In the Spotlight - Matt Raible

At the end of the first week after the LinkedIn layoffs, we all had individual opportunities, but we also had two team opportunities. The following week (last week), I flew to NYC to meet with one potential client while the other potential client flew to Denver to meet with the rest of the team. After flying to NYC, I traveled to Mountain View to do some on-site work at LinkedIn. At the end of the week, it seemed like most of the remaining tasks at LinkedIn could be done by someone else. I told them I thought it was best that I move onto other things, while staying available for support questions. On the way to the airport, I spoke with both our team opportunities. Following those conversations, I was very pumped about both projects and confident about pending offers. You can imagine my disappointment when my flight was delayed for 5 hours.

After a fun weekend with Abbie, Jack and friends, I woke up Monday morning without a job and it felt great. However, things changed quickly. Monday morning many opportunities landed in my inbox: a 3-day gig this week (helping write open-source training), a 1-week gig in December (evaluating how well Tapestry 5, Wicket and Struts 2 integrate with Dojo/Comet for a client in Europe), a 1-week training gig in Europe next year and a 3-month gig for the whole team. I accepted all these opportunities and am very happy I'll get to work with Jimbo, Country and Scotty again next year. The 3-month gig should be a lot of fun. We're helping build a SOFEA-based architecture that leverages appropriate client technologies (to be determined) to build a kick-ass web application. I look forward to talking about the technologies we use and things we learn along the way.

So the good news is I've entered The Golden Period. The Golden Period is when you don't have a job, but you do have a start date. Unemployment is absolutely blissful during this time. The Golden Period exists a couple times for me over the next 6 weeks.

I'll be traveling to Costa Rica tomorrow for a best friend's wedding. I'm leaving both my laptop and my iPhone at home. Next week, I'll be loving life with my parents in Costa Rica and Panama. The following week, I'll be working on AppFuse all week and hope to make great progress on developing 2.1. Then I have the 1-week Web Framework Analysis gig, followed by 2 weeks of vacation in Oregon. My Golden Period begins this afternoon (for 3 weeks) and happens again over Christmas (for 2 weeks).

Yeah, life is good. Damn good.

LinkedIn's top investor[1] is Sequoia Capital and they recently posted this presentation on the web.

http://www.slideshare.net/eldon/sequoia-capital-on-startups-and-the-economic-downturn-presentation?type=powerpoint

Notice the reduce head count recommendations. ... Oh well, life goes on.

Raible

[1] http://www.linkedin.com/static?key=investors

So, as of today, there is no LinkedIn Denver office. While I had a lot of fun being a UI Architect and managing Engineers, I'm somewhat happy this has happened. After all, now I get to enjoy the best perk about being an employee: the good ol' severance package!

If you're looking for good Engineers, I highly recommend all of the guys who worked for me during this journey. You can read more about the skills they possess and what they're looking for by viewing their LinkedIn Profiles:

Scott Nicholls

Bryan Noll

James Goodwill

As for me, I'm definitely in the market as well. You can view My LinkedIn Profile to see if I might be a good fit for your organization. I'm willing to travel up to 25%, but would prefer not to. After all, ski season is right around the corner.

Lastly, I just wanted to say that I really enjoyed working at LinkedIn. I've never worked with a smarter group of Engineers, nor been so excited about a company's product and vision. I know that LinkedIn will be highly successful and I hope to use their site to find gigs for many years to come.

AbstractTransactionalJUnit4..Tests vs. AbstractTransactionalDataSource..Tests
I've switched from my favorite Spring class to the annotation-happy AbstractTransactionalJUnit4SpringContextTests. However, this has presented an issue: when using ATDSSCT, I was able to call endTransaction() and startNewTransaction(). With ATJ4SCT, this doesn't seem possible. Below is a screenshot of the diff on a test method in the JPA implementation of UserDaoTest:

On the right, you'll notice that I had to comment out @ExpectedException to get the test to pass. This concerns me since this exception should be thrown. Is there a way to call endTransaction() and startNewTransaction() when subclassing AbstractTransactionalJUnit4SpringContextTests?

Instantiating GenericDao Implementations Programmatically
The second feature I tried to add is the ability to instantiate a GenericDao programatically rather than requiring a XML bean definition. In current versions of AppFuse, you can use the following bean definition to create a GenericDao for a model object.

