GWT 1.4 released - more widgets, faster startup, other enhancements

Discussions

News: GWT 1.4 released - more widgets, faster startup, other enhancements

  1. GWT 1.4 has been released. There's lots and lots of cool new stuff in GWT 1.4, so it's hard to know where to start. How about application performance? This release includes several breakthroughs that make your compiled GWT code significantly smaller and faster. Many users are reporting that after a simple recompile with 1.4, their applications are up to 30% smaller and 20%-50% faster. And startup time in particular is now highly optimized thanks to a new bootstrapping technique and the availability of image bundles. In addition to a fast initial startup, GWT code uses a caching technique to prevent applications from making unnecessary HTTP requests. As Joel Webber (Tech Lead of GWT Core Libraries) would say, "The fastest HTTP requests are those that do not, in fact, occur." Here's a broad overview of the major enhancements: New widgets and libraries
    • RichTextArea, HorizontalSplitPanel and VerticalSplitPanel, SuggestBox, DisclosurePanel, PushButton, ToggleButton, and an enhanced Image widget make advanced applications easier than ever.
    • ImageBundle automatically consolidates multiple images into a single HTTP request.
    • NumberFormat and DateTimeFormat make easy work of complex internationalization and localization.
    • You can finally use java.lang.Serializable with GWT RPC, and the GWT RPC server-side subsystem is no longer intimately tied to servlets. You can easily wire it into any Java back-end infrastructure. Spring fans, rejoice.
    • A new JUnit-based benchmarking subsystem makes measuring and comparing the speed of code snippets as easy as writing unit tests.
    New deployment options and optimizations
    • Adding GWT modules to an HTML page is now simple: just add a <script> tag.<li>You can now include GWT modules across domains. Note that including scripts from other sites that you don't fully trust is a big security risk.<li>External JavaScript files referenced from your GWT module load synchronously now, so script ready-functions are no longer needed.<li>Auto-generated RPC whitelist files are now produced during compilation to help catch accidentally responding with objects that compiled GWT client code wouldn't be able to deserialize.<li>The GWT distribution now includes a DTD for the GWT module XML format, making it easier to configure modules in an DTD-aware XML editor.</ul></script>

    Threaded Messages (52)

  2. GWT has made some mind blowing improvements on how web development is being done. I really which they'd allocate a UI expert too really jazz up the Widgets with some pretty CSS. If anyone out there has done this already, place respond to this posting. I still can't stomach the current look and feel :) Is there any word on when GWT will support JDK 1.5 syntax? Best Regards, Richard L. Burton III The quote of the day "The fastest HTTP requests are those that do not, in fact, occur." - Joel Webber
  3. Thanks, Google.[ Go to top ]

    GWT and Guice are innovative and great tools. Thanks to Google for giving them to the community! Congratulations on shipping!
  4. If anyone out there has done this already, place respond to this posting.
    You may want to consider gwt-ext: http://code.google.com/p/gwt-ext/
  5. If anyone out there has done this already, place respond to this posting.


    You may want to consider gwt-ext: http://code.google.com/p/gwt-ext/
    BTW, showcase: http://gwt-ext.googlecode.com/svn/trunk/site/samples/Showcase/www/com.gwtext.sample.showcase.Showcase/Showcase.html
  6. Wow, the gwt-ext looks very impressive. Is it planned that the gwt-ext will have a file upload component with a progress bar, a tree table component and a calendar component (that's what I can think of right now)? If yes, what's the estimated launch time for them?
  7. Wow, the gwt-ext looks very impressive. Is it planned that the gwt-ext will have a file upload component with a progress bar, a tree table component and a calendar component (that's what I can think of right now)? If yes, what's the estimated launch time for them?
    Well, I'm not one of gwtext developers (we're just using it internally for a number of projects), but it is a java/GWT wrapper around the ext javascript library, which already has at least one file uploader (cfr: http://aariadne.com/uploadform/ ). Integrating it into the library in just the same way the rest of the framework does should be relatively easy.
  8. Wow, the gwt-ext looks very impressive. Is it planned that the gwt-ext will have a file upload component with a progress bar, a tree table component and a calendar component (that's what I can think of right now)? If yes, what's the estimated launch time for them?
    Calendar: http://gwt-ext.googlecode.com/svn/trunk/site/doc/html/com.gwtext.client.widgets.DatePicker.html
  9. Thai, in terms of time estimates, you'd need to talk to the author. I've spoken to the author and offered to write some of the Third party widgets (The Ext.ux library) and he pretty much told me to go to hell. He doesn't want anyone contributing to the project in any fashion, whether it be JavaDoc's, Third party Extensions, etc. Feel free to contact him via the user group (By the way, he enabled the "Request permission to join this group" in the Google Groups.) That goes to show you his mindset. I'm thinking of branching his code base and opening it up to various developers to really speed up the process and build up a community around it. If anyone is interested, let me know. Best regards, Richard L. Burton III
  10. Thai, in terms of time estimates, you'd need to talk to the author.

