It's refreshing to see a number of people out there who "get" where Tapestry is headed.
There is always a tension between compatibility and the drive for new features. Tapestry's baggage: the base classes that begat abstract classes; the conflict between page names and class names, the plethora of lookup paths for various artifacts ... all of these things are choking Tapestry. Looking forward, where full page updates are the exception, and Ajax-oriented partial page renders are the norm, Tapestry 4 will not be able to keep up. Jesse has been proving himself as the master of this stuff, but there are just some issues buried in the DNA of Tapestry 4 that extend all the way back to the Tapestry prototype in 2000. Perhaps I should have kept quiet until I had more of T5 to show. I think everyone is going to agree that the new feature set, the new style of development, the simplicity and the power, are going to be quite compelling. The central issue is backwards compatibility. As the upgrade from 2 to 3 to 4 has shown, adding new features to Tapestry often breaks existing code. This is a reaction to the relationship between the framework code, and the application code. The fact that application classes extend framework classes means that virtually any change to the framework classes exposes client code to incompatibilities. Further, the fact that so much logic passes through user code causes its own set of problems when we want to add in more radical new features (such as true WYSIWYG preview). I've forced people to choke down the poison pill in little stages, from 2 to 3 to 4. I don't expect that to happen once 5 is out ... the annotation-based APIs are wonderfully flexible and adaptive even when the framework is changed. My goal is not to beat JSF, but to give Java developers a compelling reason to stay on Java and not jump over to Ruby on Rails. That's a tall order. On 8/1/06, Mark Stang <[EMAIL PROTECTED]> wrote:
James, One of the reasons we haven't switched is because Geoff's Spindle wasn't there. So, I agree, that tools are important. My point was, that most Tapestry Users won't migrate over to JSF just because we have to upgrade. I agree that Geoff has had an extraordinary bad time of providing an upgrade to Spindle for 4.x. However, I read his e-mails and the problems that he is encountering are, in a large part, due to the framework gone wild of 4.x. The main reason that 5.x exists is because 4.x is such a wild child. It has gotten out of control. The reason that 4.x exists AT ALL is because, as Howard was writing the next version of Tapestry, people were complaining that there wasn't any upgrade path. That the differences between 3.x and 4.x of old were so different that everbody was complaining. Everyone wanted 4.x as an intermediate version. It is there for those who couldn't use 3.x and didn't want to wait for 4.x. So, in reality, 4.x is just a stop-gap, not to denigrate all the work Jesse, et.al. have contributed (he is a wild man). The "next generation" of Tapestry is not 4.x or 3.x, it is 5.x. Should Geoff have done 4.x? I think he did it because we pushed him into it because we weren't willing to wait for 5.x. Bottom-line, 5.x is the future, 3.x and 4.x are prototypes and support will decline for both of them as 5.x becomes the standard. With that said, if Howard has some brilliant idea and abandons 5.x, all hell will break loose. But from his e-mails, the plan is to maintain and enhance 5.x for the future. All roads lead to 5.x regards, Mark -----Original Message----- From: James Carman [mailto:[EMAIL PROTECTED] Sent: Tue 8/1/2006 9:06 AM To: 'Tapestry users' Subject: RE: Tapestry 5 Discussions Mark, you also have to consider a different type of user. For me, a component/framework extension developer (Tapernate, tapestry-acegi, etc.), I am not going to want to rewrite all of my cool stuff each time a new version of Tapestry comes out. No way will I maintain a version of my components for each version of Tapestry. What about Trails, which is helping Tapestry gain some attention by providing a cool RAD environment? If innovative folks get sick of having to rewrite their stuff all the time, then they'll just stop writing components for Tapestry altogether and that'll hurt the community. Also, what about tool developers? The cognition folks have a pretty cool Eclipse plugin that will probably have to be reworked for T5. Spindle also suffered the same growing pains. I don't want to put words into Geoff's mouth, but he seemed somewhat troubled by the fact that he had to totally rework Spindle for T4 from T3. Hugo Palma is creating a TapIDEA, an Intellij IDEA plugin. He'll also be impacted by this as his IDE extension will probably have to be completely reworked. I know that some folks aren't very impressed by tools and they don't think that tool support should be the reason that people choose a platform, but to some they are very important. -----Original Message----- From: Mark Stang [mailto:[EMAIL PROTECTED] Sent: Tuesday, August 01, 2006 10:25 AM To: Tapestry users; Tapestry users Subject: RE: Tapestry 5 Discussions I don't think I agree. We switched to Tapestry from Struts because it gave us a component framework. Internally, we have three projects on Tapestry. One is 4.x and the other two are 3.x. For the 3.x projects we have looked at 4.x and while we would like to be on the latest and greatest, there isn't enough of a ROI to justify moving at this time. And since 5.x is in the "near" future we are waiting. However, we might not ever upgrade. What would cause us to upgrade? Everything works. And when we have had problems we post it to the group, which usually results in a fairly quick fix. Or if push comes to shove, we pay Howard. What more could you ask of a framework? And if you think about what brought us to Tapestry, it wasn't the upgrade path or support, it was the ability to develop components. >From everything I have read, we will still have "pages" and "components". Will we have to rewrite all of our components? I don't think we will have to do so, mainly because they are not that tied to the API. regards, Mark -----Original Message----- From: Danny Angus [mailto:[EMAIL PROTECTED] Sent: Tue 8/1/2006 7:51 AM To: Tapestry users Subject: Re: Tapestry 5 Discussions > Finally, let's take a sober look. Of all the production apps written > in T4, how many do you REALLY BELIEVE would be ported to T5? I'd say 1 > of a hundread, if that. On the other hand tapestry provides us the the ability to re-use components. If we want to write new applications in Tapestry5 do we throw away all our old components and lose their value? Or do we go to the expense of migrating them and writing new ones? For the people who are stuck requiring support for product which is likely to be ending its life the choice will be a stark one, not whether to upgrade to Tapestry 5, but what framework to migrate to. I would predict that most of the people who see their investment in components become increasingly worthless will have little loyalty left and will plump for something which is more likely to protect their investment, no matter what the technical limitations are. Look out for people offering a Tapestry4 to JSF migration path. d. **************************************************************************** *************************** The information in this e-mail is confidential and for use by the addressee(s) only. If you are not the intended recipient please delete the message from your computer. You may not copy or forward it or use or disclose its contents to any other person. As Internet communications are capable of data corruption Student Loans Company Limited does not accept any responsibility for changes made to this message after it was sent. For this reason it may be inappropriate to rely on advice or opinions contained in an e-mail without obtaining written confirmation of it. Neither Student Loans Company Limited or the sender accepts any liability or responsibility for viruses as it is your responsibility to scan attachments (if any). Opinions and views expressed in this e-mail are those of the sender and may not reflect the opinions and views of The Student Loans Company Limited. This footnote also confirms that this email message has been swept for the presence of computer viruses. **************************************************************************** **************************** --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
-- Howard M. Lewis Ship TWD Consulting, Inc. Independent J2EE / Open-Source Java Consultant Creator and PMC Chair, Apache Tapestry Creator, Apache HiveMind Professional Tapestry training, mentoring, support and project work. http://howardlewisship.com --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
