com/fluttercommunity/community

Since you've already been in contact with @roughike, it might be worth seeing if he wants to transfer the repo there if he doesn't have the time to keep this plugin up-to-date.

[roughike]

I'm alive and kicking, and have no health issues. Thanks for the concern!

We're using this plugin in Reflectly in production and have about 4 million users. It works for us. That being said, I'm aware there are some bugs, the SDKs could use a little update and the pull request & issue queues need some maintenance. I guess I'm just busy doing some other stuff ¯\_(ツ)_/¯

If someone has the context, could you describe what "dead project" means here? Unmaintained issue & PR queues, last release being ~6 months ago, or the plugin simply not working at all for you?

[feinstein]

Unmaintained issue & PR queues, last release being ~6 months ago, or the plugin simply not working at all for you?

Well, it's actually the sum of all of it. If no-one was having troubles, opening meaningful issues or submitting PRs, it would be fine for a project to not be updated.... But once all of it comes together and there's no response, it appears to be a dead project.

I am glad you are OK, but for resilience sake it would be interesting to include someone else into the project, just as a plan B person.

[noln]

Yeah, sure there needs to have someone else... But how do you assure this will happen? Who finds volunteers? Who tries to convince maintainers that they need someone else into the project? I think Google is a good answer.

I'm not sure that Google has an interest in ensuring that a competitor's login method is supported. In this case I'd say it would be advantageous for Facebook to have some developers involved, but then that'd clash with their own cross-platform app framework technology interests.

A better option might be something like Opencollective, Bountysource, or even Patreon to fund development time from open-source contributors. There are plenty of freelance devs and consultants who'd love to help with open-source, but time is money.

[feinstein]

Facebook would be ideal, but Firebase is a Google service that provides login through Facebook credentials, so yes, Google can have an actual interest.

Any doubts look at all the people from the Flutter team commenting on my post on dead projects over the main Flutter repository where I point to this repo's case. It's all about the Flutter ecosystem look strong and reliable for everyone.

[noln]

Facebook would be ideal, but Firebase is a Google service that provides login through Facebook credentials, so yes, Google can have an actual interest.

Any doubts look at all the people from the Flutter team commenting on my post on dead projects over the main Flutter repository where I point to this repo's case. It's all about the Flutter ecosystem look strong and reliable for everyone.

Perhaps..until they deprecate it and dump it in the gutter, classic Google-style. Sorry, seen too many Google projects come and go to have much faith in this one! Hopeful, but cautious.

[cpboyd]

@noln I'd imagine that since they're using Flutter on Google Home and for some of their mobile applications (and the looming Fuschia OS), Flutter's probably gonna be around for a while.

There's the same risk with anything, aside from HTML and Javascript.

[eleomilagrosa]

@roughike
Well this will be a dead project for sure.

174

it started since Aug 30, 2019 when App Store announce that they don't accept UIWebView anymore.
so this library will not be usable anymore because you'll get an error anyway from App Store.

[roughike]

I just merged a couple bugfix PRs and released 3.0.0 to Pub which updates the SDKs to latest versions. I'm also going through the issue tracker and closing obsolete / otherwise closeable issues.

Not sure how this is an actionable issue. You can continue the discussion here, but I'm closing this one for now, unless you have concrete action points for me. Have a great day!

[feinstein]

Thanks @roughike, glad to see this advancing :)