    I've spoken to the author and offered to write some of the Third party widgets (The Ext.ux library) and he pretty much told me to go to hell. He doesn't want anyone contributing to the project in any fashion, whether it be JavaDoc's, Third party Extensions, etc. Feel free to contact him via the user group (By the way, he enabled the "Request permission to join this group" in the Google Groups.) That goes to show you his mindset.

    I'm thinking of branching his code base and opening it up to various developers to really speed up the process and build up a community around it.

    If anyone is interested, let me know.

    Best regards,
    Richard L. Burton III
    I wonder if forking the project is the right approach: having a "dictator" in an open source project is not necessarily a bad thing, and since gwt-ext is relatively new it may be a reasonable way to make it grow without being hampered by endless discussions among developers. If you want to contribute extensions can't you do it with a separate project?
  11. no good deed goes unpunished[ Go to top ]

    I am the author of GWT-Ext ( http://code.google.com/p/gwt-ext/ ) and would like to clarify a few things in response to the accusations / misinformation by Richard Burton III. - I started GWT-Ext out of my own interest and based on the great community feedback I received on my blog ( http://www.jroller.com/sjivan/entry/gwt_ext_integration ) and lots of users eager to use this library, I decided to open source the project soon after ( http://www.jroller.com/sjivan/entry/gwt_ext_0_9_released ). It is licensed under LGPL and not the more restrictive GPL license only because I wanted it to be easier for users to use this library. - The initial version of the library was only released around two months ago and while it is relatively young, it has around 20K lines of code, 2500 - 3000 API's and around 350 classes and interfaces. It has taken me over 700 hours of personal time during nights and weekends to get the library to the point where it is. While Richard might try to trivialize the effort by saying its just a wrapper over Ext, anyone who has attempted similar work from scratch will know the issues and difficulties integrating the GWT component lifecycle and render phase with those of Ext's in a way that the two operate seamlessly. - The project is not closed. The source code is available to all and there is a issue tracker system on the project page too where users can file issues and submit patches. There have been some good discussions on the forums for API changes / improvements / enhancements and I have been open to user suggestions and feedback. It is true that I am presently the only commiter but this does mean that the project is falling behind or dying as Richard puts it. I have responded to all questions asked on the forum within a reasonable time (mostly hours if not minutes) and have been fixing issues and enhancements at a pace faster than most open source projects. It is worth noting that 95% of the Ext API's are already complete and available for users and I continue to work hard on this project. - Now onto what actually happened with Richard Burton III. Richard joined the forum less than two weeks ago and demanded to be added as a core developer. To this I responded saying that I am presently not planning on adding new core developers and certainly someone who started using the library less than two weeks ago. He has no history on the GWT forums, hasn't answered a single user question on the forum, or even filed a single issue (enhancement or bug fix / patch). I tell him that if he'd like to contribute, he can start of with filing issues and patches to demonstrate his commitment and technical skills and gain credibility which may lead me to consider adding him as a commiter over time. - He then sends me a nasty email (which I've included at the bottom of this post) trivializing my project and saying its not that great, yet he tries to pass the project as his own on his registered domain and talks about forking the project out here. I don't understand where that came from when 95% of the Ext API's are already included in GWT-Ext, with development moving at an extremely fast pace with an energetic and growing community. If its so trivial to write a wrapper for Ext, I wonder why he wants to fork, rather than starting from scratch. It really makes me question his intent. I have spend countless hours working on GWT-Ext though the wee hours and over entire weekends to make GWT-Ext available to the community. How many hours has he spend on GWT-Ext that he somehow feels that he rightfully owns this project and if he cant become a commiter, feels the need to hijack the project for no apparent reason. - Richard goes on to register a domain with my project name, includes samples that were written my be as a part of the GWT-Ext project yet has no link to the official GWT-Ext project page. He also posts on the GWT-Ext forum trying to rile users against me and spreading FUD that the project might die for god knows what reason. - The reason the Google Groups presently requires authorization is because it was receiving spam which I had to filter out preemptively. So I recently added forum authorization. All users wanting the join the forum have been included in a reasonable time. Just so that its known to community , I am posting relevant portions of my response to him along with this response which I have chosen not to respond to. It's a disgrace that there are such thankless users but I suppose that's part of being part of the community. Here's what I responded when he asked to be made a commiter :
    I have no intentions of opening up GWT-Ext to other developers at this point. And certainly not to someone who joined GWT-Ext less than two weeks ago. I'm not suggesting you're not competent but I have no idea about your technical skills or involvement in other open source projects. You asked the question before and I already responded saying you can help out by trying GWT-Ext and filing bug reports. There are other ways to help too - for example submitting patches and gain credibility over time which *might* lead to your being considered as a developer down the road. You don't join a project and in a matter of two weeks demand to be part of the development group. ... If users would prefer having complete Javadocs at their disposal then they are better off waiting until the Javadocs have been completed. There are numerous well documented libraries like Spring and Hibernate where there is a lot of traffic on the forums too. The simple fact is that despite being well documented, it is not possible to have isolated useful examples of each individual API. GWT-Ext is in beta now and is less than 3 months old. Early adoptors are welcome to use this library and provide their valuable input as the library shapes but should understand it is still early days and that they would probably need to get their hands dirty on occasion. And this has been the been with the most of the users. My highest priority is getting things done right and the things that are missing like Javadocs and official wiki will come in due time. If that's something you cant deal with, then check back after some time and use the library when things are further along. And as most of you know, I have been working my tail off adding new functionality at a rate that surpasses most other open source projects and also responding quickly to questions posted on the forum. If you are not satisfied with my approach or direction, please move on. And btw I also don't appreciate your registering a .com domain with my project name without atleast running it by me first. Sanjiv
    And here is Richards response to me :
    To be quite honest Jivan, you have this perception that GWT-Ext is your claim to open source fame. Your project suffers severely from the lack of documentation in any form. You absolutely no JavaDoc's and your level of arrogance is beyond deserved, leave it to people who earned the right like Gavin King and Rod Johnson. ... You seem to have a problem understanding your project it's the holy grail to GWT, but you think you'll gain some fame from it. Good luck, just don't be an ******* about it. As for the domain, how about **** you and I own it. Permission my balls. It's called trying to build a community up around an *******. The traffic has been great and well, your project isn't all that great. It's just a wrapper around the ExtJS JavaScript library (Of which those developers are very friendly and open to suggestions.) Best Regards you fucking clown. Richard L. Burton III
  12. Re: no good deed goes unpunished[ Go to top ]

