Raible Designs

Comparing Kick-Ass Web Frameworks at The Rich Web Experience

Yesterday, I delivered my Comparing Kick-Ass Web Frameworks talk at the Rich Web Experience in Orlando, Florida. Below are the slides I used:

Although it's difficult to convey a presentation in a slide deck, I can offer you my conclusion: there is no "best" web framework. I believe web frameworks are like spaghetti sauce in that everyone has different tastes and having so many choices is necessary to satisfy everyone. You can read more about the plural nature of perfection in Malcolm Gladwell's The Ketchup Conundrum (a written version of What we can learn from spaghetti sauce). Even though there is no "best" web framework, I believe GWT, Flex, Rails and Grails are frameworks that every web developer should try. They really do make it fun to develop web applications.

You can find the slides for my other RWE talk at Building SOFEA Applications with GWT and Grails.

Kudos to Jay Zimmerman for putting on a great show in Orlando this year. I had a great time talking with folks and learning in the sessions I attended. I particularly enjoyed bringing my parents and kids and staying at such a nice resort. Disney World (Magic Kingdom) and Universal Studios was very enjoyable due to the short lines. Also, the weather was perfect - especially considering the freezing cold in Denver this week.

Comparing Web Frameworks Book

A publisher recently sent me an e-mail asking some advice. They received a proposal for a book that compares CakePHP, Symfony, Zend, TurboGears, Django, Struts, RoR. Here's a quote from the proposal:

We would like to compare a couple of frameworks and present their advantages and disadvantages in various applications.

Obviously, that kind of manual would be very useful for readers who are starting their 'adventures' with web applications, as it would facilitate their choosing the best framework for their particular application. The manuscript would offer a comparison of the most popular solutions (CakePHP, Symfony, Zend Framework, TurboGears, Django, Struts, Ruby on Rails) and demonstrate the main differences between each.

Therefore, the target audience would mainly be project managers, responsible for deciding on the technologies to be used for in-house projects, as well as less experienced, web application beginners.

Another purpose of the book would be to present 'good practices' in various frameworks, such as code re-factoring, design patterns and application security. From this point of view, it could become a valuable asset for experienced and learner programmers alike.

Since I got a lot of feedback from my tweet on this subject, I figured I'd ask it here.

What do you think of such a book?

Here's my response:

How do PHP books do these days? Of the list of frameworks (CakePHP, Symfony, Zend Framework, TurboGears, Django, Struts, Ruby on Rails), I think there's interest in Django and Rails, but not so much the others. And Struts sucks, so having that as a comparison is obviously going to make it look bad. I wouldn't buy it, but I'm a Java guy that's mostly interested in web frameworks that make developing SOFEA-based applications easier. In my mind, these are Flex and GWT.

The book I'd like to see would cover developing RESTful backends and SOFEA front-ends. RoR, Grails or Django could be used to develop the backend and Flex, GWT and X could be for the front-end. In reality, this is probably a tough book to write b/c things move so fast. If you decide to do it, I'd keep it short and sweet so you can get it to market and update it quickly.

Dojo/Comet support in Java Web Frameworks

This week I'm doing a research project for a client. The main purpose of the project is to find out which Java-based web framework works best with Dojo and Comet. Here's the key requirement from the client:

It's all about Comet, we want Comet everywhere we can put it, but we want to isolate the icky bits of fiddling with pages with JavaScript. We're kind of wed to the Dojo implementation of the client-side bit, so we may as well use more of the Dojo widgets for a richer UI. For us, "works best with" needs to pay a certain amount of consideration to "fits naturally with", if you understand what I mean. I know that any framework that lets you spit out raw HTML will let you hand code in your Dojo / Comet, but that's certain to become very tiresome very quickly.

The candidate frameworks they asked me to look at are Wicket and Tapestry 5. They're willing to upgrade to Struts 2 since they're already using Struts 1. However, they don't feel that action-based frameworks naturally lead to rich UIs, so they'd prefer a component-based framework. They're currently using Seam for an administration-type application and feel it's too heavy for their customer-facing application.

Here's what I've found so far in my research. Please let me know if anything is incorrect.

