AFAIK we had to move from GWTEXT for licensing issues; conan on the #drools channel on IRC can give you more detail (as the requirement preceeded me joining the team). We have started a GITHUB repository https://github.com/ballroom/ballroom, to capture a rich (hopefully) set of UI components to be re-used within JBoss (well, current members include Drools and JBoss AS). Please feel free to contribute
I now understand the problem of license for commercial use. I think that is right to use pure GWT. I want to propose you the following idea:
· To Build Widgets's extensions on the pure GWT
· To take advantage of the reuse of the new generated widgets.
· To work collaboratively and extensively on these widgets, to exploit the reuse of these, in different projects.
· Generation of a powerful library of rich widgets, with their respective API.
The idea is to take advantage of the visual enrichment that these widgets are having, as the community evolves and improves them. In this way we might have the visual components centralized and reused.
In case of the project Guvnor, we might take the widgets that were injected into this project and they would be published in a project “Powerful Widgets GWT “for improvement and evolution of them. And this way, all projects that need these widgets, they could use and inject this powerful library of rich widgets, such as for the project "Tasks or Processes Console ".
Marcelo Martini (Compañía OSDE)
Perfect. I am going to comment to Mariano de Maio, that he is working with the FormBuilder, in order to he shares all the visual components that developed in your project.
And we could unify all the visual components (GWT JBOSS Visual Components) of all projects.
Firstly we are advised by Red Hat legal not to use software by the EXTJS organisation. After it's bait and switch and then litigious action on any one that didn't comply it really isn't a company you want to risk dong business with. In short EXTJS promoted the belief their software was under LGPL, then switched to GPL and told everyone that actually it wasn't LGPL after all. They threatened anyone legally who did not comply with their warped understanding of the LGPL and GPL too. They did not understand that you cannot change the LGPL license and still call it LGPL, and the nonsense of js code meaning all server side code is GPL too.
In short We in the Drools&jBPM communtiy wouldn't touch them with a barge pole. Guvnor is now 100% off EXTJS. We are working on a version of Oryx that will also remove EXTJS.
And no matter how much EXTJS people will try and tell you. An ASL project cannot use a GPL project and remain ASL.
Anyway onto the technical arguments. GWT gets stronger each release and more pleasing on the eye. I think if you look at the latest Guvnor releases and what Michael has done on the decision tables that with minimal work it can look nice. By being pure GWT we get something much easier to maintain too, as each GWT release will "just work" and you get all the improvements. Pure GWT applications are much easier to debug too. And finally pure GWT applications will use less memory, have less memory leaks and run faster.
There are 3rd part widgets that natively extend GWT that we can use and we can always develop our own if need be.