    This site isn't a site designed to run flame wars. To be honest and open about the complete thing, I never demanded to be made a part of the core team (Feel free to check the mailing list for my offers to help on the JavaDoc's or help work on the Ext.ux extension for third party widgets. It doesn't make logical sense to have another project for this when GWT-Ext should also contain this project as well. Secondly, you are very abusive to your members on the mailing list (Refer to http:// groups.google.com/group/gwt-ext/browse_thread/thread/ad2754e715b88f1a/ dbb179c81268ff5dd if it is still posted), which is out of line. Lastly, you failed to mention that you closed off the mailing list after you discontinued my membership, which it fine by me. You never received any spamming on the mailing list, you only have a 100 members. I'm not here on this site to cause a flame war, but let me just state the following facts and I'll leave it at that. 1. I never demanded to be made a core member of GWT-Ext, I did however offer my help to work on the Third party extensions that people contribute to EXTJS.com or assist in working on the JavaDocs. Please refer to the mailing list for proof. 2. I am rather concern about how abusive you are to the members on your mailing list. It isn't how you build a community, instead it's how you destroy one. Please see http://groups.google.com/group/gwt-ext/browse_thread/thread/ad2754e715b88f1a/dbb179c81268ff5dd for an example. 3. My membership from the mailing list was discontinued after sending the below posting on the mailing list. Then the approving of members on the mailing list was enabled. There are only 100 members to the mailing list and LOW usage of it, no spamming. Sanjiv Jivan, The GWT-Ext project is a nice wrapper around the ExtJS JavaScript framework, but in order for your project to really pick up steam, I do believe you need to open it up to other developers. The lack of documentation has caused people to query the mailing list for even the most basic examples which seems to offend you (http:// groups.google.com/group/gwt-ext/browse_thread/thread/ad2754e715b88f1a/ dbb179c81268ff5dd). To anyone who is using the Gwt-Ext project, feel free to post atomic code examples to the www.GwtExt.com Wiki. Over the course of the next month I intend to post various examples for each widget with an introduction to the what the code is doing. I really hope to help get this project moving. In doing so, I'll need the assistance from everyone who has an interest in the project. Best Regards, Richard L. Burton III 4. The part you turned into ... is In regards to responding to how you respond, anyone on the mailing list should be offended by your response. Most projects are more open and accepting to user feedback, input and also help. You seem to have a problem understanding your project it's the holy grail to GWT, but you think you'll gain some fame from it. Good luck, just don't be an ******* about it. I won't continue this any further here in respect to the other members of this site. I did offer to help get your project exposure in various forms, which you so kindly disregarded. Best Regards, Richard L. Burton III
  13. Re: no good deed goes unpunished[ Go to top ]