<bean id="personDao" class="org.appfuse.dao.hibernate.GenericDaoHibernate">
    <constructor-arg value="org.appfuse.tutorial.model.Person"/> 
    <property name="sessionFactory" ref="sessionFactory"/>
</bean> 

When moving to a no-XML required architecture, it'd be nice to allow users to create GenericDao's programmatically. Below is the easiest way I've found to do this in a test:

GenericDao<User, Long> genericDao;
@Autowired
SessionFactory sessionFactory;

@Before
public void setUp() {
    genericDao = new GenericDaoHibernate<User, Long>(User.class);
    genericDao.setSessionFactory(sessionFactory);
}

However, there's a couple problems with this. First of all, mixing constructor injection and setter injection probably isn't a good idea. Changing the constructor to take a SessionFactory solves this problem, but now all subclasses need to have a more verbose constructor:

@Autowired
public UserDaoHibernate(SessionFactory sessionFactory) {
    super(User.class, sessionFactory);
}

Whereas before they had:

public UserDaoHibernate() {
    super(User.class);
}

In an ideal world, I could call new GenericDaoHibernate<User, Long>(User.class) and the SessionFactory would be wired in auto-magically. Is this possible with Spring 2.5?

The 2nd problem this presents is your client code will now be dependent on an implementation rather than the interface. I don't know how to solve that one, but I'd love to figure out a way to create GenericDaos with no XML and no implementation details in the client. Any ideas are most welcome.

If you'd like to see all the changes I made in converting from XML to Annotations, please see this patch.

Photo by Toshio

It looks like Jason is going to be doing a similar session in a full-day format at ApacheCon next week. The good news is it's free, the bad news is you have to be in New Orleans. Hopefully Jason and team will post their slides from this event and we can all learn how to use Maven more effectively.

In other Maven-related news, I hope to complete AppFuse 2.1 around the same time as Spring 3.0 (early next year). As part of that release, I hope to start using Nexus for AppFuse's repository as well as restructure its archetypes to allow using m2eclipse (with WTP) more effectively.

I believe the major mistake we made in AppFuse 2.x was making it easy for user's to upgrade their applications. We currently use the maven-war-plugin and our own maven-warpath-plugin to make it possible to include AppFuse classes and assets in your project. You can easily start a new project w/o having a whole bunch of files in your project. The problem is, you can't easily use "mvn jetty:run" to work on your project. Of course, you can use "mvn appfuse:full-source" to solve this, but I'm starting to think more and more that "full-source" should be the default. This is what we did in 1.x and it seems to be the more natural pattern for folks using AppFuse.

That hard part about moving to "full-source" by default is coming up with a way to share common assets and classes among wars and war projects. Sure, I can copy all the shared images, css and js into each project - but that could become a maintenance nightmare. Subversion 1.5 with relative svn:externals might solve this, but it still seems kinda hacky. I don't want to use the maven-war-plugin because the overlay is kinda hokey and I think it's easier for users to understand when everything is in their project. AppFuse's current directory structure in SVN looks as follows. I've added indicators of what is in each directory.

Rather than using AppFuse's current (manual) archetype-creation process, I'd like to move to a more automated creation process using the maven-archetype-plugin's create-from-project feature. I'd like to figure out a way where I can have the source code and assets from web/common included in each of the other web/* projects (both when using "jetty:run" and "archetype:create-from-project"). One idea I thought of is to make Jetty/Maven aware of multiple src/war directories for "jetty:run" and "package" and then somehow hook into the archetype plugin at creation time to pull in the shared resources. I don't know if something like this is possible. If you know of a good solution to this shared web assets issue, I'd love to hear about it.

Back to AppFuse Light. If I can figure out how to solve shared resources in web modules, I can use this in AppFuse Light to move to a modular SVN structure vs. its current "use Ant to create different combinations" setup. If a modular structure (like appfuse/web/*) is possible for AppFuse Light, I believe it makes sense to move its source into AppFuse's SVN repository. Below is how the directory structure might look after this move.

With this addition and "archetype:create-from-project", we should be able to create all the basic and light archetypes automatically. We'll probably still need a manual archetype-creation process for modular archetypes, but I'm OK with that.

The last thing I'm struggling with is figuring out the best way to create archetypes for something like AppFuse. In the past, we've used dependencies to allow users to inherit dependencies and their versions. This works, but it results in a lot of duplicate XML (in projects and archetypes) for developers. Last night, I tried using a parent project instead of dependencies and it seems to work much better. Not only do you inherit dependencies, but you also inherit plugins, profiles and properties. If you inherit, you can override, which is slick.

If you're an AppFuse user, how would you feel about having an AppFuse module as your project's parent? Would you prefer that, dependencies on AppFuse or full-source with no dependencies on AppFuse? Regardless of parent vs. dependencies, I think running "appfuse:full-source" should allow you to de-couple your project from AppFuse.