Discussions

News: GWT 1.5 release candidate posted

  1. GWT 1.5 release candidate posted (43 messages)

    The 1.5 release candidate of the Google Web Toolkit (GWT) is available. This release adds features for compatibility with Java 5 annotations and templates instead of GWT's earlier annotation-like annotations for type safety. Also included: themes and more widgets. The normal bug fixes and enhancements announcements apply. If you don't know what GWT is: GWT takes fairly normal Java code and translates parts of it into JavaScript, providing a very shallow ramp to rich internet applications. It also provides an RPC library (as one would expect, with AJAX), a rich widget library, and an active ecosystem providing more widgets and capabilities. The move to Java 5 syntax is a huge step up for GWT: congratulations, guys.

    Threaded Messages (43)

  2. Pattern expressions on [ Go to top ]

    sorely needed. Having to move classes around purely because there is no mechanism to include only certain files is painful if there are too many to refactor.
  3. cool, now I can take advantage of the latest gwt-incubator and gwt-dnd projects. I'm hoping the Google Summer-of-love turns out a mature set of flexy, extjsy visual components. Remember, Haight is not a Google value.
  4. Last week I had a session with my managers to sell them the idea of GWT over flex I showed them a sample application and how we can benefit from writing java code in both server and client etc. But the look and feel of Flex is just something that GWT can't compete with now. Yes we can customize the look and feel and frameworks can be written to support all the nice effects that flex have, but having a nice UI with nice widgets, out of the box, can be good selling point to managers. I can't compete with flex advocates when they show their demo (Accordion, Transitions, Cool effects, etc, etc). But the truth is that Google will never invest into GWT for all of those nice features, after all it is a free beer. So I am thinking maybe google should take the adobe approach with GWT. Free engine and paid service for Enterprise messaging (which GWT doesn't have at the moment) and nice UI designer then they will have funding for it. Regards Behrang Javaherian http://www.beyondng.com
  5. Last week I had a session with my managers to sell them the idea of GWT over flex I showed them a sample application and how we can benefit from writing java code in both server and client etc. But the look and feel of Flex is just something that GWT can't compete with now. Yes we can customize the look and feel and frameworks can be written to support all the nice effects that flex have, but having a nice UI with nice widgets, out of the box, can be good selling point to managers. I can't compete with flex advocates when they show their demo (Accordion, Transitions, Cool effects, etc, etc).
    But the truth is that Google will never invest into GWT for all of those nice features, after all it is a free beer. So I am thinking maybe google should take the adobe approach with GWT. Free engine and paid service for Enterprise messaging (which GWT doesn't have at the moment) and nice UI designer then they will have funding for it.

    Regards
    Behrang Javaherian
    http://www.beyondng.com
    You need to look for third-party GWT.widgets. Ext-GWT and others have those types of effects.
  6. You need to look for third-party GWT.widgets. Ext-GWT and others have those types of effects.
    I did but the flex look and feel is superior. Flex people can drag and drop component on the screen, attach a validator to them and press the run button and get a great user interface that would take few days to build even with Ext-GWT, I can't see why GWT can't get there. Although I am fan of open source and using a lots of open source projects, I think a commercial support and funding will help GWT to be able to compete with flex in productivity and user experience. Cheers Behrang Javaherian http://www.beyondng.com
  7. You need to look for third-party GWT.widgets. Ext-GWT and others have those types of effects.


    I did but the flex look and feel is superior. Flex people can drag and drop component on the screen, attach a validator to them and press the run button and get a great user interface that would take few days to build even with Ext-GWT, I can't see why GWT can't get there. Although I am fan of open source and using a lots of open source projects, I think a commercial support and funding will help GWT to be able to compete with flex in productivity and user experience.