    It's too bad that this kind of thing has to happen. I blame it on the impersonal/toneless nature of email/mailing lists/forums etc. I bet if the two of you had met for a beer or something and talked, none of this would have happened. I know that this is rarely geographically practical. I'm just sayin'.. Cheers, Freddy
  14. Re: no good deed goes unpunished[ Go to top ]

    Richard is clearly off his rocker. Looking at the gwt-ext forum shows that Sanjiv has been very responsive to posted questions. Even in the thread Richard points out (newbie asks a very basic question), Sanjiv provides the answer, including a code fragment that succinctly answers the question. Sanjiv does mention the term 'spoonfed', but that hardly qualifies as being 'very abusive' (Richard's characterization). As to Richard getting kicked off the gwt-ext discussion forum (I have no idea if that is true), I'd kick him off as well if he went and 'pre-emptively' registered a domain name that had the name of the open source project I'd created. Or I'd kick him off when he started displaying abusive language --> life is too short to deal with this if you don't have to. I think what Sanjiv suggested to gain gwt-ext credibility (start filing bug reports/submitting patches) is very reasonable - 'committer' status needs to be achieved, not merely requested. We've all probably had the experience of cleaning up after well-intentioned but inexperienced developers (which is why 'committer' access is restricted in the first place). And Richard may be a GREAT developer (I have no idea), it is just that he has not demonstrated he'd be an asset to the gwt-ext effort at this point (judging from his posts to the forum). And a beer might have helped the situation a couple of weeks ago, but given Richard's language/actions I think that moment might have passed... Which is unfortunate, as Richard clearly has (had?) a lot of enthusiasm for the gwt-ext technology. cheers, Brad
  15. Re: no good deed goes unpunished[ Go to top ]

    Brad to be completely honest, I enjoy using ExtJS because it provides very nice JavaScript widgets, and using GWT-Ext just makes working with it and GWT transparent. Feel free to join the mailing list, search for my postings and let me know if my posts are if anything supporting and enthusiastic about the project. In regards to my abusive response, well that only came about after I was abruptly remove from the mailing list without just caused. He removed me from the mailing list, I believed because he simply had the power to, but not the reason. Which is perfectly acceptable, just don't send that person an e-mail thinking they won't be offend by that action and stake a claim that I can't register a domain (Which I did mention in the mailing list and you can look that up as well, but it received no response from him.) to build up the community around the project. That is also a display of support for the project. Any good project succeeds because of documentation and community support. You can expect people to ask some very basic fundamental questions about our framework when you don't provide any means of documentation. In short, if the project is not fully documented then either let others answer the questions you feel are ignorant, create the documentation to support the project to avoid this issue, or let others help in build the documentation and accept the fact that letting other people contribute (Not run the project), only helps the project. I have no problem with working on bugs and submitting patches. In fact, there is a posting on the mailing list that expressed an interest in writing implementing some third party ExtJS widgets as an add-on to the project which can could then accept or integrate (Read this as a way to build credibility). His answer was he wanted a consistent code base. I translate that into being "I don't want anyone even contributing to the project." A beer would have been an excellent way to open the channels of communication; I guess it's better than removing people and closing down your mailing list as a means of building support. I know the foul language wasn't necessary, but in all fairness, Jivan isn't the most socially inclined person in the world. I have nothing to hide, that's why I posted the rest of the e-mails on here and also links to the threads that he felt is helpful for him to ignore. Bottom line, GWT-Ext is a nice project because ExtJS is very clean and professional. No credit was given to the developers of ExtJS on his main page, and I didn't post a link back to his project, since I'm in the process of building up the Wiki. Please feel free to add what ever you want to the Wiki. The examples I posted there are locked to avoid people breaking the code samples. Wikipedia.com has such high traffic that any errors are corrected in seconds, so errors in my wiki would degrade the value of the site. I do appreciate your point of view. This will also help give the GWT-Ext project more exposure and I hope that more documentation is the direct result of all this and not a huge flame war. I could care less how people view me online. I do care about community support. Best Regards, Richard L. Burton III
  16. No credit was given to the developers of ExtJS on his main page, and I didn't post a link back to his project
    Last time I have checked Sanjiv's blog (http://jroller.com/sjivan) there are lot of references to the Ext library and direct links to Ext page. The same is true for gwt-ext project page (http://code.google.com/p/gwt-ext) along with several brief explanations on Ext and GWT. In fact in one of his last blog entires (http://jroller.com/sjivan/entry/ext_vs_dojo) he praises the genius of Jack Slocum (the author of Ext) several times. So, honestly, I do not understand why you have written about that. And, frankly, let's make a distinction between LGPL license that allows people to use product in certain way and project organization approach and direction, that does not entitle everyone to get a committer status. LGPL does not define how project has to be developed.
  17. The reference was in regards to his statement that I didn't give a direct link to his project on my wiki. I talk about GWT-Ext project and even linked to the show case, so I don't understand his statement about there being no direct link to his project. And, frankly, I never *asked* to be a committer on the project. I would love for someone to show me where I've asked and stop implying this. Until then, I see no purpose claiming I've said this. Best Regards, Richard L. Burton III
  18. Re: no good deed goes unpunished[ Go to top ]

