On Nov 1, 2007 4:44 PM, Don Brown <[EMAIL PROTECTED]> wrote: > Here is the problem I've having - we are writing a book, and since > this whole issue seems far from resolution, we've been using the XML > configuration throughout the book (it is almost done). What I'd > rather have done is use the convention stuff to cut the code size and > complexity of the examples down, which, IMO, would have made it easier > to learn. Therefore, I'd like to get this resolved asap.
I'd suggest that convention-over-configuration might be an easier learning curve for people new to Struts 1 or 2. But, I'd wager that people who have been using Struts 1 for several years will find Struts 2 easier to learn using the tried-and-true struts.xml. I'd expect that the first wave of Struts 2 adopters are going to be people with a Struts 1 background. After that, we will start to see more newcomers, and more people who would be ready for convention-based development. If it were me, I'd finish the book using struts.xml, and go to work on a second edition as soon as SmartURLs goes to 1.0 (even if the first edition isn't done yet). Getting a couple of solid Struts 2.0 books out there is the best way to drum up marketshare for a Struts 2.1 edition. In my experience, there are huge discrepancies between the developers who subscribe to lists like this one, and frequent blogs, and attend seminars, and try new platforms, like Ruby on Rails, and the other 95% of developers, who just work for a living :) When I'm training out in the field, many of the people in the room haven't even visited the Struts website. Most have never even heard of Ruby on Rails, and, of those that have, maybe one person has actually tried it or even read a book about it. Everyone here is very aware of the pain-points of XML and non-POJO Actions, but the rest of the world: not so much. Out in the field, people tell me that the biggest problem they have with Struts 1 is that it's so easy and so reliable that the developers are bored! I think a lot of product managers are happy to have that problem! (As the curse goes: "May you live through interesting times!") Of course, the silent majority is only silent when they are bored. There are a lot of war stories out there about teams who tried frameworks like JSF and went screaming back to Struts, because JSF was too different. Despite Suns' best efforts, today more job posting cite Struts experience that all other frameworks combined. Heck, Rick Hightower is afraid to even put Struts on an "absolute" chart. Why? Because in absolute terms it dwarfs all the others. * http://www.indeed.com/jobtrends?q=ejb+java%2C+hibernate+java%2C+spring+java%2C+struts+java%2C+jsf+java In absolute terms, Hibernate and Spring are converging with Struts, and JSF is trailing behind (way behind). The lesson of JSF is that when there is a dominant paradigm, evolution trumps revolution. In terms of the first Struts 2 books, I'd be careful to present S2 as the same thing only better, rather than as a brave new way to develop. -Ted. <http://www.husted.com/ted/blog/> --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