    Cheers
    Behrang Javaherian
    http://www.beyondng.com
    Look at GWT-Designer. That's what I used. For third-party controls, I just dropped a GWT standard control on the form and did pretty much what you did. Then, after the code was generated, I refactored, them swapped in the third-party control and pow! Now, I have java code but was able to put the screen together with a wysiwyg app. I had a pretty complex app, tabs, dialogs, scrollable regions, menus, the works in about a week. And this was all in java code.
  8. 'You need to look for third-party GWT.widgets. Ext-GWT and others have those types of effects' After what they did to Ext-GWT(It started with LGPL and turned to GPL) licensing can you dare to look at any other third party library? It is safe to rely on GWT default widgets,gwt-incubator and build on top of it. Going for third party libraries just for the look is a big risk.
  9. After what they did to Ext-GWT(It started with LGPL and turned to GPL) licensing can you dare to look at any other third party library?
    +1 Change of ExtJs licensing was showstopper for us as well and we decided to go for FLEX instead. Once ExtJs is not free it just can't compete with FLEX which is much superior solution for RIA applications.
  10. Flex is great but I can't force my clients to upgrade to the latest version of Flash. Extjs is a nice library but I really see gwt-incubator as an up-and-comer. A little more skinning, dnd, and animation and I think you are there. Check out gwt-fx as an animation library. Pretty slick. Now that GWT has made the infrastructure jump to 1.5, we should see it start to visually close the gap with Flex.
  11. After what they did to Ext-GWT(It started with LGPL and turned to GPL) licensing can you dare to look at any other third party library?

    +1

    Change of ExtJs licensing was showstopper for us as well and we decided to go for FLEX instead. Once ExtJs is not free it just can't compete with FLEX which is much superior solution for RIA applications.
    And hope Flex doesn't change their licenses. You aren't immune.
  12. MyGWT is a big problem[ Go to top ]

    After what they did to Ext-GWT(It started with LGPL and turned to GPL) licensing can you dare to look at any other third party library?

    +1

    Change of ExtJs licensing was showstopper for us as well and we decided to go for FLEX instead. Once ExtJs is not free it just can't compete with FLEX which is much superior solution for RIA applications.
    NO! The problem with MyGWT is that it was NOT actually LGPL and it is not now actually GPL. Parts of it are proprietary and they have this weird sort of preconditioned LGPL/GPL. You have to meet those conditions before you receive the LGPG/GPL. This seems to be designed to inhibit forking (I think it fails to achieve that, but it's certainly annoying, and somebody would have to stare their lawyers down to make a fork work). Normally, if you somebody tries to move to a more restrictive license (eg LGPL->GPL) then it doesn't work because the community would simply fork. With MyGWT you have the problem that the license was misunderstood at best and misrepresented at worst. Google needs to help fix this, in my view.
  13. Re: MyGWT is a big problem[ Go to top ]

    It certainly was a bum move. Something popular enough that did this could have had a chilling effect on OS adoption back in the day.
  14. Re: MyGWT is a big problem[ Go to top ]

    Lest we forget that GWT and Java have actually become more open and recently too. Google code hosting has some very specific licenses that you can pick from. MyGWT and extjs definitely suckered a lot of people since nobody is going to pay a bunch of lawyers to defend a code fork.
  15. MyGWT and extjs definitely suckered a lot of people.
    Follow the malpractices of Ext and what people like John Resig, Dion Almer and Graeme Rocher have to say. http://extisevil.blogspot.com/
  16. License Risk[ Go to top ]

    'You need to look for third-party GWT.widgets. Ext-GWT and others have those types of effects'

    After what they did to Ext-GWT(It started with LGPL and turned to GPL) licensing can you dare to look at any other third party library? It is safe to rely on GWT default widgets,gwt-incubator and build on top of it. Going for third party libraries just for the look is a big risk.
    There are ways to avoid this. The simplest way is do not participate in a project that insists on recpiprocal copyright assignments and make sure it has an active community of committers, if they do not own all the copyright they cannot re-license without all community contributors agreeing. Mark http://blog.athico.com/ Drools Blog
  17. 'You need to look for third-party GWT.widgets. Ext-GWT and others have those types of effects'

    After what they did to Ext-GWT(It started with LGPL and turned to GPL) licensing can you dare to look at any other third party library? It is safe to rely on GWT default widgets,gwt-incubator and build on top of it. Going for third party libraries just for the look is a big risk.
    That's a problem for any third-party anything. It sucks,but you can use the pre-GPL stuff or purchase it. Really, how is that any different from just buying something? What if Google starts charging for GWT? No difference.
  18. 'You need to look for third-party GWT.widgets. Ext-GWT and others have those types of effects'

    After what they did to Ext-GWT(It started with LGPL and turned to GPL) licensing can you dare to look at any other third party library? It is safe to rely on GWT default widgets,gwt-incubator and build on top of it. Going for third party libraries just for the look is a big risk.


    That's a problem for any third-party anything. It sucks,but you can use the pre-GPL stuff or purchase it.