• Tapestry 5 doesn't have Dojo or Comet support (Prototype and Scriptaculous are the baked-in Ajax frameworks).
• Struts 2 has old (version 0.4.3) and somewhat deprecated Dojo support. The developers seem to be in favor of removing it and promoting people hand-code Dojo instead. Struts 2 doesn't have support for Comet.
• Wicket has support for Dojo 1.1 that includes Comet support. This was written by Stefan Fußenegger and posted to the mailing list last month. I e-mailed Stefan and asked him about documentation. His response: "I lost my ambition to document it properly since I didn't receive any feedback on the mailing list. :)"

At this point, it seems that if the client really wants to use Dojo, they should use Wicket, and possibly pay Stefan to document it properly. However, they're willing to consider other options, as long as they have Comet support.

One option I thought of is to use DWR and its Reverse Ajax/Comet support. Another option would be to add better Dojo support to Tapestry 5. However, I don't think this is possible since the Prototype/Scriptaculous code is generated by the framework and would likely require a changes to switch it to Dojo.

Are there any other Java-based web frameworks that support easily creating Dojo widgets and working with Comet? Keith Donald tweeted that Spring MVC has Dojo support. However, I believe it's only for widgets and it still requires you to write JavaScript. If your framework doesn't have Dojo/Comet support, how hard would it be to add it?

Update: I also posted this question on LinkedIn. Make sure and check my question for additional thoughts from folks.

AppFuse Light 1.8.2 Released

AppFuse Light 1.8.2 is a bug fixes release that includes upgrades for Spring, Spring Security, Hibernate, Wicket, Tapestry and many others. In addition, Spring bean definitions were replaced with annotations (@Repository, @Service and @Controller). See the Release Notes for more information on what's changed since the last release.

AppFuse Light now offers 60 possible combinations for download:

• Web Frameworks: JSF (MyFaces), Spring MVC (with Ajax, Acegi Security, JSP, FreeMarker or Velocity), Stripes, Struts 1.x, Struts 2.x, Tapestry, WebWork, Wicket
• Persistence Frameworks: Hibernate, iBATIS, JDO (JPOX), OJB, Spring JDBC

If you have any questions about this release, please subscribe to the AppFuse user mailing list by sending a blank e-mail to [email protected]. You can also post questions in a forum-like fashion using Nabble: http://appfuse.org/forum/user.

AppFuse Light 1.8.1 Released: includes upgrades to Spring 2.5 and Wicket 1.3

AppFuse Light 1.8.1 is a bug fixes release that includes an upgrade to Spring 2.5 and Wicket 1.3 RC1. See the Release Notes for more information on what's changed since the last release.

What is AppFuse Light? Click here to find out.

AppFuse Light is a lightweight version of AppFuse. I was inspired to create it while writing Spring Live and looking at the struts-blank and webapp-minimal applications that ship with Struts and Spring, respectively. These "starter" apps were not robust enough for me, and I wanted something like AppFuse, only simpler.

AppFuse Light is designed to show Java Web Developers how to start a bare-bones webapp using a Spring-managed middle-tier backend and Hibernate for persistence. By default, AppFuse Light uses Spring for its MVC framework, but you can change it to JSF/MyFaces, Stripes, Struts 1.x, Struts 2.x, WebWork, Tapestry or Wicket. In addition, there's a number of extras for Spring MVC, including Velocity and FreeMarker versions, Ajax support and Acegi Security support.

This project was formerly named Equinox and has been under development since April 2004.

AppFuse Light now offers 60 possible combinations for download:

• Web Frameworks: JSF (MyFaces), Spring MVC (with Ajax, Acegi Security, JSP, FreeMarker or Velocity), Stripes, Struts 1.x, Struts 2.x, Tapestry, WebWork, Wicket
• Persistence Frameworks: Hibernate, iBATIS, JDO (JPOX), OJB, Spring JDBC

If you have any questions about this release, please subscribe to the AppFuse user mailing list by sending a blank e-mail to [email protected]. You can also post questions in a forum-like fashion using Nabble: http://appfuse.org/forum/user.

If you're a developer of one of the frameworks that AppFuse Light uses - I'd love a code review to make sure I'm "up to snuff" on how to use your framework. I'm also more than willing to give commit rights if you'd like to improve the implementation of your framework.

Live demos are available at:

• Spring · Spring + Ajax · Spring + Acegi
• MyFaces + Facelets
• Stripes
• Struts 1 · Struts 2
• Tapestry
• WebWork
• Wicket

What's on tap for AppFuse Light 2.0? Here's what I'm hoping to do:

1. Drop the seldom-used persistence frameworks: JDBC, JDO and OJB.
2. Drop Struts 1.x and WebWork as web frameworks (replaced by Struts 2).
3. Support the same persistence frameworks as AppFuse: Hibernate, iBATIS and JPA.
4. Re-use appfuse-service, appfuse-hibernate, appfuse-ibatis and appfuse-jpa in AppFuse Light. I'll likely include the core classes (User, Role) since AppFuse Light is more "raw" than AppFuse.
5. Require Java 5.

