As said in previous chapters, the agent platform is the environment where the agents are deployed. Think of it both as the "home" of the agents and the main server of a network from a distributed application. In our case, that platform is SPADE, an agent platform based on the Jabber/XMPP protocols and technology.
The SPADE platform is installed and executed on a host. That host becomes the centre of the multi-agent system that we are going to build. In order to properly deploy the SPADE platform, it needs to be configured first to suit the host where it is going to be. Along with the SPADE distribution comes a tool called configure.py which is a script to automatically configure SPADE for you. In most scenarios, the only configuration you need to do for SPADE to work is to invoke this tool with the desired hostname to use as the platform hostname. E.g:
$ python configure.py myhost.myprovider.com
Then, SPADE will be configured to work as a platform serving at the myhost.myprovider.com domain.
Internally, the configure.py script creates the configuration files that SPADE needs to work. This configuration is stored in the etc/ subdirectory inside the SPADE installation path.
After a successful configuration, you can try to run SPADE for the first time. To do so, use the runspade.py script that is provided with the SPADE distribution. This script will launch the SPADE platform and set it ready to start working with agents. You should see something along the lines of:
$ python runspade.py SPADE 2.0 <gusarba@gmail.com> - http://spade2.googleprojects.com Starting SPADE...... [done]
You may also see some warnings and additional messages while the platform is starting. That's fine, we will discuss them later. You are now ready to start loading agents into the platform.
To learn how to make your first SPADE agents, head on to the next chapter.