    Well if you say you didn't ask to be a developer, I don't see what your problem is and all this whining about. The author said that you can contribute by submitting patches, didn't he? He has put out an awesome library that we are using internally to build a prototype. As far as being open, I do remember when you originally started your wiki, it had the words "Welcome to the Unofficial GWT-EXT wiki" along with a link to the forum. I now see that you have removed the word Unofficial and also the link to the developer forum. So if your intent is noble I am not clear on why you are posing to be the official wiki using a domain name of the project that you don't own and have a beef with the author for no good reason. GWT-EXT rocks. Sanjiv, keep up the good work. Rob
  19. Re: no good deed goes unpunished[ Go to top ]

    You're right. I removed that intentionally and I'm not going deny that. The link to his website is there as well to his show cases are there. I removed the link to the mailing list, well, I'm not a member and his intentions are not noble since you need to be approved. Rob, I suggest you go into the mailing list and read the entire thread. Once you're done, identify where I lost all respect for Sanjiv. Hmmm I wonder when? :) I'm addicted to responding to this thread =)
  20. No credit was given to the developers of ExtJS on his main page, and I didn't post a link back to his project

    Last time I have checked Sanjiv's blog (http://jroller.com/sjivan) there are lot of references to the Ext library and direct links to Ext page.
    The same is true for gwt-ext project page (http://code.google.com/p/gwt-ext) along with several brief explanations on Ext and GWT.
    In fact in one of his last blog entires (http://jroller.com/sjivan/entry/ext_vs_dojo) he praises the genius of Jack Slocum (the author of Ext) several times.
    So, honestly, I do not understand why you have written about that.

    And, frankly, let's make a distinction between LGPL license that allows people to use product in certain way and project organization approach and direction, that does not entitle everyone to get a committer status. LGPL does not define how project has to be developed.
    That reminds me :) http://jroller.com/rburton/entry/gwt_widgets_problem_and_solution Cheers!
  21. Re: no good deed goes unpunished[ Go to top ]

    I'm not sure I agree. Bottom line is the Home Page of gwt-ext does credit (and link) to both GWT and Ext, calling the latter highly polished. And his blog entry that kicked off the whole gwt-ext initiative (http://www.jroller.com/sjivan/entry/gwt_ext_integration) gushes about both. And while Sanjiv may be standing on the shoulder of giants (GWT & EXT respectively), it is no small feat to get them to play nicely together... Quick search through the ExtJS forum shows that Sanjiv has 150+ posts, mostly helping people, including posting fixes that Jack (ExtJS 'owner') indicates he'll incorporate into ExtJS. That, in my opinion, is how you build credibility in the open source community --> Sanjiv posts a fix in the ExtJS forum, Jack looks at it and thinks it makes sense, and then Jack decides how to best integrate it into the ExtJS codebase. It is also how it seems to work in the gwt-ext forum - there are lots of posts where users show explicitly track down bugs to specific offending lines of code, and sometimes even post the fix. When Sanjiv says he wants a consistent codebase for gwt-ext, I read it as he prefers working in the common open-source style of reviewing submitted patches/fixes, as opposed to letting people commit directly --> I don't read it at all as "I don't want anyone even contributing to the project" as you do. Especially since he explicitly states that is what you should do to gain credibility. Best of luck in starting your wiki. Documentation, esp. doc aimed at newbies, will be very helpful when gwt-ext gets out of beta and becomes more widely used. cheers, Brad
  22. Re: no good deed goes unpunished[ Go to top ]

