Results 1 to 7 of 7

Thread: GWT 3.0, Sencha GXT and the future of the widget eco system ....

  1. #1
    Ext GWT Premium Member
    Join Date
    Aug 2010
    Germany, Solingen

    Default GWT 3.0, Sencha GXT and the future of the widget eco system ....

    A few month ago the GWT Steering Communittee annouced some informations about the future of GWT and the things we can expect in GWT 3.0.

    It looks like some parts of GWT will be removed in version 3.0:

    * UiBinder (I will not miss it ... ;-))
    * RCP (RestyGWT is a good alternativ)
    * Request factory
    * Generators (most of the things can be done with annotation processors)
    * the widget eco system (that's hard)

    Most of the things GWT 3.0 will no longer support I will not miss. But loosing the widget eco system will be hard and replacing it with JsInterop and writing JavaScript widgets does not seams to me as an alternative. Of course JsInterop is cool, but writing web frontend in Java is really cool.

    I am writing enterprise applications. Of course the application have to look nice, but it is the business logic that counts. I have written some widget libs and know how much work it is. That's one of the reasons I prefer using a commercial library like Sencha GXT (to be honest, it is the only one). In most cases setting up a custom theme is enough to create a nice looking individuell app.

    Thinking about GWT 3.0 and widgets, it might be possible to create a open source project that contains the necessary widget classes so it will be possible to write widgets in Java. This would enable developers to work with widgets as they did it before. But as long as there are no detail infos about the next GWT version (3.0) it is hard to decide weather it is a good idea or not. (In my opinion it must be possible to write such a module which will work with GWT 2.x & 3.0) Also, I would like to see a Sencha GXT version which supports GWT 3.0. And I would like to see widget base classes (Open Source), so that everyone can create widgets as they did it before (GWT < 3.0). I think a lot of developers will like this module.

    Also I would like to know what will happen to Sencha GXT. Will there be a version that supports GWT 3.0.

    It would be nice if Sencha can provide some informations related to this questions.

  2. #2


    Thanks for the kinds words about GXT. We still don't have any more info on GWT 3 than what has been publicly shared at GWT.create in January and the subsequent meetup in May, but we definitely plan to support GWT 3 one way or another. When Google speaks of "removing widgets," I think their first concern is to reduce the support burden of all the individual Widgets currently in GWT, which few Google projects actually use. However, even if they remove the Widget and Cell base classes altogether, it should still be possible to write new base classes using the new DOM impl in GWT 3 as you point out, and this would indeed make a nice open source project. GXT widgets already extend our own base class (Component) which extends Widget, so it is theoretically straightforward to rebase our widgets to use a new DOM implementation. We are still evaluating alternatives at this point, but it is always our goal to make it as easy as possible for GXT customers to keep up with the latest versions of GWT.Given my understanding of the motivation for the proposed removal of Widget in GWT 3, I would hope that Google would be open to leaving some Widget base classes in GWT itself, but if not, I think a separate project is a great idea. If I were you, though, I would probably wait to see what GWT 3 actually brings. We're still waiting for 2.8, after all.

  3. #3
    Ext GWT Premium Member
    Join Date
    Aug 2010
    Germany, Solingen


    At the moment it is frustrating that there are no more informations about the upcoming GWT 3 release.

    I agree that it should be possible to add the widget functionality to GWT. In my opinion it would be better if this classes are part of GWT. But if not there should be a valuable solution that it is possible to use GWT 3 in the same way as GWT 2.x regarding the widget system in the future.

    Pointing out that Sencha GXT will support GWT 3 is a really good news! It should be good to promote this information in the GWT community.

  4. #4
    Sencha User
    Join Date
    May 2015


    Thanks for the response drfibonacci, I was at the GWT.create conference in Munich and I sensed some unease around the 'widgets are the past' announcements.

    It makes sense, the other technology we were shown looked to be much closer to the metal and more... 'native' in the browser... I just fear that software companies like mine who have developed these huge suites of applications in GWT are being left a bit bewildered.