    Really, how is that any different from just buying something? What if Google starts charging for GWT? No difference.
    If I'm the author of an Open Source project and I deeply depend on another Open Source project - which is what the Open Source eco-system is about. And that project stops being Open Source or changes it's license in a way that I can't consume it, then it leaves me and my entire community without a paddle. The emphasis here is on the Open Source eco-system and how the myriad of projects creates a body of works great than the sum of all it's parts. If many people using the project thought they where using it under the terms of the LGPL, and then it turns out that this wasn't quite true - due to poor communication and continual changing of licenses, which did start of as LGPL. If the assets, the CSS, are in fact proprietary (not communicated on the web license page, but only in a nested file in the archive) a fork won't be easy and as CSS is effectively code, without a clean room re-implementation (expensive) it's not possible for the open source projects which depend on this project to continue. Mark http://blog.drools.org The Drools BLog
  19. We are all knowingly, putting ourselves at risk using ANY OS project. We've all been using for OS for so long, that some may have forgotten those initial conversations with our bosses when we were trying to convince them to use Struts, Ant, Hibernate, Spring, whatever. MyGWT is that nightmare scenario, but there is nothing to stop any of them from switching if they so choose. Isn't that why we decouple or cry "Standards!" or whatever? Fortunately, most of the projects don't do this, but it can happen to any of them. To me, saying don't use third-party widgets, but still using any other open sources on the hopes that they don't change the license is naive. Dropped support, changed licenses, bad code, those are all the dangers, be it MyGWT, Hibernate, or Spring.
  20. Dropped support by 3rd party vendors[ Go to top ]

    In my past experience, I had a lot more problems with commercial parties dropping a product (or at least stop active development/support) or firms being bought (forced into migration/upgrade path if you want to keep using)/ going of business.
  21. Interesting idea... BTW, have you checked jQuery JS and its huge plugin library? Some plugins look amazing: Mac like menus: http://www.wizzud.com/jqDock/ FlexGrid: http://webplicity.net/flexigrid/
  22. Last week I had a session with my managers to sell them the idea of GWT over flex I showed them a sample application and how we can benefit from writing java code in both server and client etc. But the look and feel of Flex is just something that GWT can't compete with now. Yes we can customize the look and feel and frameworks can be written to support all the nice effects that flex have, but having a nice UI with nice widgets, out of the box, can be good selling point to managers. I can't compete with flex advocates when they show their demo (Accordion, Transitions, Cool effects, etc, etc).
    But the truth is that Google will never invest into GWT for all of those nice features, after all it is a free beer. So I am thinking maybe google should take the adobe approach with GWT. Free engine and paid service for Enterprise messaging (which GWT doesn't have at the moment) and nice UI designer then they will have funding for it.

    Regards
    Behrang Javaherian
    http://www.beyondng.com
    I think at the end of the day we're dealing with some inherent limitations of today's stock browsers. I guess SVG would help that, but...
  23. I think at the end of the day we're dealing with some inherent limitations of today's stock browsers. I guess SVG would help that, but...
    Check out the GWT Canvas component. It works with GWT 1.5 already, and in all current browsers; it uses the "canvas" HTML 5 tag in Firefox 1.5+ and VML in IE 6+ (I don't know what it uses in Opera and Safari). IMO, with current advances in web technology like the HTML 5 W3C standard and Firefox 3.0 (both approaching final releases), Flash/Flex may well become obsolete or at least unnecessary five years from now. Another great tool from Google and with a GWT API is Google Gears, which provides, among other things, a JDBC-like SQL API backed by SQLite, a very capable and fast embedded SQL database. Rogério
  24. IMO, with current advances in web technology like the HTML 5 W3C standard and Firefox 3.0 (both approaching final releases), Flash/Flex may well become obsolete or at least unnecessary five years from now.
    This assumes Adobe will freeze flex/flash development for 5 years and wait for competing solutions to take its market share. Questionable assumption I'd say. That said, I'm all for advances and diversity in web technologies (any sort of technologies). Natural selection will let the best survive.
  25. GWT Widgets[ Go to top ]