    That page (His home page) has been updated various times. Once again, I never asked to be a committer. The topic at hand isn't the 'complexity' of the project. It isn't about a request to be a committer (Not sure where that came from), and lastly it isn't about me telling him how to run his project. I'm simply trying to build a community to help support his project. If I'm at fault for spending my money, time and effort to help build a community, then there is something logically wrong with that claim. Reviewing this thread, the claims about me are a) A demand to be committer. Show me the thread or the e-mail in which reflects this. b) My foul language and abusive nature was uncalled for. I have the write to voice my opinion and it was done in private to him, not done on his mailing list or in some public fashion. Closing off a mailing list, sending a personal e-mail, be insulting about it, and then claiming the mailing list was closed because of spam on *Googles* mailing list. priceless! I'm done replying to this thread because its really degrading the value of the initial post and it's rather pointless. Not to mention that I broke my initial promise. Best Regards, Richard L. Burton III
  23. Re: no good deed goes unpunished[ Go to top ]

    Just to be clear.You kept prodding him on "Opening up the project" when he said that you are free to post patches through the issue tracker. Maybe I misread what you meant, but if you don't want to submit patches, and don't want to be a committer, what does opening up the project mean? You could and I am sure still can submit patches through the issue tracker. I believe there is already an open bug for Javadocs and another user has said he will be providing patches. I will be interested to see if you are going to submit any patches or if this is just a publicity stunt. Rob
  24. Re: no good deed goes unpunished[ Go to top ]

    About the JavaDocs, that's when I got the "Consistent code base" response. I'm working with GWT-Ext and when I come across a bug or something that needs to be patched, I may submit it, but I'd first consider the interaction with the developer first. Not sure I'd want to even bother interacting with him.
  25. Re: no good deed goes unpunished[ Go to top ]

    Richard, You seem to be at odds with the author so just create your own ajaxy-webservice-jms open source project and run it the way you want. Or better, do something useful and try contributing to GWT or Ext. Why don't you just leave gwt-ext alone if you are so bothered by it. After reading the entire thread its pretty evident what you were trying to do. You say you want the project to be opened up, yet deny you ever wanted to become a member commiter. You say you want to further this project and although the author invited you to submit patches, you still find reason to complain and not do anything useful like contribute anything concrete. You're one of those guys that like to take credit for others work - and I have seen plenty during my 15+ years in the field. Finally this thread is about GWT 1.4, not about Ext or gwt-ext. Stop posting off topic!! -Andrew
  26. Re: no good deed goes unpunished[ Go to top ]

    In 15 years of software programming, you seem to learn about about psychology. Maybe you're in the wrong field. Thanks for continuing the thread :)
  27. Re: no good deed goes unpunished[ Go to top ]

    In 15 years of software programming, you seem to learn about about psychology. Maybe you're in the wrong field. Thanks for continuing the thread :)
    TSS, please boot this 13yr old girl.
  28. GWT extension[ Go to top ]

    What I would REALLY like to see as an extension to gwt would be a generified AST that the gwt compiler could read, so that people could develop in other languages and still use the gwt compiler to spit out the .js stuff! Don't know if it is feasible, but that sure would be nice! By the way, great job with the gwt-ext stuff guys!
  29. Re: GWT extension[ Go to top ]

    Miguel, A co-worker and I were having the same discussion about this a week back. Java has always lead the market in terms of innovation, but we're also seeing companies like interface 21 and JBoss also getting into the .Net game as well.
  30. Re: no good deed goes unpunished[ Go to top ]

    It's too bad that this kind of thing has to happen. I blame it on the impersonal/toneless nature of email/mailing lists/forums etc. I bet if the two of you had met for a beer or something and talked, none of this would have happened. I know that this is rarely geographically practical. I'm just sayin'..