    All eyes are increasingly on you guys to see what we do I guess.

    The future is looking increasingly like restful micro services with a JS front-end, unfortunately

  5. #5
    Sencha Sr Product Manager
    Join Date
    Jan 2012
    Arlington, WA


    Just a note from my point of view. Our executive team is very sensitive in supporting the massive apps folks have built. In other words, we won't abandon our valued customers. The Jackel compiler won't make us get rid of the current way of designing apps.

    We've have large to small customers that have huge applications that have taken many folks to develop and and maintain. Given that we have so many customers using the current GWT widget, we won't abandon it. But on the other hand I suspect our internal research and development eventually might come up with strategies to implement both widget and jackel systems together for mixed old and new development together.

    Our customers are important and we make tools that help them get the job done.

    Thanks for sharing your opinion!

  6. #6


    Quote Originally Posted by branflake2267 View Post
    Just a note from my point of view. Our executive team is very sensitive in supporting the massive apps folks have built. In other words, we won't abandon our valued customers. The Jackel compiler won't make us get rid of the current way of designing apps.
    I tried to get updates to this topic but haven't found much. We have a ton of code dependent on GWT and GXT widgets and are faced with the unfortunate lack of any apparent future with GWT 3.0 or otherwise. Some scary facts was able to find:

    1. Sencha no longer has any presence in the GWT steering committee. Colin Alworth used to be there, but is no longer in either the steering comittee or Sencha. Justin Hickman seems to have taken his place but is no longer with, again, either Sencha or the steering committee as of about Feb 2017.
    2. GWT 3.0 can't begin yet before J2CL is out and it isn't. It may be close, but indications are that it will be severely limited. It is questionable how much / whether any code generation, deferred binding or related functionality will be there and how much of it would be even possible. Sencha not having any presence in the steering committee does not help the situation. There are notes indicating that a lot of JSNI code that GXT or other widgets may have internally (I don't know, we don't use it much) needs to be converted into JSInterop, starting with GWT widgets, then up to GXT.
    3. A lot of people do not understand the benefit of abstraction widgets provide and seem to think that it is always better to be as close to the platform as possible. Yes, if you are writing a game requiring super fast animations that is correct. However, if you are writing a solution that is trying to abstract the presentation layer from the enhancements to the presentation itself across versions and do more with this, then that is not the case.

    Just being able to create some similar widgets but require code changes isn't really an option for all of us. We, in particular, are building a foundation on top of which unknown developers build add-ons and sell them to customers. Customers may separately upgrade the foundation and these add-ons. New versions of the foundation are supposed to be compatible with add-ons built for some prior releases. You can think of those almost as apps on top of OS. GWT and GXT are exposed to all (with appropriate licencing). This means that we can't just change everything internally and be happy because we will have broken compatibility with those add-ons.

    I started a related discussion at the GWT contributors forum earlier - there is a lot of discussion there, some of it related to widgets, UI binder and code gen, other to RPC (also important to me).

    Business proposition of GWT 3.0 - what is it good for vs. other solutions out there?!

  7. #7


    No responses?

Similar Threads

  1. Future of Sencha Touch???
    By patrik62 in forum Sencha Touch 2.x: Q&A
    Replies: 15
    Last Post: 18 Jun 2015, 11:50 PM
  2. Future of Sencha Touch
    By Lukappa in forum Ext JS 6.x Q&A
    Replies: 1
    Last Post: 11 Apr 2015, 7:52 AM
  3. With Sencha Desktop Package discontinued - does SA have a future?
    By tangix in forum Sencha Architect 3.x: Q&A
    Replies: 2
    Last Post: 30 Dec 2013, 7:41 AM
  4. Future Roadmap of Sencha Touch
    By findajit in forum Sencha Touch 1.x: Discussion
    Replies: 0
    Last Post: 26 Jun 2011, 4:38 AM
  5. Future direction of Sencha/Ext JS
    By jonhobbs in forum Sencha Touch 1.x: Discussion
    Replies: 5
    Last Post: 18 Jul 2010, 7:38 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts