Raible Designs

RE: Life above the Service Tier

Yesterday I wrote the following:

I hope to develop with Flex, Grails, GWT or YUI + Struts 2 in the next 6 months. These seem like the most exciting technologies for Java web development in 2008.

This post is meant to explain why I think these are the most exciting technologies going forward.

A few weeks ago, a very interesting paper was posted on TSS: Life above the Service Tier. In this paper, Ganesh Prasad, Rajat Taneja and Vikrant Todankar introduce a new architectural style they're calling SOFEA, for Service-Oriented Front-End Architecture. To summarize:

The principles of SOFEA are:

0. Decouple the three orthogonal Presentation Tier processes of Application Download, Presentation Flow and Data Interchange. This is the foundational principle of SOFEA.

1. Explore various Application Download options to exploit usefully contrary trade-offs around client footprint, startup time, offline capability and a number of security-related parameters.

2. Presentation Flow must be driven by a client-side component and never by a server-side component.

3. Data Interchange between the Presentation Tier and the Service Tier must not become the weakest link in the end-to-end application chain of data integrity.

4. Model-View-Controller (MVC) is a good pattern to use to build the Presentation Tier.

Their paper can be downloaded from Life above the Service Tier.

I read this paper earlier this week and enjoyed reading it as well as thinking about the concepts it introduces. First of all, I believe SOFEA only applies to web applications and isn't a valid architecture pattern for web sites. While it may work for web sites, the traditional mechanisms (serving pages from the server side) seems to work well and isn't going away anytime soon.

So if SOFEA is the way of the future for developing web applications, where does that leave all the web frameworks that serve up pages server-side? This includes all Java web frameworks, Ruby on Rails and PHP. I think it leaves them in an interesting situation. They can still be usable if they can serve up the Application Download and the Data Interchange, but otherwise, they seem pretty much useless with this new architecture.

Is a SOFEA architecture a silver-bullet? I doubt it as there's still a lot of unanswered questions. How does SOFEA solve i18n and validation? Is it possible to re-use server-side validation rules in the client-side architecture? Granted this is probably a client-side framework feature rather than a SOFEA concept, but I still think it deserves some thought.

I don't know if i18n is that much of an issue for most applications. Most of the gigs I've consulted on in recent years were English-only, with no plans for internationalization. Validation is often server-side too. However, I believe server-side validation is often done simply because the web framework being used did not provide good client-side validation. How does i18n work in a JavaScript application? Can you bundle i18n scripts in the Application Download and have those read on the client-side - or do you serve up a different version of the application for different locales?

I think the most interesting part of SOFEA is how simple the backend becomes. With Spring and Hibernate (and some type of remoting) it should be easy to develop your SOA backend. But how do you publish those services? Do you still use a web framework on top to handle validation and such, or do you just markup POJOs with @WebService annotations?

Will 2008 be the year for SOFEA applications? It's definitely possible. I'm thinking of starting a Denver SOFEA user group to discuss and promote this architecture style. If I did - would you be interested in attending?

Draft Specification for automatically mapping URLs to Controllers and Views

Ted Husted has been working on a draft specification for "Heuristic Request/Response Mappings", based on technologies used by Rails, Struts 2 and Stripes.

The central idea is that instead of creating explicit mappings, the framework applies a series of heuristic transformations to match a URI with a code component and a view component.

The first part of the draft is available here:

* http://code.google.com/p/web-alignment-group/wiki/WAG_RFC_001

Before doing any more work on the description, I'd be very interested on feedback as to whether I'm making any sense, or whether the draft has turned into opaque gibberish :)

If you're developing a web framework (in any language) and use conventions to auto-map URLs-to-controllers and controllers-to-views, it'd be great to hear your feedback on this draft. It'd be pretty cool if you could switch frameworks/languages and use the same conventions across the board.

Xebia Web Framework Contest

I found an interesting blog post today about a contest (English translation) a French company (Xebia) had with some Java web frameworks.