Let me know if you disagree with any of these items or would like to see other enhancements.

AppFuse Light 1.8 Released

AppFuse Light 1.8 adds CSS Framework integration, as well as support for Stripes (1.4.2) and Wicket (1.2.6). It also has significant upgrades for JSF and Tapestry; to versions 1.2 and 4.1.3 respectively. See the Release Notes for more information on what's changed since the the beta release of 1.8.

What is AppFuse Light? Click here to find out.

AppFuse Light is a lightweight version of AppFuse. I was inspired to create it while writing Spring Live and looking at the struts-blank and webapp-minimal applications that ship with Struts and Spring, respectively. These "starter" apps were not robust enough for me, and I wanted something like AppFuse, only simpler.

AppFuse Light is designed to show Java Web Developers how to start a bare-bones webapp using a Spring-managed middle-tier backend and Hibernate for persistence. By default, AppFuse Light uses Spring for its MVC framework, but you can change it to JSF/MyFaces, Stripes, Struts 1.x, Struts 2.x, WebWork, Tapestry or Wicket. In addition, there's a number of extras for Spring MVC, including Velocity and FreeMarker versions, Ajax support and Acegi Security support.

This project was formerly named Equinox and has been under development since April 2004.

AppFuse Light now offers 60 possible combinations for download:

• Web Frameworks: JSF (MyFaces), Spring MVC (with Ajax, Acegi Security, JSP, FreeMarker or Velocity), Stripes, Struts 1.x, Struts 2.x, Tapestry, WebWork, Wicket
• Persistence Frameworks: Hibernate, iBATIS, JDO (JPOX), OJB, Spring JDBC

If you have any questions about this release, please subscribe to the AppFuse user mailing list by sending a blank e-mail to [email protected]. You can also post questions in a forum-like fashion using Nabble: http://appfuse.org/forum/user.

If you're a developer of one of the frameworks that AppFuse Light uses - I'd love a code review to make sure I'm "up to snuff" on how to use your framework. I'm also more than willing to give commit rights if you'd like to improve the implementation of your framework.

Live demos are available at:

• Spring · Spring + Ajax · Spring + Acegi
• MyFaces + Facelets
• Stripes
• Struts 1 · Struts 2
• Tapestry
• WebWork
• Wicket

Yes, I realize that 60 combinations is ridiculous. I didn't create the frameworks, I'm just integrating them so you don't have to.

Unfortunately, it's a real pain to create Maven archetypes or they'd all be as easy as mvn archetype:create. Rumor is that the archetype plugin will allow you to create-from-project in the future. When that happens, I'll make sure all the combinations are available as archetypes.

Does Struts 2 suck?

As far as I can tell, Struts 2 sucks. To be fair, so does Stripes. Why? Because there's no developer feedback for invalid properties or OGNL Expressions. What does this mean? It means if you fat-finger a property name, nothing happens. The OGNL exception is swallowed and you never know you did anything wrong. Furthermore, no one seems to care. The XWork folks will help you build, but not solve the problem. This seems like a major deal-breaker to me, However, I also believe it can be fixed - so maybe there's hope.

To demonstrate the problem, I did an experiment. I used the "user details" page in AppFuse Light to fat-finger a property name for the following frameworks: Struts 1, WebWork, Struts 2, JSF, Spring MVC, Stripes, Tapestry and Wicket. First, I tried changing the "lastName" property to "LastName" to see if the framework's property evaluation was case-sensitive. I found that with WebWork/Struts 2, Stripes and Tapestry, the property is not case-sensitive. I prefer case-sensitivity, but maybe that's because I prefer Unix over Windows.

The 2nd thing I tried was changing "lastName" to "pastName" to see if I'd get an error. An error occurred for all the frameworks mentioned, except for WebWork/Struts 2 and Stripes. This makes me believe these frameworks suck. The both use OGNL, so they could blame it on that, but Tapestry uses OGNL and it presents an error message. After this small experiment, my conclusion is the following frameworks have the best developer feedback:

• Struts 1
• JSF
• Spring MVC
• Tapestry
• Wicket*

* Wicket seems like it needs some work as all it presents is "Internal Error" and makes you dig through your log files to find the problem.

