John Ferguson Smart has announced JSF Jumpstarter, an online book with a small yearly subscription. The book attempts to give readers a simple introduction to workable JSF, using Exadel Studio Pro (now part of JBoss Tools) as an IDE. The yearly subscription from the book's home page is $9.95USD per year; the announcement implies that this is an introductory subscription rate. The book doesn't go into great exhaustive detail into the entire specification, from the table of contents. However, a gentle and effective introduction to JSF is still needed, so a book like this might be entirely appropriate. As of 8:20 a.m. EST, the sample chapter isn't available - access denied - but that should be fixed soon.

Sample chapter gives 404 message.

Any recommendation for a JSF (paper) book? I'd like to learn the JSF basics within a weekend but don't mind if the book contains additional advanced information. It should explain basic concepts and not merely provide a cookbook with screenshots.

Any recommendation for a JSF (paper) book? I'd like to learn the JSF basics within a weekend but don't mind if the book contains additional advanced information. It should explain basic concepts and not merely provide a cookbook with screenshots.

Have you looked at The Complete Reference series on JavaServer Faces? It's available on Amazon.

I'd like to learn the JSF basics within a weekend but don't mind if the book contains additional advanced information.

Take a look at http://jsftutorials.net. It has a "Golden Collection for JSF Newbie" section at the top. The links to JSF book's free samples are also there: http://www.jsftutorials.net/jsf-books/free-chapters.html. You can look at them and pick up the one you want to buy.

The JSF reference is actually very good. Core JSF second edition is also out.

The JSF reference is actually very good. Core JSF second edition is also out.

I honestly can't stand JSF. I thought it would have been dead a year or so ago, but I guess I was wrong. Frameworks like Wicket, Tapestry 5, Struts 2.x, WebWork, Stripes, RIFE, etc. are much more productive than JSF. Best Regards, Richard L. Burton III

JSF isn't so bad when you put Seam on top of it. Mind you, I wouldn't mind seeing Seam ported to some other frameworks in order to not have to deal with JSF, though it would have to reinvent a fair bit of what JSF already does for it. I find few other frameworks aside from Tapestry that make form handling as powerful as JSF. On the other hand I find few other frameworks that make complex workflow as obnoxious and tedious as JSF -- but that's one area that Seam addresses handily. How is Struts 2.x significantly different than JSF? (Actual question, not rhetorical -- I was under the impression they were kissing cousins).

"JSF isn't so bad when you put Seam on top of it.". Right. And I thought hierarchies were great for genealogy representations. What else do you want on top of Seam? Sourcream?

How is Struts 2.x significantly different than JSF? (Actual question, not rhetorical -- I was under the impression they were kissing cousins).

Struts 2 is more or less a re-named WebWork. It is the culmination of the WebWork/Struts merger, and (if memory serves), uses the WebWork "guts" with a Spring IOC engine underneath. So, more of an MVC II based framework than a component based. From the struts website: "Apache Struts 2 was originally known as WebWork 2. After working independently for several years, the WebWork and Struts communities joined forces to create Struts2. This new version of Struts is simpler to use and closer to how Struts was always meant to be." (http://struts.apache.org/2.x/)

The JSF reference is actually very good. Core JSF second edition is also out.

I honestly can't stand JSF. I thought it would have been dead a year or so ago, but I guess I was wrong. Frameworks like Wicket, Tapestry 5, Struts 2.x, WebWork, Stripes, RIFE, etc. are much more productive than JSF.

Best Regards,
Richard L. Burton III

Oh, come on now. JSF is just fine if you get a lobotomy first. JSF 1.2 is EJB 1.0. It's just what you need for the "enterprise-level" application.

I honestly can't stand JSF. I thought it would have been dead a year or so ago, but I guess I was wrong. Frameworks like Wicket, Tapestry 5, Struts 2.x, WebWork, Stripes, RIFE, etc. are much more productive than JSF.

