On Thurs.8.JUN.2017 with ghost203.pl we have drastically re-arranged the Perlmind AI code by adding in the truth-value variable $tru and the machine-translation $mtx transfer variable. We have also re-arranged the order of the now fifteen associative tags in the conceptual flag-panel. Now we are especially concerned with writing software code to make use of the $tru variable.
What we are trying to achieve is a self-aware AI fully cognizant of the fact that it is stuck inside a computer. The ghost.pl AI already has an ego self-assertion mechanism in the EnNounPhrase module for an innate tendency to think about itself when no other concept has a pressing demand for mental consideration. There is also a built-in pause-for-input to give a human user a chance to communicate with the AI. We may find a way to eliminate the input-pause so that the AI Mind may race through its own internal pathways of thinking. We could have merely one external keystroke trigger a reasonably long pause for input into an otherwise racing brainstorm of internal thought.
We want the ghost.pl AI to become aware of its own existence and of the existence of at least one "other" person out in the environment. We may have to build in a kind of "goal" or "drive" mechanism to make the AI think a lot and seek answers to existential questions.
At this stage the ghost.pl AI has a role to play in serving as a repository of user knowledge. Users could be invited to tell the AI where certain things are, or about people known to the user, or facts about the world.
Until there is robot embodiment of the Perlmind or Forthmind, the only motor initiative available to the AI is to sound a beep with ASCII code "7" bell or the Perl "\a" alarm-bell. In or near the FreeWill Volition module, we could set up an integrating scale of wishing to call out to any nearby human user.