Overall, I think it's a good summary of the strengths and weaknesses of the various frameworks.

Introduction to Grails with Scott Davis at the Colorado Software Summit

According to Scott, today's Java-based web frameworks are partial solutions at best. They're only solving one piece of the puzzle - you still need to manage persistence, deployment, etc. all by yourself.

We're moving into a new era of web frameworks. The expectation now is a full-stack solution. Grails is a fully integrated modern Java web application in a box. It contains Spring, Hibernate, Quartz, Log4J, Jetty, HSQL, JUnit and Ant. You're not limited to using Jetty, you can type "grails war" and create a WAR that you can deploy to any application server. In a single zip/tar, you get the whole thing - including the database and servlet container. You get a lot of good default for free, but you're not limited to those defaults.

What does "modern" in a framework mean? It means it uses Convention over Configuration - Grails autowires components together based on naming conventions. Struts 1.x uses Configuration over Configuration - the more XML the better.

The Wisdom of Crowds - why are the many smarter than the few? What is the wisdom of crowds when it comes to web frameworks. Struts is the wisdom of crowds when it comes to web frameworks. It's based on sound principles (MVC) and was written by Craig McClanahan. He was the architect of "Catalina" Tomcat 4 and wrote Struts shortly thereafter. David Geary was contributor #2 to Struts. It has a proven track record and has a 60%-70% market share. Struts must be the perfect framework - especially since it has such a great pedigree.

So what's wrong with Struts?

• It's 7 years old
• It's verbose and overly complex (reflective of the EJB 2.x era)
• Splintered community: Craig moved on to JSF -> created Shale, then the WebWork merger
• While Struts 1.x was an unqualified success, Struts 2.x can't seem to build that critical mass - it can't seem to reach The Tipping Point like Struts 1.x did

The Recipe for a Tipping Point: Contagiousness / viral, tiny "tweaks" to a proven model can yield big, disproportionate effects, the rise is not "slow and steady" - the effect is dramatic and immediate.

Ruby on Rails won the hype award. It's largely worth the hype, but it's not revolutionary - it's evolutionary. It has tiny tweaks to the proven MVC-driven approach: Convention over Configuration, Scaffolding and Unified Technology Stack.

The one thing that Rails is lacking is Java support. Grails, on the other hand, offers the same experience using known, proven Java solutions.

• Rails => "replacement"
• Grails => "upgrade"

Scott drank the Rails Kool-Aid for a while and enjoyed it, but found it difficult to switch from Ruby in the morning to Java in the afternoon. With Grails, he doesn't have to do as much context switching, as well as all the Java libraries are available - the ones you know and love.

You can drop groovy.jar into your classpath and write Groovy code. One nice thing about Groovy is you can rename your existing .java files to .groovy and they'll work just fine. Included Ajax support: Script.aculo.us and Protoype as well as YUI. YUI is battle-tested since it's used by Yahoo and very well documented. You can use "grails install-dojo" to install the Dojo toolkit. Grails has a wealth of plugins available at http://grails.org/plugins.

Now we're going to crack our nuckles and build some code - slides are over.

 grails create-app conference
 cd conference
 grails create-domain-class Speaker (add some fields)
 grails generate-all Speaker
 grails run-app

The impressive things about this set of commands is I was able to 1) download Grails and 2) run all these commands at the same time that Scott did. I was definitely impressed (I knew I would be). Auto-scaffolding - you can get the same thing as "generate-all", but it generates controllers and views at runtime in-memory. All you need to do is create a domain object (i.e. Talk) and then create a TalkController that has the following line in it:

def scaffold = Talk

As a test, I tried this at the same time that Scott did and got the following error. It looks like Grails/Jetty isn't smart enough to pick up new classes as they're added.

