This article is an interesting read if you get some time; it talks about abstractions and how they�re not always perfect. Specifically, it talks about how as time goes on, computer abstractions get better and allow programmers to do more and more amazing things, but it comes at a cost.

Leaky Abstractions

<?xml version='1.0' encoding='utf-8'?>
<tomcat-users>
  <role rolename="xxx"/>
  ...
  <user username="foo_user" password="foo_password" roles="xxx,yyy"/>
  ...
</tomcat-users>

In the web.xml file of the webapp:

    <security-constraint>
        <web-resource-collection>
            <web-resource-name>Basic Authentication</web-resource-name>
            <url-pattern>*</url-pattern> <!-- you can put as many specific patterns as you wish -->
            <http-method>GET</http-method>
            <http-method>POST</http-method>
        </web-resource-collection>
        <auth-constraint>
            <role-name>manager</role-name> <!-- this maps to a role in the user xml file -->
        </auth-constraint>
    </security-constraint>

    <login-config>
        <auth-method>BASIC</auth-method>
        <realm-name>Authentication Area</realm-name>
    </login-config>

    <security-role>
        <role-name>manager</role-name> <!-- this maps to a role in the user xml file; same as above -->
    </security-role>

Click here for the building instructions.

The completed project:

Another first was my use of the 1.5" flagstone floor tiles. These are an integral part of the "big" project I plan to make (more on this later).

And the final first was my use of cardboard for a base. Don't do it. It warps... As you can probably tell from the picture; everything is crooked. :(

Oh wait - there was another first. This was a painting technique. Around the sconce, you'll notice that the light from the flame is reflecting on the wall; it looks better in person. I'm still working on this technique, but it shows great promise.

The overall agenda is to create an enterprise level blogging 'solution' that we as a company can use. But whenever anyone starts talking about features, there is no concensus. Everyone wants different features. That's ok - features are good. But what are the core features and then what are add-on features to add later? You'd think that by the 3rd meeting that would have been decided.

And after features, let's not even talk about implementation, because no one can seem to agree on this either. Personally I think we should decide on features first.

But hey - at least we have decided on one thing: The project code-name: sayzpool. That's cool! :)

If/when we decide on anything contcrete, i'll write more.

Click here

In the matrix. Check it out. It talks about how the focus of cosmology is starting to shift from the "WHAT" to the "WHY".