I read through it quickly, still planning on a deeper analysis later. One thing I have to say is "view actions". I wanted them badly in JSF 2.0 and now Lincoln has done a great job of presenting to a good portion of this expert group why they are useful (in his PrettyFaces talk). Once again, we see it in this article. So, Lincoln and I will touch base on a proposal to complete the view metadata processing on initial request to include view action support. I've been meaning to provide a patch for this, but I'm glad that I've got Lincoln to work with since he has some other ideas beyond what I was thinking. (action item) Second, conditional validation just keeps biting us. I realize graph validation is complex, but we really need to just consider the use case that this guy is bring up (which is the same use case I hear over and over) and just try to think how we would solve it if we were developing the app, then try to reduce that complexity. Bean Validation is excellent for field-level validation, but it really doesn't deal with the "if this, then that" or "this before that" situation. I proposed one solution in the JSF 2 and beyond slides using the postValidate event, but really that isn't much prettier. -Dan On Fri, Dec 11, 2009 at 12:02 AM, Dan Allen <dan.j.allen(a)gmail.com&gt; wrote: -- Dan Allen Senior Software Engineer, Red Hat | Author of Seam in Action Registered Linux User #231597 http://mojavelinux.com http://mojavelinux.com/seaminaction http://www.google.com/profiles/dan.j.allen

On Fri, Dec 11, 2009 at 2:58 AM, Ken Paulsen <Ken.Paulsen(a)sun.com&gt; wrote: Sadly there is no video. But I'm going to make sure we capture the audio the next time Lincoln presents it. I believe he is putting together a blog entry about it in a couple of weeks. I think the best approach here is for us to simply work up our proposal. If I start explaining what is different, then I'll be saying the same thing in the proposal, but with an ad-hoc response. There is a BeforeRenderEvent for both the view (preRenderView) and component (beforeRender). Again, we'll cite those in the proposal. I should also mention that Andy Schwartz has also been an advocate of the view action idea and will likely be working with Lincoln and I to whip up the proposal. I'll try to get that out as soon as I can. Thanks for being patient ;) -Dan -- Dan Allen Senior Software Engineer, Red Hat | Author of Seam in Action Registered Linux User #231597 http://mojavelinux.com http://mojavelinux.com/seaminaction http://www.google.com/profiles/dan.j.allen

On Fri, Dec 11, 2009 at 1:15 PM, Jim Driscoll <Jim.Driscoll(a)sun.com&gt; wrote: As did I. Naturally, they look a little naked w/o styling, but still, the base components should be sufficient for doing something. I think you look at your app afterwords and say "man, this could really use some richness". Then you reach for libraries. -Dan -- Dan Allen Senior Software Engineer, Red Hat | Author of Seam in Action Registered Linux User #231597 http://mojavelinux.com http://mojavelinux.com/seaminaction http://www.google.com/profiles/dan.j.allen

On Fri, Dec 11, 2009 at 4:14 PM, Ken Paulsen <Ken.Paulsen(a)sun.com&gt; wrote: Gotcha. Of course. We just need to make sure that they are a good starter set. Nice to see Jim filing that issue for a better panelGrid ;) -Dan -- Dan Allen Senior Software Engineer, Red Hat | Author of Seam in Action Registered Linux User #231597 http://mojavelinux.com http://mojavelinux.com/seaminaction http://www.google.com/profiles/dan.j.allen

On 12/11/09 10:15 AM, Jim Driscoll wrote: Done. https://javaserverfaces-spec-public.dev.java.net/issues/show_bug.cgi?id=693

I got as far as: "JSF 2.0, which does address some – yet by n means all – of my biggest issues with JSF is totally out of the question, since it requires JEE6 and the specification has not been finalized until very recently." and pretty much gave up. I had to wonder how hard it would have been to find out that you can use JSF 2 with Java EE 5. I have to say I admire Dan's persistence in taking the other points he makes seriously, I'll have to review the doc over my vacation to extract points from it. One obvious point is that we don't state clearly on our website that JSF 2.0 only requires Servlet 2.5 and JSTL 1.2. I'll take that AI. His email is karl.banke(a)iternum.com Anyone want to volunteer to contact him, to correct the factual errors (including the fact that he could probably could have avoided using a 3 1/2 year old spec in favor of a recent one), and tell him we're examining his comments for concrete solutions (or better, turn him to suggesting concrete solutions, with APIs?) Jim

For what it is worth, Gavin's mostly positive comments on JSF 2: "On the other hand, there was a lot to be gained by improving the standard option that already existed as part of the platform. There are a lot of JSF shops out there, and they are all going to get a great boost out of JSF2. And let's not forget that there are a significant number of people out there who are fans of JSF (including me, with several reservations). I've tried most of the leading Java web frameworks, and honestly, they are all very far from perfect. JSF has an overall architecture that I'm very comfortable with, so I've been willing to overlook some of the flaws. I personally do not want to go back to an action-oriented framework like struts2 or stripes. (Perhaps I am in the minority, but that's how I feel.) Now, JSF2 is the real test. We've done most of what we realistically can to "fix" it. If that's not enough, then I'm prepared to throw in the towel on JSF and put our energy into a competitor like Wicket. But I do believe that 2.0 is a big improvement, especially when you are using it with CDI. There's no piece of JSF which has not been totally refreshed in 2.0" http://www.infoq.com/news/2009/11/weld10 -Dan -- Dan Allen Senior Software Engineer, Red Hat | Author of Seam in Action Registered Linux User #231597 http://mojavelinux.com http://mojavelinux.com/seaminaction http://www.google.com/profiles/dan.j.allen