On Fri, 22 Nov 2002, edgar wrote:
> Date: Fri, 22 Nov 2002 09:09:41 -0500 > From: edgar <[EMAIL PROTECTED]> > Reply-To: Struts Users Mailing List <[EMAIL PROTECTED]>, > [EMAIL PROTECTED] > To: 'Struts Users Mailing List' <[EMAIL PROTECTED]>, > [EMAIL PROTECTED] > Subject: RE: future of struts > > It is wonderful to be misunderstood. > I don't misunderstand you -- I just think you are wrong :-). > Because struts does so much of the work a big chunk of what is left is > the tags. I would compare it to the fit and finish of a car. How many > car purchasing decisions are made because the driving is pleasurable or > the color is correct. > > I can understand the groups wanting to minimize the tags importance > since you are comparing a non MVC situation to a situation with MVC. > That is missing the point, since now you have a framework (many thanks > to those who have spent the time putting it together). > > My point is that the tags should not be second class citizens. Yes you > want to minimize and focus them so that the effort put into them is > worthwhile. But we as a user community shouldn't have to rewrite the > tags to make them useful to our projects or spend silly time triing to > work through gaps in the usefullness. > I would take a slightly different view of this. One of the reasons people have told me that they chose Struts is the fact that it imposes *no* restrictions on how you implement your model tier. Choose DAOs or direct JDBC or EJBs or JDO or XML or ... anything you want, and it will work fine. Struts does not impose any "you have to use OUR model tier interface in order to use the rest of the framework", like many other frameworks do. In the view tier, however, the picture is much more confused. People look at the Struts tag libraries and don't clearly understand that you *can* use these (and, if you do, there is some pretty compelling ease-of-use features), but you don't have to -- as everyone who uses Velocity or XML or other non-JSP solutions. The tags themselves are ***not*** the most valuable thing about the framework -- they are just a convenience for implementing one of the view tier approaches. In today's world, the JSTL equivalents for the Struts tags in the bean and logic libraries are ***much*** more powerful than their Struts counterparts. Given that, it makes natural sense for Struts users to start migrating towards JSTL. The same thing is going to happen at the UI component level with JavaServer Faces, when it is finally released. Why should users continue to wish for proprietary (to Struts) tags that are less functional, and work only with JSP (JavaServer Faces is not limited to use in JSP pages)? >From my Struts developer perspective, I have a limited number of hours to devote to Struts. I would prefer that my hours (and those of other committers, but that is really their business to decide) be spent enhancing the unique values that Struts brings (the controller architecture), and not trying to compete with current and emerging Java standard APIs that are much better than what we already have. It's open source -- committers can work on the parts they want. Users can submit patches on the things they care about. But I can guarantee you that *I* am not going to spend anything more than maintenance time on the current Struts tag libraries in releases after 1.1. > Anyway, I will stop swimming upstream. > > Edgar Craig -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
