A Quick GOFAI Intro For Young Ladies
The new Artificial Intelligence has one programming flaw, it’s open ended reasoning leads to blurred aims.
The good old fashioned Artificial intelligence (GOFAI), had two flaws in its beauty.
It was unable to have great ideas, and any one type of intelligence was not enough.
You have a great advantage now, use the new intelligence. to combine the old intelligence solutions to create focused mechanical minds.
At first, to learn, explore and master, focus the minds on managing the flow and state of web applications.
When asking your favorite AI for specific GOFAI type code examples, ask for both XML based configuration and JavaScript implementation.
XML reads like a folder tree, it will help you see and manage the mind.
Some code editors even allow you to collapse XML nodes, like in a mind or concept map.
I repeat, any one conventional AI system on its own is not enough, use state of the art artificial intelligence to unite several GOFAI solutions.
Use XML for configuring the AI, use JavaScript for the engine, if you choose to build a framework ask AI for plugin architecture.
This way it will be able to implement the most useful AI first, and leave the extra-fancy stuff for the future.
Ask you AI to consider making the central applicaion object, by simply extending and EventEMitter.
And implement the plugins through a simple applicaion.use(pluginObject);
Finally, I will list the names of the old AI’s, highlighting that the Blackboard pattern does not require any code.
Expert Systems, Finite State Machines, Decision Trees, Blackboard Systems, Case-Based Reasoning, RETE / Production Rule Matching
Use your AI, to connect them together, and form a local-first web application intelligence.
Links
- GOFAI https://en.wikipedia.org/wiki/GOFAI
- build a framework files/gofai-primer.txt
- Expert Systems https://en.wikipedia.org/wiki/Expert_Systems
- Finite State Machines https://en.wikipedia.org/wiki/Finite_State_Machines
- Decision Trees https://en.wikipedia.org/wiki/Decision_Trees
- Blackboard Systems https://en.wikipedia.org/wiki/Blackboard_system
- Case-Based Reasoning https://en.wikipedia.org/wiki/Case-based_reasoning
- RETE / Production Rule Matching https://en.wikipedia.org/wiki/Rete_algorithm