Meh.

I've built several small projects in Flutter. Google doesn't Dogfood it properly, and just cut a significant number of staff working on it.

I want Flutter to succeed, but I don't have too much faith it long term. That said, if I had to hack out a quick app right now for a client I'd probably use Flutter. Development is fast and easy. You can just write code as if you're writing JS and learn a bit of extra syntax.

React Native has it's own issues, but I have more faith in Meta to keep funding it.

It turns out, Google didn't actually cut a significant number of staff working on it:

> No change in team size; some DevOps roles moving to new locations.

From the Product Manager of Flutter and Dart.

https://twitter.com/MiSvTh/status/1785767966815985893

If you outsource people your still cutting the team.

I just don't have too much faith in Google here.

What a strange article.

Obviously the question that really needs answered is “does Flutter have a future?”

Given Google is notorious for killing popular products, the fact they just fired a bunch of people dedicated to make this product work and the rather niche nature of Dart and Flutter… all signs point to it being becoming an open source project.

Can it survive as an OSS project?

In my experience Google’s OSS projects have a high bar and have a less kafkaesque leadership than Google products. Now, one issue is that all tech that’s derivative of Google, ie Bazel, Go or k8s, tends to have “blind spots” where some parts are ignored because it’s not a need internally. The prime example would be Go’s utter garbage dependency management prior to go mod, which came very late. This wasn’t much of a priority for Google because they had their monorepo. (Nowadays it’s all dandy though). Conversely, what oss language writes sanitizers, race detectors, coverage checkers, migration tooling etc? Only the big guys do, and that can be quite nice for the rest of us.

Anyway, Flutter! Yes technically it’s an “open source project” but really it’s halfway to product/platform, no? That’s probably a bad sign (not Googles fault but run-of-the-mill oss maintainers don’t typically excel at product and design decision making). Dart? Also probably a bad sign for obvious reasons. And if Google abandons it, I imagine there’s a lot of infra (for cross platform testing etc) that goes away, and that’s assuming Google gives flutter away to the community instead of staffing it with a skeleton crew until it fizzles out. (Maybe I’m missing something in how it’s governed though, let me know)

Yeah, personally I don’t think flutter can survive without the Google breathing apparatus. But it’s not impossible. We really need and deserve something like flutter. But is there enough value over Tauri/Electron etc where people can reuse their frontend stack sanely? I think it depends on how well the competition catches up with mobile support, which is the current pain point.

Disclaimer: ex-google, somewhat recent contributor to (and user of) tauri.

Tauri has a big community. Flutter has a big community. Do you believe Tauri can survive without a direct corporate backing, while Flutter can't?

It's not a rhetoric question. I'm in need of a native application toolkit, and I gravitate towards Tauri for technical reasons. However, I'm not sure if the Tauri project has enough cash flow to sustain a durable alternative to Flutter, Qt, Avalonia, JavaFX, Compose Multiplatform, and likes.

JavaFX is a pretty dead project. Especially on mobile where it has zero users. It actually proves parents point, without Google flutter will be too big. Yes, there will be companies who will pick it up and compile it. That's hard, but doable. The problem is that there's a ton of moving parts. Testing alone is a huge task.

Some companies might pick it but then they would only test the use cases they need or a personal fork. It would also sink without the Google brand to back it.

Saw a tweet a few years back about how in China there was a lot of interest in cross-platform solutions. Both Alibaba and Tencent use Flutter in some capacity. It’s possible they might fork it and continue using it. Unfortunately, between that might being an internal-only project, and the seemingly siloed separation between Chinese development communities and the rest of the world, that might not help Flutter’s status, should Google end its patronage.

https://web.archive.org/web/20211218043000/https://twitter.c...

JavaFX still has some life into it for desktop projects, and a company that lives out of it, people pay them to keep JavaFX going, including mobile.

https://gluonhq.com/

Now, I agree if Gluon ever goes away then JavaFX is certainly done.

Oracle still maintains JavaFX along with Gluon and others. New features and bug fixes are delivered on each release.

Java and JavaFX are being ported to iOS and Android as Project Mobile under the official OpenJDK umbrella.

JavaFX fresh builds are also seen on jdk.java.net now.

Yes, although that was quite recent, as Oracle was trying to breath new life into OpenFX, after it was taken outside of the JDK, keeping Swing the best option for those that don't want to deal with OpenFX distribution of native dependencies.

As far as I am aware, Project Mobile doesn't have much uptake, outside companies that are already working with Gluon.

Indeed, Johan Vos of Gluon is the leader of the Project Mobile.

Swing remains a decent GUI toolkit, but it still requires native dependencies for stuff like OS file dialogs and other platform APIs, audio/video codecs, hardware-accelerated graphics.

Swing ships with the JDK, so whatever native dependencies are required, application developers don't need to care about them, unless they are making use of jlink, and to this day many still prefer to push to some JRE being installed, than making use of jlink and jpackage.

Which is why alternative JDK distributions still offer JRE variants, while officially from Oracle's side one should either use plain JDK or jlink/jpackage.

It is somehow ironic that Java, the pioneering "cross-platform" solution, has virtually no decent contender in this race.

Similar for Python.

There's Codename One, it's much better than anything available for Python etc. but since it's an independent OSS project from a small company it sadly never got the traction of flutter/react native.

Late reply but for the record…

> Do you believe Tauri can survive without a direct corporate backing, while Flutter can't?

Currently, I think tauri can survive yes. But I know much deeper the tauri community than flutter, where I’m just speculating as a bystander.

In fact, Tauri losing money would be much less bad than losing 1-3 of the core workhorses that are volunteers primarily afaik. This in fact is the biggest risk, imo. If any of them stop working it's gonna be really hard to recover, because they’re so autonomous and make sensible decisions.

But mostly flutter seems like a much more ambitious project, because they define UI and rely on Dart. Tauri and electron piggyback on browsers which imo is strategically and technically the only reasonable way to do it long term. They can almost exclusive focus on cross platform issues (which btw is an enormous undertaking on its own).

That's kind of my concern as well. Hoping it does survive.

The current project manager of the flutter team has said that their developer headcount, roadmap, etc. hasn’t changed and only some devops positions have been relocated.

Such a simple question, yet complex answer