The psiKeds Product Backlog
(C* = Complexity [S, M, L, XL, XXL, XXXL] | R* = Responsible [KR, MJ, NN] | P* = Priority/Release [TC, AG, AU, PT, IR, OS, ...])
| Label | Task | Specification | C. | R. | P. |
|---|---|---|---|---|---|
| PBL-x01 | Organic Money Datatype | Money shall be offered as own datatype which must be usable by comparations (in releations) and computtations (in functions)). | L | KR | TC-4 |
| PBL-x02 | Organic Range Datatype | Ranges are no longer treated as explicitly generated sets of definite values (which is not acceptable for larger or infinitive ranges) but as own data types. This data type must allow to exclude single values or sub ranges (because of relations) [Refactoring following the prepared description of MJ). | XL | KR | TC-3 |
| PBL-x03 | psiKeds Brass Example Databases | For all features of the release brass psikeds offer simply to use and understand example databases (which also might be read as proof of the implementation) | XL | KR | TC-1 |
| PBL-x04 | Database PKB Dump Function | For being able to debug the inference machine and its knowledge base logically, the inference machine is able to rewrite (dump) the loaded knowledge base in the eclipse xtext specific PKB format. | M | KR | TC-2 |
| PBL-x05 | Eclipse-Plugin-Editor | The xText-based PKB-Editor for eclipse is transferred into a real eclipse pluging and can be installed manually | S | KR | TC-0 |
| PBL-x06 | Eclispe-Plugin-Editor-Distribution | The Eclipse-Plugin-Edtor is distributed by the method Eclipse itself prefers to distribute its plugins | S | KR | TC-5 |
| PBL-y01 | Range to (Range|Value) -Relations | Relations operating on real 'organic' ranges can be used by knowledge engineers. | L | KR | AG |
| PBL-y02 | Range to (Range|Value) -Functions | Functions taking 'organic' ranges and/or values as arguments and one 'organic' range as result can be used by knowledge engineers: the result value is computed on the base of the 'argument' decisions [Moodus Ponens] or the last argument decision is determined by all other argument decisions and the 'proleptic' result decision | XXL | NN | AU |
| PBL-y03 | Implied Variants | Based on the concepts of a manually inserted or automatically computed (emprirically evoked) secondary concept net of a system, clusters of variants can be inserted into logical rules by using the defined concepts as premise or conclusion events. | XL | NN | AG |
| PBL-y04 | Conditioned Relations | Relations as whole can become the conclusion of a rule so that they are only taken into account if the premise of the rule has become true [Modus Ponens] or that the premise will become false, if the relation is not fulfilled [Modus Tollens] | L | NN | AG |
| PBL-y05 | Complete Path (Event) Decision | The inference machine accepts a complete path (= event) as a collection of decisions, takes all steps of this row as single decisions, and infers all corresponding conclusions | XL | NN | AU |
| PBL-y06 | Look-1-Step-Forward-Request | The inference machine takes one decision, computes the conclusions as a simulation of the decision | L | NN | AG |
| PBL-y07 | Database XML Dump Function | For being able to debug the inference machine and its knowledge base logically, the inference machine is able to rewrite (dump) the loaded knowledge base in the well known xml format. | M | KR | AG |
| PBL-y08 | Performance test | Automatic computation of a knowledge base in form of a deep tree containing all possible events in all contexts and corresponding rules [solution must be computable]. Using this knowledge base in multi query agent scenarios. | L | NN | AU |
| PBL-y09 | Dump all valid entity descriptions | Function shall 'only' be offered for the knowledge engineer and shall not become part of the deduction dialogue between resolution engine and query agent. The function shall be able to work top down and bottom up. | XXL | NN | AU |
| PBL-y10 | Complex number of PS / components | In opposite to the simple interpretation, a complex number of components shall allow that different PVs can together fulfill the n times requested PS | L | NN | AU |
| PBL-z01 | Derived Secondary Concepts | An improved support tool for knowledge engineers or the graphical knowledge base editor itself computes for each variant in the pv-ps-net all empirically evoked secondary concepts and for each system of the ps-pv-net it computes the net of all derived secondary concepts based on the primary concepts of all variants of the specific system. | L | NN | IR |
| PBL-z02 | Knowledge Engineering Tutorial | XXL | KR | OS | |
| PBL-z03 | Program Documentation | psikeds offers an administrator handout | XL | KR | OS |
| PBL-z04 | Improved HTML Query Agent | Another HTML Query Agent which do not use JSF | XXL | NN | OS |
| PBL-z05 | Graphical knowledge base editor | Instead developing psiKeds knowledge bases directly by using an xml/text Editor or one of the Xtext based generated editors, a knowledge engineer shall be able to link / to set up psiKeds elements graphically , especially the purpose systems, variants, features, concepts, events, and rules | XXXL | NN | OS |
| PBL-z06 | Graphical query agent | A knowledge base user shall interact with the resolution engine by reading and modifying resulty graphically. | XXXL | NN | OS |
| PBL-z07 | Arguing engine | The query agent and its presentor shall be able to explain why a component has been selected and why other possibilities are not offered | XXXL | NN | OS |