Best Regards,
Richard L. Burton III

Second that, JSF is probably the worst ever framework to come out of Java-land: convoluted tag-language (why would I want to learn yet another one, what's wrong with HTML?), slow, high learning curve, nonsensical stack-traces and above all bleeding and bad abstractions. I'd take Struts2, Stripes or Wicket anyday over JSF, hell I'd even take Struts 1! It is thoroughly incomprehensible that vendors keep trying to flog the dead-horse that is JSF.. / Wille Blog: Buzzword Bingo

It is thoroughly incomprehensible that vendors keep trying to flog the dead-horse that is JSF.

I'm not sure JSF is a dead horse, but the vendor buyin is because: * it's "standard" * You almost *NEED* vendor tools to work with it. It seemed to have designed FOR vendors.

I use JSF and I don't use any tools besides Eclipse + WTP. I don't see how JSF differs from other web frameworks related to this subject.

It is thoroughly incomprehensible that vendors keep trying to flog the dead-horse that is JSF..

At least face reality. You may not like it, but JSF is far from a dead horse. JSF has experienced dramatic take-up in the past year. I realise that popularity is not a measure of quality, but it is certainly a measure of lack of deadness! If you want to use HTML, use facelets with JSF. Extensibility is one of the huge advantages of JSF.

It is thoroughly incomprehensible that vendors keep trying to flog the dead-horse that is JSF..

At least face reality. You may not like it, but JSF is far from a dead horse. JSF has experienced dramatic take-up in the past year. I realise that popularity is not a measure of quality, but it is certainly a measure of lack of deadness!

If you want to use HTML, use facelets with JSF. Extensibility is one of the huge advantages of JSF.

We moved to .NET technology rather than chasing framework after framework in the Java world. There was a time I was happy (perhaps excited) to learn these frameworks. Now that I have a family and a life, I made a choice to learn something we can use for our RAD approach and deliver the product on time. Why should we be learning Facelets and then Seam on top of JSF? What else are we missing from this list? JSP was/is widely used and why couldn't the JSF authors couldn't get it working well with JSP? My argument is not just confined to JSF. Java space has gotten inundated with any number of frameworks. It has almost become a fashion statement to enumerate all the frameworks that one is versed with. This is a pointless peregrination. I enjoyed working with Hibernate and Spring. However, there is no good presentation framework in Java space that is comparable to Web Forms in .NET. As someone said in an earlier post, JSF may have experienced a dramatic take-up in the past. Well, past few years have been "this is the year for JSF". Too little too late. I moved to .NET and am quite happy with the decision. You can enjoy JSF's extensibility, Java's portability, and the great choice of application servers. In retrospect, how often did we need these? Good luck.

You can enjoy JSF's extensibility, Java's portability, and the great choice of application servers. In retrospect, how often did we need these?

Good luck.

Well... For example, I guess that the many shops that were using Orion in the past were quite happy they could very easily migrate their J2EE application to another Java AS, when Ironflare decided they just weren't going to release new versions of Orion. Portability is great too. In our company some people develop on Mac OS X, some on Windows, while the others simply use Linux. ASP.NET is a great technology, but you do need Windows for both development and deployment. Not all people agree that Windows is the best OS for either of those two functions. Extensibility is certainly a virtue. You don't -have- to extend JSF but if you want to do some things a little different that the standard implementation, JSF offers you lots of hooks to inject your own code into the life cycle. The fact that JSF's extensibility leads to 'frameworks' being build on it isn't just the community's response to faults in JSF, but was intentional and by-design. Of course I do agree that it's not very helpful if we would end up with 10 different alternatives for JSP, but at least most of these alternatives would be able to work with the same set of other functionalities in JSF. Many of those other frameworks in Java land often completely replace each other.

. However, there is no good presentation framework in Java space that is comparable to Web Forms in .NET

Then you've not done Echo2 (or the like). I've done plenty of Web Forms. For small apps with not too complicated functionality it (Web Forms) is fine. Can it do more? Sure. But with split classes and the page paradigm, it doesn't come close to Echo2.