[263102] commons.DefaultGrailsApplication Class not found attempting to load class Talk
java.lang.ClassNotFoundException: Talk
        at java.net.URLClassLoader$1.run(URLClassLoader.java:200)

Grails comes with a number of environments. Dev (the default) auto-reloads changes to Controllers, Views and even the Model (this is helpful for rapid development). Prod loads all items statically for maximum performance. To change the environments - you can change DataSource.groovy.

For some reason, adding/removing some properties on my Talk object and my application hasn't been the same since. Now when I try to access my TalkController, I see the following stack trace:

org.springframework.orm.hibernate3.HibernateQueryException: could not resolve property: difficulty of: Talk; nested exception is org.hibernate.QueryException: could not resolve property: difficulty of: Talk
	at org.springframework.orm.hibernate3.SessionFactoryUtils.convertHibernateAccessException(SessionFactoryUtils.java:640)
	at org.springframework.orm.hibernate3.HibernateAccessor.convertHibernateAccessException(HibernateAccessor.java:412)
	at org.springframework.orm.hibernate3.HibernateTemplate.execute(HibernateTemplate.java:378)
	at org.springframework.orm.hibernate3.HibernateTemplate.executeFind(HibernateTemplate.java:342)
	at org.codehaus.groovy.grails.scaffolding.DefaultScaffoldDomain.list(DefaultScaffoldDomain.java:112)

I asked Scott about this error and he proved that removing properties from domain objects should work. I told him restarting Jetty didn't fix the problem and he suggested dropping the "Talk" table and letting Grails re-create it. Unfortunately, I have no idea where this database is, so that's difficult to do. Doh - now I realize what was causing the problem. Before I dropped the "difficulty" property, I had clicked on the column and that property was still referenced in the URL. When I'd refresh the browser, the stack trace occurred. I don't know if I'd consider this a bug or not.

dbCreate = hibernate.hbm2ddl.auto is used. When in development create-drop is used. In production and test environments, it uses update and saves the data between restarts.

You can turn off Hibernate's automatic schema alteration by commenting out "dbCreate" line in DataSource.groovy.

To run your application in Tomcat instead of Jetty, you can run "grails war" and copy the WAR to Tomcat. The WAR is created with the production environment by default, so you may need to pass in arguments or set environment variables if you want the WAR created in dev mode.

At this point, my battery died. Scott continued to cover how to order form elements (when using scaffolding) with "static constraints" and how to add validation rules. It was an excellent presentation and Grails definitely looks like a really cool web framework. The best part is I learned most of what you need to know to use it - in an hour!

I might have to try Grails soon - I love the concept of Life above the Service Tier and Grails would work nicely for serving up REST. I think YUI, GWT-Ext and Flex are probably the best frameworks for developing a SOFEA client. The question is - when using YUI, how do you download all pages in the application at once?

RE: Trolling with Java Web Frameworks

Fred Daoud has an excellent post titled Trolling with Java Web Frameworks. Well written Fred!

Choosing a JVM Web Framework

This morning I delivered my Choosing a JVM Web Framework presentation at the Colorado Software Summit. One of my goals for the talk was to get more audience participation and stories of how folks thought it best to choose a web framework. The room was packed and the crowd was interested, so we barely even used the slides I prepared. One of the most interesting things about the audience was that over 60% (of 50-60 people) were using Struts 1. Most of them came to learn about frameworks they might think about migrating too. Unfortunately, I didn't talk a whole lot about the frameworks and their features, but a few members had advice concerning frameworks *not* to use based on their experiences. Overall, it was a lot of fun to interact so much with the audience and hear their thoughts on web development for the Java Platform.

You can download a PDF version of my presentation from my presentations page. Thanks to all the folks who responded to my Stories Wanted post - I used many of these comments as part of the "Case Studies" in the presentation.

Colorado Software Summit - are you coming?