Without good developer feedback, how can you have good productivity?

Dear Struts 2 and Stripes Developers,

What do you think about improving your error messages for invalid properties and expressions? Is this a feature you think you could add? We'd love it if you did.

Sincerely,

Your Users

Click here for some screenshots of how a fat-fingered property looks in various frameworks:

JSF

Struts 1

WebWork

Struts 2

Spring MVC

Tapestry

Stripes

Wicket

Update: Stripes doesn't suck and Wicket has excellent error reporting. See my comment below for more details.

Update 2: I've created a patch to (hopefully) solve this issue in XWork. If you have any feedback on ways to improve this patch, I'd love to hear about it.

AppFuse 2.0 RC1 Released

The AppFuse Team is pleased to announce the release of AppFuse 2.0 RC1! This release marks a huge step in the march to releasing AppFuse 2.0. This release puts the finishing touches on the AppFuse Maven Plugin (AMP), which offers CRUD generation, as well as the ability to change AppFuse from "embedded mode" to "full source" (like 1.x). In addition, we've addressed over 100 issues in preparation for the final 2.0 release. We hope to fix any bugs related to this release and release 2.0 Final in the next week or two.

The videos still represent how M5 works, but things have been simplified (now you don't need to run appfuse:install after appfuse:gen).

AppFuse 2.0 is available as a Maven archetype. For information on creating a new project using this release, please see the QuickStart Guide or the Hello World video.

If you've used AppFuse 1.x, but not 2.x, you'll want to read the FAQ. Join the user mailing list if you have any questions. The Maven Reference Guide has a map of Ant » Maven commands. Maven for Newbies might also be useful if you've never used Maven before. There is some support for Ant in this release.

For more information, please see the 2.0 RC1 Release Notes. The 2.0 series of AppFuse has a minimum requirement of the following specification versions:

• Java Servlet 2.4 and JSP 2.0 (2.1 for JSF)
• Java 5+

We appreciate the time and effort everyone has put toward contributing code and documentation, posting to the mailing lists, and logging issues.

We also greatly appreciate the help from our sponsors, particularly Atlassian, Contegix, JetBrains, and Java.net. Atlassian and Contegix are especially awesome: Atlassian has donated licenses to all its products and Contegix has donated an entire server to the AppFuse project. Thanks guys - you rock!

Comments and issues should be posted to the mailing list.

Update: I've uploaded a 247-page PDF version of the RC1 documentation to java.net. This PDF contains the relevant pages from the wiki that help you develop with AppFuse 2.0. Who knew I'd end up writing another book?

Open Source Web Frameworks' Mailing List Traffic - June 2007

Who knows if these stats mean anything, but it does make a pretty graph. Current mailing list traffic leaders in the web framework space: Rails, Flex and GWT. For those frameworks with dev and users lists, these stats are from the users lists. If you find these numbers to be inaccurate, please let me know.

Here's the numbers in case you want to create your own graphs:

• Rails: 4056
• Flex: 3558
• GWT: 2305
• Django: 1951
• Wicket: 1718
• Struts: 1689
• Grails: 1307
• MyFaces: 1283
• Tapestry: 1268
• TurbyGears: 797
• Stripes: 206
• OpenLaszlo: 189

How popular is your web framework?

From the Struts user mailing list:

Since its release in June 2001, Apache Struts has become the most popular web framework for Java. Six years later, by any objective measure, Struts is still Java's most popular web framework.

In February and March 2007, the group released both Struts 1.3.8 and Struts 2.0.6 to the general public, and Struts downloads zoomed to over 340,000 a month from the Apache site alone. And this is just the tip of the iceberg. Most copies of Struts are downloaded from an network of mirrors or obtained from Maven repositories.

So how popular is Struts compared to the other heavy hitters like Spring and Hibernate? Spring has about 1/2 as many (80K) downloads in the same period and so does Hibernate. How do MyFaces, Wicket and Tapestry stack up? Here's their best download numbers in the past few months:

• MyFaces: ~12K (and that's only because of a rate 3-times-normal spike)
• Tapestry: ~12K
• Wicket: ~10K

Sorry JSF, you appear to be losing. Badly. This is an incorrect statement as pointed out by commentors. Thanks for keeping me honest guys.

Disclaimer: Yes, I realize that these statistics are not very accurate, especially considering Maven. Unfortunately, until Maven has repository download stats, this information is the best we've got.