On Sunday 26 July 2009 08:12:05 am Rene Gielen wrote: > That is what I also regognize from the disussions a while ago. > > We all know that the Dojo tags turned out as a nightmare. Dojo looked > promising back in these days, with all it's supporters and some good > solutions when there were only few alternative frameworks around. > Nevertheless, not only the programming model was hard to get and to > understand, also the API was in constant flux without ever coming to a > stable state for years. Today we can consider Dojo as the biggest loser > in the JS framework run. > > jQuery nowadays has turned out to be stable both in functionality and > API stability, compact, fast, mature and easy to understand. It can > really be considered as _the_ JS/AJAX framework besides Prototype, and > the adoption is overwhelming. Many Java web delvelopers, including a > couple a S2 community members I know of, have already incorporated plain > jQuery into their projects and built up a good knowledge about it. Most > of our "competitors" incorporate it either directly or via mature > plugins, have a look at Myfaces, Richfaces, Icefaces or Wicket. > > My feeling was that there was a consensus that jQuery would be the way > to go if we wanted to go provide a good AJAX solution, but we had not > enough resources to build a plugin from scratch. Now Johannes stepped up > and presented a complete plugin, and IMO we should see this as a great > opportunity. If we don't want to fall terribly behind the other web > frameworks, we need a decent and state of the art AJAX support - pure > Web 1.0 interfaces are more and more going away, and people tend to > judge a web framework for consideration in their projects also by it's > AJAX capabilities - e.g. when Rainer and I did some Struts 2 talks the > last couple of months for JUG Cologne, that was a very clear feedback we > got from many people in the audience. > > Once the plugin is there, I don't see why it would be any harder to > maintain than oher external library plugins we have, let's say Spring or > Sitemesh or what not - especially since I suspect a lot of S2 developers > and community members are using jQuery on regular basis and will be able > to help with their experience. Also, we don't have to fear an API > breaking with every minor release as we had with Dojo, which made > maintaining it that hard. > > - René >
I am going to back Rene 100% on this. The big thing to me is the API stability. I think it should say something that M$ includes JQuery with their Visual Studio products. Personally, I am using my jquery plugin, and will quickly port my current project to this new one (meaning it will get bugfixes along the way if/when I find them) and I know Musachy is familiar with JQuery from his help when we started the other plugin a while back. I posted the high-level issues earlier, but I am going to take a little bit to browse the code today and see if there are any technical issues. Johannes mentioned a lack of familiarity with maven, but I am pretty sure that won't be too hard to work around, I've ported a few projects to maven recently and feel like I am getting pretty handy with it. Another thing that no one has mentioned yet is that we are going to indirectly end up supporting JQuery if we continue to allow these "grassroots" efforts come along without adopting an official support plugin. Right now, there is my plugin (which I'm using, but I'd guess that's it), but there is the other plugin that my buddy Eric is working on here - http://code.google.com/p/struts2-jquery-plugin/ Eric's plugin is becoming somewhat mature as well, but due to his geographical location it is hard for him to participate in the mailing lists. IMO, more and more users are going to start using these plugins. I think we are at a point with all of these efforts that we could try to combine them and get a best-of- all-worlds type of plugin, then keep it working between releases by building a new showcase with selenium tests. It'd be a big effort, but probably have a huge return in that we could start regaining users that we have likely lost to other faster-moving frameworks (Spring MVC). To address Martin's issue directly, I think the main reason we wanted to move away from Ajax with the 2.1 release was the difficulty to support Dojo. I know, personally, I was all for bringing in things like the JSON result type and taking a framework-neutral approach to future development, but as Al mentioned, this is a pie-in-sky goal. I think that if we push through the next effort as a plugin and showcase app, we are still going to enhancing the struts2-core to support Ajax, without tying us to JQuery. But, by adopting a somewhat official plugin, we can provide first-class Ajax support. -Wes -- Wes Wannemacher Author - Struts 2 In Practice Includes coverage of Struts 2.1, Spring, JPA, JQuery, Sitemesh and more http://www.manning.com/wannemacher --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@struts.apache.org For additional commands, e-mail: dev-h...@struts.apache.org