Web Forms are working well for us. What complicated functionality were you referring to that Web Forms can't do the job? If Echo is that good, how come it didn't catch on? I am not doubting the functionality that Echo has to offer. I am just wondering why it didn't catch on. For us to embrace a technology, we would like to see good documentation and books and industry-wide acceptance. We don't look forward to bleed by adopting cutting-edge technologies. Java server-side technologies are very robust and have been happily embraced for building large applications (banking applications, for example). The applications we build are not complex at such level and hence .NET has been excellent. I would have loved to use JSF, stay in Java world and bank on my years of Java expertise. Synergy is lacking in Java space. How long does it take to have industry-level JSF components that .NET has? I saw this space getting almost polluted with so many frameworks and I made a decision to move on. May be, I am wrong in moving to .NET. It has been alright thus far. We will see. Someone mentioned in an earlier post that portability is a good thing to have. They can develop on Macs, Unix, and Windows. I am not denying it. It is not something that we are overly concerned with.

Web Forms are working well for us. What complicated functionality were you referring to that Web Forms can't do the job?

I am sure it is working well for you. And I didn't say it could do the job.

If Echo is that good, how come it didn't catch on? I am not doubting the functionality that Echo has to offer. I am just wondering why it didn't catch on.

Struts. Fickle developers. Developers who think the Struts type model is better. People doing web start/applets instead. Too many choices (the reason you left). You name. They all contribute. The thing about web ui development - All the types have a place. Certain things are done better different ways. With Webforms - you have no choice. It cuts both ways.

For us to embrace a technology, we would like to see good documentation and books and industry-wide acceptance.

I understand that. Not sure Webforms fits the bill, but it will do.

We don't look forward to bleed by adopting cutting-edge technologies.

Echo has been around for quite a while.

The applications we build are not complex at such level and hence .NET has been excellent.

Exactly my point. It seems to be perfect for you. It has done well for some things for me too. But oddly, I have used a few Java port APIs to help me out (iTextSharp, DotLucene, NHibernate, Spring.net).

I would have loved to use JSF, stay in Java world and bank on my years of Java expertise.

Me too. I've done a few apps in JSF. I had some issues so I will reevaluate that on the next project or the redo's.

Synergy is lacking in Java space. How long does it take to have industry-level JSF components that .NET has?

JSF has quite a few. How many do you have at the JSF price. :)

I saw this space getting almost polluted with so many frameworks and I made a decision to move on.

May be, I am wrong in moving to .NET.

I have one (minor) Webform app that I converted to Echo2 because of all the things available in Java and the need to integrate. The little I gained in .Net was minor compared to what I lost. I also have a large Winform app that i am converting to Swing. Why? Because there is nothing like Netbeans (or Eclipse). And the plethora of widgets at a great price is incredible.

Mark, thanks for sharing your expertise/experience. I am also keeping an eye on Web Start / Swing and also on .NET's WPF/E. HTML based webapps seem to be reaching their limitations.

You can enjoy JSF's extensibility, Java's portability, and the great choice of application servers. In retrospect, how often did we need these?

Good luck.

I can't believe that in 2007 this kind of argument is still being put forward. Even Microsoft is drifting towards a limited level of cross-platform support these days with a subset of .NET. Extensibility has been key to the success of JSF. It means that elegant systems like Facelets and Seam can be produced. It means that the same component set can be used with or without AJAX capabilities for example. And as for portability; well, I am going to state the totally obvious here. (Well, I thought it was obvious). I am currently developing a substantial project. I have been developing on XP, along with others who use Ubuntu. Our final deployment platform is Solaris. Best of all, we were free to choose our deployment platform so as to best suit the requirements, not the only one that will run the code. The answer to how often we need these things is, for me, just about daily.

Java space has gotten inundated with any number of frameworks.

I don't understand this argument. This is a sign of the health of the Java environment. If you want to go for commercial support and 'standards' pick JSF. But having choice is great. It may impose a requirement on a developer to research things on more depth, but that is a good thing, IMHO.

At least face reality. You may not like it, but JSF is far from a dead horse. JSF has experienced dramatic take-up in the past year. I realise that popularity is not a measure of quality, but it is certainly a measure of lack of deadness!

