Is GWTP Still Maintained?

[yadavharsh]

Hi,

We are working on a large scale application using GWT/GWTP. However, we are not very sure if we should continue with it, given the fact that the GWTP repository looks stale with no recent commits or releases.

Is GWTP still maintained or internally used by ArcBees?

Just saw that the acrbees.com website no longer uses GWT. This is a little concerning, and before proceeding with GWTP we wanna be sure if we should be switching to other framework, if its no longer going to be maintained/open sourced/sponsored by ArcBees.

Thanks!

[christiangoudreau]

We've reached a stage where it is stable with our current clients. The lack of activity is mostly due on two things: we're waiting to see how GWT 3.0 will evolve and external contributions that has slowed over the last few years.

As of our consulting activities, we want to be leaders in the AI field. Implementing solutions in AI will still need tools like GWTP though :)

We tried a couple years ago to use paid support plan as a tool to speed up our own contributions, but it hasn't been successful and to this day, what kept us alive (and GWTP) is our clients on the consulting side.

Don't hesitate to participate, contribute or reach me: christian.goudreau@arcbees.com if you think we can help!

[yadavharsh]

Thanks for the reply Christian.

So in a nutshell there would not be any internal development (or a roadmap) for GWTP from within ArcBees (until GWT 3.0 is out which might be after 2-3 years). And there is not going to be a GWTP 2.0 (or not anytime soon) with Dagger as mentioned in: #783

I understand that as a company you have to decide in whats best fits. However, a very clear announcement or a roadmap from ArcBees would help a lot of people in the open source community using GWTP right now, about whether they should be continue using it or make switch. Sad if it would be latter, as its a great framework and simplifies many things.

[christiangoudreau]

That decision is still under discussions internally. What I can tell you is that the investment we had from other businesses 4-5 years ago is almost inexistant now. While we know that big businesses are still investing and using GWT/GWTP, we're not part of that equation.

You are right to think that we need to consider what is the best for our future as a business. One thing for sure, our commitment to open source will always remain and while we think about our future, we would welcome anyone that want to help.

[mduesbury]

Hi...

Any possibility of an update on GWTP, especially with regards to supporting GWT 3.0 in the future? I'm aware of quite a few large applications built with GWTP, so I would imagine that in time there will be quite some demand for a version of GWTP that supports GWT 3.0.

But some work could happen before that though to remove generators for example and replace with annotation based processors.

Is there still a roadmap/pipeline for this stuff?

[christiangoudreau]

We will consider it, it was one of our strong commitment two years ago. But we don't have any idea when that will happen (the GWT 3.0 release)

[aeromac]

It's a shame because GWTP was very well maintained and very well documented - too good to fade away

[christiangoudreau]

I wouldn't say fade away... it has been stable since a couple of years. Help is welcome once GWT 3.0 is out and before!

In the meanwhile, it is a question that we've been asking ourselves. Truth is that we didn't had any new customers since at least 3 years and a half using GWT or GWTP, which has forced us to use other technologies.

With Typescript, react and angular, a lot of us has come to the conclusion that most of the reasons that made us choose GWT in 2008 and led us to build GWTP in the first place can now be answered with modern javascript. We've seen a lot of our old GWTP users move and use React or Angular as well.

I'm sad to say this, but this project, while stable, is indeed being replaced by other stable and equivalent technologies outside the GWT world.

[ibaca]

No surprise, anyway this is a zero bug project, so IMO anyone is safe to use this project for years, it has been that way the last 3 or more years, and it is a pretty productive, complete front-end library!

This looks like an incredible enterprise project, not too modern, stable, but perfect for business. Sad your clients, instead of seen the value of this super stable solution for client apps, has gone to libraries that born and die in a period that this library does not even need to have a single change to still being a complete and productive solution. I think that Java, GWT, and GWTP are all super slow changing technologies that solve actuals problems, that it unique fault is not to born each 3 years. I think people value libraries by the number of features added per-year, so hehe if this is how you choose, then for sure Java, GWT and GWTP won't be in your list. Libraries MUST be valued by its features (period). Sound a bit absurd, but yep, it's the unique reason to under-valued these super stable technologies. I suppose that all the people that actually value stable solutions still work in C 😛.

OTOH, IMO GWT is all about sharing java code with the JVM, if you are only creating a client-side app, using GWT is... probably not the best option.

[rcbeuker]

What about your "Reactivity" project ibaca? Maybe ArcBees could support this in some way to keep this great library alive...

[aeromac]

I've looked into using React and Angular etc and, IMO, GWT is so much better. The Widgets system has aged - it would be good if GWTP Views didn't depend on them - instead, allowing custom solutions. I'm not sure what would be required - but changes like this, IMO, would keep GWTP relevant as we move towards GWT3 and away from Widgets.

[rcbeuker]

@aeromac that's what @ibaca is trying to achieve (gwtp (thus gwt) with new Java-React add-ons). At this moment all my project are build around gwtp so obviously I'm very interested in a gwtp future ...