Tragically L33T

Seth Godin is a really smart guy. More importantly, he takes his ideas and does something with them.

I came across Seth in an interview he did with Merlin Mann. He talked a lot about fear and the lizard brain: the part of our brain that is only hungry, scared, selfish and horny. Its the part of our brains that ruled in high school.

It will also sabotage us if we let it.

Whenever we get close to completing something big or interesting or important to us, the lizard brain senses a threat. “What if we fail?” “Ridicule is painful!” “Better to go along in life quietly, then do something that might get us noticed.”

The lizard brain tells us to quit when success is near. Seth mentioned in his interview that he listens to his lizard brain… and then does the opposite. If his lizard brain is quiet, he probably isn’t doing anything important. If his lizard brain is screaming bloody murder, he knows he is on the right track.

I picked up a couple of Seth’s books, including Tribes, The Dip, and his new book, Linchpin. As I come across more interesting tidbits I will share them.

In the past I have not blogged very often. In fact I seem to blog less often than Wordpress releases a security patch. This was making me nervous and, combined with the issues of writing posts offline at events like NFJS, I decided a change was in order.

Enter Jekyll, the static page blog generator behind Github Pages. So far the workflow of managing text files in a Git repository is working well for me. Not being able to leave well enough alone I created a Rakefile to manage certain tasks like create a tagcloud for the sidebar, creating tag specific pages listing posts, and creating a draft post.

Creating a draft post was pretty straightforward:

desc 'create a new post in draft mode'
task :new => [:require_input] do
  title = ask("Title: ")
  filename = title.downcase.gsub(/[^a-z0-9]/,"-")
  template=File.read "lib/post_template.markdown"
  File.open("_drafts/#{filename}.markdown", 'w+') do |f| 
    f << template.gsub(/POST_TITLE/, title)
  end
  sh "git add _drafts/#{filename}.markdown"
end

Publishing a draft to the blog will consist of a git mv of the draft file to the _posts directory with the data appended to the filename.

As for comments I have switched over to Disqus, which allowed me to import my Wordpress comments and link to them on my Jekyll blog.

So I have been working on my project, but Tapestry has been kicking my butt. I have decided to step back and reset. I have useable DAOs and crappy DTOs and not much else.

So I am resetting. Grabbing the code that works, dropping Tapestry (the documentation on the IoC container and testing is not great) and going back to basics. Hibernate, Spring, Sitemesh…

Yesterday (11/1) I:

• Created a basic project structure using a maven archetype
• Checked that structure into source control
• Setup a Hudson job to run continuous integration

Today I have a light load (since it falls to me to feed the wife). I want to get my first page to render in Tapestry and continue research into the elements I will need like custom dispatching.

This is going to be harder and easier than I thought.

So Tapestry supplies its own IoC container, so no Spring at this point. Also there is a nice Hibernate integration, so I am starting to mess with my entities and figure out the needed annotations.

Maven profiles seem to manage the different configuration aspects of my application such as properties for development versus a Hudson build. So I can find it in the future, here is the profile snippet for development with a in memory database:

<profile>
    <id>development</id>
    <properties>
        <hibernate.dialect>org.hibernate.dialect.HSQLDialect</hibernate.dialect>
        <hibernate.connection.driver_class>org.hsqldb.jdbcDriver</hibernate.connection.driver_class>
        <hibernate.connection.url>jdbc:hsqldb:mem:autoblog</hibernate.connection.url>
        <hibernate.connection.username>sa</hibernate.connection.username>
        <hibernate.connection.password></hibernate.connection.password>
        <hibernate.connection.pool_size></hibernate.connection.pool_size>
        <hibernate.connection.autocommit>true</hibernate.connection.autocommit>
        <hibernate.cache.provider_class>org.hibernate.cache.HashtableCacheProvider
        </hibernate.cache.provider_class>
        <hibernate.hbm2ddl.auto>create-drop</hibernate.hbm2ddl.auto>
        <hibernate.show_sql>true</hibernate.show_sql>
    </properties>
</profile>