Well, at least it is on life-support (by the vendors). EJB 1.1 saw a dramatic uptake in use when vendors pushed it enough, didn't mean it was a good thing, just sufficiently uninformed senior managers making technical decisions they shouldn't have been making by justifying it with "strategic technical platform" or something to that effect. The fact that you need Facelets to use HTML, Ajax4Jsf to use AJAX and Seam to make it usable is just brain-dead. And that still doesn't shield you from having to know everything about the JSF lifecycle to deal with everyday problems that should be really simple. / Wille
Blog: Buzzword Bingo

The fact that you need Facelets to use HTML, Ajax4Jsf to use AJAX and Seam to make it usable is just brain-dead.

I don't agree. I like the fact I have freedom to choose technologies to work with JSF. I like these different ideas that comes from community. Some of Seam, Ajax4JSF etc ideas will have its place in some new JSR and become stadard (WebBeans...). Just like Hibernate enabled EJB3. And this would not be possible if JSF is not open or if it "dead horse" as someone said. As of comparison to EJB1, I cannot remember we had such healthy ecosystem with EJB (1, 2) as we now have with JSF.

And that still doesn't shield you from having to know everything about the JSF lifecycle to deal with everyday problems that should be really simple.

This is not true. I was able to start and complete my first project using Seam without being JSF expert.

As of comparison to EJB1, I cannot remember we had such healthy ecosystem with EJB (1, 2) as we now have with JSF.

EJB1 & 2 had a much stronger "ecosystem" than JSF has today. There are always people who like to solve problems in more complicated (than necessary) ways and more "innovative" (i.e. fashioned) ways. And these technology serves these people well. But they don’t stick around once becoming out of fashion.

Agreed. The only reason JSF is not dead is because it is bundled with jee 5. When you compare JSF with newer webframework such as wicket or spring mvc 2.5, JSF looks like EJB 1.

The notion that you need a tool to make JSF pages is rubbish, I'm new to web development and write JSF pages with ease. Whats wrong with HTML? Nothing, but you never use just HTML, you still need to add something somewhere (either in code or in the page) for 2 way databinding and all the css and javascript to handle non-trivial user interfaces. With JSF, it's all rolled into one component tag. The claim that it is over-engineered is not without merit. It is, but only because it gives developers so many entry points to extend the framework in useful ways. One only has to look at Seam & facelets to see what can be done. Seam & Facelets (and no JSP) takes all the pain out of JSF including error reporting, and frankly our productivity is fine with it. When you consider the conversations, state management, workflow management, and security built into Seam, the sum is definitely greater than the parts making it one cohesive framework. I don't believe you can have a framework on Seam, because it is a framework that glues all the other elements together.

Well, personally I don't buy the idea that 'use another framework to fix framework'. for EJB2, what we really needs is Spring or EJB3, not xdoclet. for Hibernate2, what we need is Hibernate3, not Spring hibernateTemplate. for JSF, what we need is a better web framework, not rescue by Seam/Facelets ...etc. Java stack is already high and complex. More abstraction layer would just scare developers away in web 2.0 era.

or shouldnt there be set an example proving that the simple introduction to workable JSF worked...

Wow, we're all so hell bent on slamming the technology that we can't even have a rational discussion on the topic. Would anyone really pay ANYTHING, let alone $9.95/yr for a subscription to what is essentially a Tutorial ? I know that I wouldn't. If you don't like JSF, go to another Thread If you prefer echo2 or WebWork or whatever... go to another Thread and post about it there. This Thread is about a particular JSF online book. If you don't have a comment about this topic, please go to another Thread. Some of us might actually be interested in this Thread.

Wow, we're all so hell bent on slamming the technology that we can't even have a rational discussion on the topic.

Would anyone really pay ANYTHING, let alone $9.95/yr for a subscription to what is essentially a Tutorial ?

I know that I wouldn't.

If you don't like JSF, go to another Thread
If you prefer echo2 or WebWork or whatever... go to another Thread and post about it there.

This Thread is about a particular JSF online book. If you don't have a comment about this topic, please go to another Thread. Some of us might actually be interested in this Thread.

JSF is a mess and we're comparing it to other frameworks which is on point. Trust me, users of JSF will need more than a 'jumpstart', perhaps a defibrillator, to get started using it. Best Regards, Richard L. Burton III