    What I miss in GWT is a framework on top of the toolkit. Something to enable easy forms, pagination, validation, data binding, before/after/around advice for rpc calls, etc. As for the widgets, GWT is getting there: check out the rolodex for instance: http://code.google.com/p/gwt-rolodex/ I started developing a google-calendar like app to learn GWT, the source looks horrible (contrary to intuition, it helps ALOT if you know your way around with DOM/HTML when using GWT; I knew nothing xD) but I was getting there. http://code.google.com/p/gwt-scheduler/ I never understood why there isn't a google calendar-like open source proj in GWT/Flex/ExtJS, or any other library/toolkit/framework.
  26. Re: GWT Widgets[ Go to top ]

    Well, it sits on top of DOM and HTML, so it would make sense that you would have to be versed in that stuff to created certain types of components. However, I'm not sure if you need a framework on top of GWT. Most of the stuff you mentioned is pretty easy to do out of the box.
  27. Re: GWT Widgets[ Go to top ]

    I was naive back then. I've been avoiding javascript and css for too long :) But I think GWT would benefit from a framework. At first glance, you can do those things out of the box. But you'll eventually end up repeating lots of code if you not design carefully. Let's say I want to show a loading icon everytime an rpc is being executed. The simplest way would be subclassing the AsynCallback and enforcing all my rpcs to use the subclass so I can make some sort of before/after advice. If you think this AFTER your app is developed, this would be much harder to accomplish. Besides, data binding is hard because GWT does not support reflection, you end up writing JSNI to do that for you or use generators. I just want a framework to avoid all the plumbing code, for me GWT right now feels like how servlets were in the beginning, you can do most of things but you end up writing alot of glue code.
  28. Framework on top of GWT[ Go to top ]

    Go take a look at IT Mill Toolkit - we have forms, dynamically loading tables (pagination is so web 1.0 :)), validation, data-bindings, ... as well as many nice widgets on top of GWT. Framework has been in production use since year 2001 and is evolving rapidly. Everything is released with Apache 2.0 license. http://itmill.com/
  29. Leopard Support[ Go to top ]

    Its great!
  30. Features required[ Go to top ]

    Before using GWT, we still need two feature to somehow present: 1. Search engine/web crawler support. At the moment there is no right way to create a GWT application that can be indexed by any search engine. 2. On-demand application loading. If you have a large application, it will be loaded once upfront, which is good sometimes, but I'd like to see the option that only portion of pages are loaded. Yes I know, I can separate my app into modules, but if you have ever tried that, you know it is pain... Any plans on these to come?
  31. Re: Features required[ Go to top ]

    Before using GWT, we still need two feature to somehow present:

    1. Search engine/web crawler support. At the moment there is no right way to create a GWT application that can be indexed by any search engine.

    2. On-demand application loading. If you have a large application, it will be loaded once upfront, which is good sometimes, but I'd like to see the option that only portion of pages are loaded. Yes I know, I can separate my app into modules, but if you have ever tried that, you know it is pain...

    Any plans on these to come?
    It's easy to do both of these. Jeff Dwyer, in "Pro Web 2.0 Application Development with GWT," discusses SEO and GWT, as well as a lot of other issues - including OpenID integration, application size, etc. It's a really good book, I've found, although it relies quite heavily on Spring-MVC and Spring Security. (That's not a *real* complaint, although I would have liked to have seen less Spring, just because of the tie-in. But let's be real: Spring's not THAT MUCH of a tie-in, being only slightly more of a tie-in than Java EE.)
  32. Good book indeed![ Go to top ]

    Jeff Dwyer, in "Pro Web 2.0 Application Development with GWT," discusses SEO and GWT, as well as a lot of other issues - including OpenID integration, application size, etc. It's a really good book, I've found, although it relies quite heavily on Spring-MVC and Spring Security. (That's not a *real* complaint, although I would have liked to have seen less Spring, just because of the tie-in. But let's be real: Spring's not THAT MUCH of a tie-in, being only slightly more of a tie-in than Java EE.)
    I've checked that book and especially the source codes. It is pretty good indeed, with some really good idea how to mix the different worlds, I can only recommend it. At some points I think he is not really using the real capabilities of Spring MVC (like annotations), at least our team has done it better (will be open sourced sooner or later). But anyway, anyone seriously working with GWT should check that book!
  33. Ajax technology is obsolete[ Go to top ]