Are you coming to the Colorado Software Summit this year? I'm excited to go because I wrote new presentations and I think they'll be a lot of fun to deliver. Also, as I've said before, I really enjoy this conference because it's so relaxing. It's a full-week long, which is a tough commitment, but I like to think of it as a vacation. You do have to deliver your talks 3 times each, so you still have to work every day, but there's also a great opportunity to learn from other speakers. And you don't feel rushed since each talk is given 3 times. This means you can treat some days like real vacation days where you only work a couple hours and others you can pack it in and get a brain full of stuff.

Here's my Choosing a JVM Web Framework abstract?

One of the most difficult things to do (in Java web development) today is to pick which web framework to use when developing an application. A few years ago, there were over 50 Java web frameworks available, most of them open source. Since then, the number hasn't gone down, but the quality of choices has certainly improved. Should you use the standard JSF, or something like Tapestry or Wicket? What about Struts' successor ? is Struts 2 better than Spring MVC or Stripes? And what about the slick-looking applications that Flex and OpenLaszlo can create? Should you use Rails on GlassFish or Grails with Groovy? Is ZK really the next best thing? Where does RIFE fit into all of this? The choice hasn't gotten easier over the years.

This session is a discussion about choosing the best tool for the job. Not only will various frameworks and their features be discussed, but so will important factors for choosing a web framework. Is ease of development more important, or future maintenance? Is the project community an important factor? All of these questions will be discussed and answers will be provided. If you are about to choose a web framework, or if you have an opinion about a web framework, this session is for you.

I think it's important to note that this talk is going to be a discussion. I don't plan on offering my opinions as much as I plan on extracting them from others. This talk probably wouldn't work with the Norway crowd (they don't like to participate much), but I think it'll work with the Colorado folks.

If you're attending ApacheCon this year, which talk would you rather attend - Comparing or Choosing? Or maybe "choosing" would fit in better as a BOF?

Does becoming a committer hurt?

I've recently been offered the opportunity to become a committer on one of the open source web frameworks I often talk about. From a professional and career standpoint, I'd be a fool to turn it down. My clients would benefit and it would look good on my resume. However, I'm wondering if it would hurt how folks perceive me when I talk about web frameworks?

Personally, I don't think it would change my views. Instead, it would probably eliminate my complaining about said framework because its developers would say "If you have a problem with that - fix it!" What do you think - if I were to become a committer on one of Java Web Framework projects, would you still think I'm agnostic when I talk about Java (or other open source) web frameworks? Of course, it's possible you don't think I'm agnostic now, so maybe I'm worried for no reason.

Choosing a JVM Web Framework: Stories Wanted

My last post on choosing a web framework got quite a few comments. Some seemed to like the application categorization technique as a means to narrow the choices. However, others seemed to disagree. So if application categorization is not a good methodology for narrowing the choices, what is?

I think one of the best ways to figure out a good methodology is to find out what people have done to choose their web framework. I'm looking for stories from developers who have evaluated 2-3+ frameworks for a project. I'd like to come up with 3-5 stories as part of my talk to highlight how some teams have chosen their web framework. What were your important criteria? What made you choose the one you did? Was it a tight race between a few of them? Did industry buzz or application categorization play a part in your decision?

Please send any stories you'd like to share to [email protected]. Of course, you can also post your story in the comments - but an e-mail gives it a bit more validity. If you'd like to share your company name, that'd be great, but it's by no means required. I haven't decided if I'm going to prevent all cases as anonymous companies or not. If you do send a story, I'll make sure and ask your permission before I share any of your personal/company information. Thanks!

Do we even need web frameworks?

On the Struts mailing list, Ted Husted writes:

IMHO, if we had today's JavaScript/CSS/HTML environment available to us in 1998, then JSP, ASP, JSF, PHP, and all the rest of it, would not exist. We did all these things to make up for the shortcomings of the client-side environment, and, since then, the key shortcomings have been addressed.

I believe Ted has a good point. However, today's Ajax development toolkits don't help those applications that require Section 508 compliance. Of course, if your toolkit embraces 508 compliance, I'd love to hear about it.