    Cheers,
    Freddy
    That's a good point.
  31. Re: no good deed goes unpunished[ Go to top ]

    This site isn't a site designed to run flame wars.
    Perhaps you should not have started one then? :) You were pretty harsh to the other dude, and i think he had the full right to defend himself. That said, majority of these kind of things are caused by the limitations of electronic communication, and i wouldn't stress about it.
  32. Re: no good deed goes unpunished[ Go to top ]

    I'm not starting a flame war here, just clearing up the facts based on your negative posts. I have never been abusive to any users on any of the mailing lists of projects I've been involved with (in the capacity of a commiter, or a user asking or answering questions). And this includes a lot of forums including AppFuse, Lingo, Spring, Hibernate, Ext, AjaxAnywhere and several others. Feel free to look me up in any forum. My user id is "sjivan" It is your interpretation that in the one thread you keep pointing to, that I was very abusive and turn this around into a big deal that I'm "very abusive" to members on the mailing list. Anyone joining the GWT-Ext forums can decide for themselves. I have already pointed out that in that particular case I was merely pointing out that the user make an attempt to use the API when he had a fairly good idea of what it may have been and ask questions if he ran into any problem. And seriously, why are you the peacemaker / moderator on a list that are just joined. And from your email to me we know that you're not the most polite / considerate guy around. I can understand you may be upset that I turned down your offer to be a commiter, but please not spread FUD about me or my project to further whatever agenda seem to have (something that i fail to understand myself).
  33. The problem with GWT-Ext is that core developer doesn't want anyone to contribute to the project in any fashion besides using it. I'm in the process of building a Wiki around GWT-Ext to show examples of various widgets and explain the logic, but more support from the developer is required. Otherwise its a one man project which could die very fast. Best Regards, Richard L. Burton III
  34. Would it be more sense to do something with Yahoo UI library instead?
  35. ExtJS.com uses both YUI and JQuery.
  36. The contributor list makes it seem like the 1.5 stuff is moving along pretty swimmingly so far. The plan is to have the 1.5 release be focussed almost solely on getting 1.5 syntax functional. The trunk doesn't compile right now, which my optimistic side is taking as a sign of huge progress. I expect builds suitable for burning my fingers on soon. For me this is the last stepping stone before gwt induced nirvana ;) -Jeff Dwyer http://www.myhippocampus.com
  37. GWT 1.4[ Go to top ]

    Does it Support Java 5.0 yet .. this has been the biggest drawback for me
  38. Re: GWT 1.4[ Go to top ]

    Does it Support Java 5.0 yet .. this has been the biggest drawback for me
    No, but there has been official word that it's in progress and high/critical priority.
  39. GWT in JBoss Drools[ Go to top ]

    We've used GWT for all our web based stuff in the Drools project and are really happy with it. We've detailed our work using GWT in our Business Rules Management Studio, with lost of screenshots and a screencast, which you can read here http://blog.athico.com/search/label/BRMS. Further screenshots, towards the bottom of the page, can be seen here http://labs.jboss.com/drools/featuresandscreenshots.html. You can also download the Drools BRMS standalone project and see GWT in action out of the box http://labs.jboss.com/drools/downloads.html. We'll be upgrading to GWT 1.4 shortly and will be implementing grid based decision tables before xmas using GWT. Mark http://blog.athico.com
  40. Re: GWT in JBoss Drools[ Go to top ]

    Thanks for the information. I'm looking into GWT and those are great links. Are you looking into using a particular new feature from 1.4 for Drools? Or the upgrade will be mostly for the performance gains? Cheers, Paul Casal Sr Developer jbilling.com The Enterprise Open Source Billing System
  41. Re: GWT in JBoss Drools[ Go to top ]

    Thanks for the information. I'm looking into GWT and those are great links. Are you looking into using a particular new feature from 1.4 for Drools? Or the upgrade will be mostly for the performance gains?

    Cheers,

    Paul Casal
    Sr Developer
    jbilling.com
    The Enterprise Open Source Billing System
    We haven't anything specific yet, as we are just winding down 4.0.2 (for next week), then we will look at the next stage of R&D. We'll first upgrade to 1.4 and then see what we can take advantage off, I need to see if the current grids will do what I need or if I need to build a new one - this will all be determined over the coming weeks. Mark http://blog.athico.com
  42. Re: GWT in JBoss Drools[ Go to top ]