    Since Flex, Silverlight and maybe javaFX matured, Ajax based solutions became obsolete. The three mentioned technologies are far superior to the inferior browser based application model. Ajax was nice and a good patch for a shortcomming of the browser. But to invest in such technologies that utilize Ajax is investing your money in the past without revenue in the future.
  34. all technologies you mention are properiaty
  35. JavaFX is open source[ Go to top ]

    all technologies you mention are properiaty
    JavaFX is open source: http://www.sun.com/software/javafx/index.jsp ---- Openness and Compatibility Matter Like the rest of the Java platform, JavaFX Script will be available under the GPL license. ----
  36. Re: JavaFX is open source[ Go to top ]

    For a lot of companies GPL will be an insuperable burden. The point is that html, css and javascript dont have licenses anyway, but maybe you are right and JavaFX will take over everything
  37. That is not the issue Iam talking about. That is a legal issue and not a technical issue. Iam only discussing the technical issue here. But to enlighten you a bit. Flex is opensource, only the visual IDE is closed source. Furthermore actionscript is close or almost ecmascript. It uses XML a layout language. It uses CSS for design.
  38. The way I see it, you have to make a architectural desision. You must make this desision over the 3 basic layers. In my point of view html+ajax is a inferior solution to Flex. One of my own arguments is that html+ajax is still subject to a specific browser in its implementation. Flex is not. One design truly looks the same on many browsers and desktops. Furthermore Flex AIR will give you many offline capabilities, that html+ajax will never be able to do. Also it is infinitly easier to design a screen/application layout in Flex opposite to html. I know it is always difficult to let go off technologies that you have used alot, are comfortable etc.., but you have to let go of old tech or be left behind. :):)
  39. It is hard to make the argument that AJAX is dead or even that Flex is the successor. What is dead is page-level calls. People will migrate to REST, JSON, GWT-RPC, and yes BlazeDS services that feed browser side clients. What flavor you choose doesn't matter as always with web frameworks. Services will make the client side technology even more throw-away so your switching costs will drop.
  40. Re: Ajax technology is obsolete[ Go to top ]

    The way I see it, you have to make a architectural desision. [...] Also it is infinitly easier to design a screen/application layout in Flex opposite to html.
    I know it is always difficult to let go off technologies that you have used alot, are comfortable etc.., but you have to let go of old tech or be left behind. :):)
    Agreed on the architectural decision, however as you tend to advertise flex, I'd like to ask a question. What is the best way to implement e.g. wikipedia in Flex? Features required: bookmarkable urls, quick link followup, search-engine compatibility, easy edit... I know ways to do this in GWT, it will be fast, testable, easy setup, even the drag-and-drop editor is there and so on, and I'm seeking the benefit of Flex over it, but I can see little at the moment...
  41. Flex and javaFX is open source[ Go to top ]

    Flex is also open source and JavaFX. ONLY Silverlight from Microsoft which is not open source by any angle
  42. Since Flex, Silverlight and maybe javaFX matured, Ajax based solutions became obsolete. The three mentioned technologies are far superior to the inferior browser based application model. Ajax was nice and a good patch for a shortcomming of the browser.

    But to invest in such technologies that utilize Ajax is investing your money in the past without revenue in the future.
    I think you're not quite comparing Apples to Apples here, AJAX as a technique is just one piece of the puzzle. I would argue that even without AJAX you can have RIA without using Flex, Silverlight, JavaFX, etc.
  43. re: Ajax technology is obsolete[ Go to top ]

    Since Flex, Silverlight and maybe javaFX matured, Ajax based solutions became obsolete. The three mentioned technologies are far superior to the inferior browser based application model. Ajax was nice and a good patch for a shortcomming of the browser.

    But to invest in such technologies that utilize Ajax is investing your money in the past without revenue in the future.
    Do you see that one size fits them all? There might be only one from Flex, Silverlight or JavaFX, but there is still place for HTML+possibly Ajax solutions. Less is more sometimes :)
  44. I wonder if anyone has considered using gwt's nice java api on top of ordinary html pages? I mean, apparently everyone is busy creating the whole page with gwt, but what I'm after is to create ordinary web pages with whatever js library or jsf or whatever, and then in cases where I need to write js code to do fancy things on the client side use gwt's nice java api and its java-to-js compiler to write java code instead of loads of js. Or in other words this: http://groups.google.com/group/Google-Web-Toolkit/browse_thread/thread/7afc8c738191bd32/f6e64fbd9672041e?lnk=gst&q=abrahamian#f6e64fbd9672041e Has anyone considered such a usage? It's particularly useful in creating web "sites" with some ajaxyness here and there, rather than big web "apps". Ara.