> That's a non-starter. While JSF and Struts are > already accepted by the market, customers and mgmt, > Shale is still too new (near zero market share, still > under early development not release...). While I'd > love to start playing with Shale, it's not going to > make into my or my customers' deployed apps anytime > soon, while JSF and common JSF extension tags (like > MyFaces Extensions) could.
I don't really consider Shale to be any more "experimental" then the code I am writing on a daily basis to solve the same problems as Shale is attempting to solve. Your biggest risk in using Shale is that some of the approaches they are using to solve certain problems may change and so you may have to refactor your code. Take the new dialog stuff for example. If you already have a solution for that then fine, you might want to wait until Shale stabilizes in order to limit the number of rewrites. On the other hand you may not have a good solution or you may be adventurous (as in my case) and want to try out the new stuff. If you didn't have a solution to begin with there isn't much downside in working *together* with smart people like Craig and David to develop a superior one. Don't think of it as using experimental code. Think of it as working together with talented people to write code that you would otherwise have to write entirely yourself. If you are using JSF (which is an official standard) there are some gaping holes there (intentionally) that projects like Shale and MyFaces are trying to fill in. You can choose to turn your back on 50% of the solutions by ignoring Shale but that's between you, your managers and your clients. > Completely disagree with you on that one. Adding > something shale-specific to MyFaces makes no sense. I didn't say I was in favor of that. > Adding a JSF tag(s) that would be of use to all users > of any JSF (since JSF validators ootb leave a bit to > be desired) across all JSF implementations (JSF is > supposed to be a standard, so apps you build with JSF > should run on all JSF impls where possible) to MyFaces > Extensions (not MyFaces core) makes complete sense. I believe that in the case of the validators added to Shale, that they do not actually require Shale to run. I personally don't care which jar file they sit in or whether I type <s:foo> or <x:foo> > If Shale had an extension library that had useful JSF > tags that could be used across all JSF > implementations, then I could see your point. Well they basically do have some things that can be used across JSF implementations (and without Shale.) <s:token> is a prime example. In the case of s:token and the validators I would agree that you could make the case that they would find a nice home in the MyFaces components package. I suspect there are a few reasons why this has not happened. I'd also be interested in working with the Shale team to see if we can work together on stuff that could run independently of Shale. > Struts didn't include commons-validator within Struts > core, it's a separate subproject so that it could be > used across other projects that had similar needs. Right and that is what David Geary has done by porting them to JSF. That took a little bit of work to pull off I'm sure. I have no desire to repeat that work when I can just use "import org.apache.shale ..." > JSF is shipping in several commercial IDEs already, > and thus has market and mindshare. Shale, while cool > and interesting does not yet have that luxury. I > hope that it does at some point, but I would still > like to see generally useful JSF tags included in the > JSF project extensions not in a superset framework > project. Shale is an add-on to JSF. Its no less of a standard than the JSF extras in MyFaces. Shale is being lead by one of the lead people on the JSF standard. Those two technologies couldn't be more perfectly alligned. > Thanks, and I know these questions are more oriented > toward Craig and David than you and the MyFaces PMC, > but I was hoping someone on the MyFaces Extensions > contributor team would offer to add this functionality > to the MyFaces extensions (at which point Craig and > David could decide whether to keep their own copy in > the Shale "core" or to depend on the MyFaces > extensions like other JSF apps and JSF supersets > could). Like I said 0% chance of us duplicating Shale work. Both projects are short on help as it is so there is no time to duplicate effort between two Apache projects. sean