    Hi Paul. Well from a quick look at GWT 1.4, there is the rich text widget, and the splitter stuff which will be handy. Probably a few other things we can replace with "out of the box" widgets. Also the new improved RPC (using Serializable) etc. will allow us to rip out some legacy code to work around the limitations. But haven't looked further. I havent seen much change in the Grid stuff, so we will need a lot more work in that area (I believe the Grid stuff, while competent and powerful in GWT - is still pretty basic - you have to customise a lot to do what you get for free with other frameworks, which is probably the only complaint I have with GWT right now).
  43. Re: GWT in JBoss Drools[ Go to top ]

    Hey Mark, I spoken to some of the folks who work on the Drools project (It's me rburton). My problem is I lack the ability to do UI :) FYI the Wiki for GWT-Ext is www.GWTExt.com and anyone can help contribute examples there as well. In fact, I encourage it! The GWT-Ext project is just a wrapper around the ExtJS.com framework with basically a one to one mapping to the API. I know some people don't like wrappers around JavaScript frameworks like this. Best Regards, Richard L. Burton III
  44. Re: GWT in JBoss Drools[ Go to top ]

    Just my curiosity :) Why didn't you choose RichFaces / Ajax4jsf for Drools instead of GWT?
  45. Re: GWT in JBoss Drools[ Go to top ]

    Well at the time we started, RichFaces wasn't available under a suitable licence. Also, we needed something that made it easy to create custom/fiddly widgets. Most toolkits are designed to make it very easy to re-use "out of the box" widgets (and rightly so, that is what most people want). Also, we have some GUI code (ie code that is compiled by GWT) that is re-used in the eclipse plug in (increasingly) - which is a somewhat unique requirement which GWT allows (kind of wacky, but useful !).
  46. Re: GWT in JBoss Drools[ Go to top ]

    Just my curiosity :) Why didn't you choose RichFaces / Ajax4jsf for Drools instead of GWT?
    As mic said and also Ajax4jsf doesn't(maybe didn't) scale as well and is more oriented to CRUD type applications; we wanted something more swing like. Mark http://blog.athico.com
  47. Re: GWT in JBoss Drools[ Go to top ]

    Just my curiosity :) Why didn't you choose RichFaces / Ajax4jsf for Drools instead of GWT?


    As mic said and also Ajax4jsf doesn't(maybe didn't) scale as well and is more oriented to CRUD type applications; we wanted something more swing like.

    Mark
    http://blog.athico.com
    Did you also looked into the Eclipse RAP? It's not swing... but it's much more desktop like (swt in this case).
  48. Re: GWT in JBoss Drools[ Go to top ]

    Just my curiosity :) Why didn't you choose RichFaces / Ajax4jsf for Drools instead of GWT?


    As mic said and also Ajax4jsf doesn't(maybe didn't) scale as well and is more oriented to CRUD type applications; we wanted something more swing like.

    Mark
    http://blog.athico.com


    Did you also looked into the Eclipse RAP? It's not swing... but it's much more desktop like (swt in this case).
    I've heard of RAP, but haven't looked into it. I did look into another javascript api that used the SWT api, seemed great in principle and had nice demos, but again bloated a bit. One of the things that makes GWT different is the API is designed for the web and cab be compiled down to have the minimum included that your application needs.
  49. Re: GWT in JBoss Drools[ Go to top ]

    I just tried the RAP demo, it's so slow to be unusable. I think this shows the strength of the GWT toolkit, it's a ground up approach that pushes state client side and delivers minimal sized dependencies all in a scalable system.
  50. Re: GWT in JBoss Drools[ Go to top ]

    ... and also Ajax4jsf doesn't(maybe didn't) scale as well ...
    Mark
    http://blog.athico.com
    What do you really mean? I don't understand. If a4j has some defect, I'd like to know it. One of the things I don't like in GWT is that there's no way to connect a bean property to a text box like the JSP EL (in GWT, I have always use myBean.setMyProperty(myTextBox.getText())). Or is there some way to do that that I'm not aware of?
  51. Re: GWT in JBoss Drools[ Go to top ]

    Here is an example of GWT databinding: http://www.zenika.com/blog/2007/08/03/gwt-data-binding-english-version/
  52. What is RAP? The RAP project aims to enable developers to build rich, Ajax-enabled Web applications by using the Eclipse development model, plug-ins with the well known Eclipse workbench extenstion points and a widget toolkit with SWT API (plus JFace). http://www.eclipse.org/rap/
  53. Serializable[ Go to top ]

    Excellent to see that we no longer need to be coupled to any GWT interface and we can now use simple serialization. This was most certainly holding me back from using this toolkit. Claudio