IN THE CURRENT TECHNOLOGICAL INFORMATION EXPLOSION, COMPUTERS ARE INCREASINGLY MORE AVAILABLE.

IN THE CURRENT TECHNOLOGICAL INFORMATION EXPLOSION, COMPUTERS ARE INCREASINGLY MORE AVAILABLE. AS MORE AND MORE TASKS ARE PERFORMED BY COMPUTERS, MOST INDUSTRIAL AND MILITARY ORGANIZATIONS THEREFORE FACE A PROLIFERATION OF COMPUTERS OF DIFFERENT COMPUTING POWER, MAKE, AND MODEL. TO BE ABLE TO USE A COMPUTER SYSTEM EFFECTIVELY, A PROGRAMMER NEEDS TO BE FAMILIAR WITH THE SPECIFICS IN THE OPERATION OF THE COMPUTER. MUCH OF THE KNOWLEDGE ON A COMPUTER SYSTEM IS LEARNED THROUGH TRIAL AND ERROR, AND GOES WELL BEYOND WHAT IS DESCRIBED IN THE USERS DOCUMENTATIONS. AS ONE COMPUTER SYSTEM IS USUALLY SUFFICIENTLY DIFFERENT FROM ANOTHER, THE PROGRAMMER'S KNOWLEDGE AND EXPERIENCE OF ONE SYSTEM IS GENERALLY NOT APPLICABLE TO ANOTHER. THE CURRENT STATE OF THE ART IN ARTIFICIAL INTELLIGENCE ALLOWS THE SYSTEM SPECIFIC KNOWLEDGE OF A PROGRAMMER TO BE ECODED INTO A KNOWLEDGE BASE. THE PROPOSAL ADDRESSES THE ENCODING OF THE DIAGNOSTIC KNOWLEDGE OF A PROGRAMMER. THE EXPERT SYSTEM PROPOSED WILL ACCEPT A SOFTWARE FAULT, FORMULATES A HYPOTHESIS AS TO THE CAUSE OF THE FAULT,THEN GENERATES A PLAN TO RESOLVE THE DIFFICULTY. IF A PLAN FAILS TO GENERATE AND ACCEPTABLE RESULT, A META-PLAN CONTROL OF THE REVISION OF EITHER THE HYPOTHESIS, OF THE PLAN, OR THE PLAN, OR BOTH.