...connolly99^1.1

The other limitations are: poor representation of 'real world' entities, semantic overloading, poor support for integrity and enterprise constraints, homogeneous data structure, limited operations, impedance mismatch, problems with concurrency, schema changes and navigational access. They are mainly due to the restricted data model.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... capabilities^2.1

This problem is addressed by Object-Relational Database Systems and partially by SQL3 [7].

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... process^6.1

It reminds me of the EMH: the Emergency Medical Hologram introduced in StarTrek Voyager series. It was a computer simulation which was able to mimic the behavior of a physician. Every time the EMH was started, he requested: ``Please state a nature of the medical emergency''. So, it demanded to specify the goal for its AI inference subroutines.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...res/1^6.2

The algorithm is pretty ineffective, from the complexity point of view, however it has a nice form.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... View^8.1

I would like to thank Joanna Pacyna for giving a unique name of Jelly to the project. The name resembles the nature of the problem and the solution. A jelly is: smooth and tasty and easy applicable. What is more, a Jelly View glues the relational databases and the inference engine, and last but not least it mixes well with Computer Chips and perhaps even with Java Beans (beans-and-jelly sounds strange, but who knows, it could be tasty, after all turkey with cranberries is great)

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... knowledge^8.2

Easy modification of extensional knowledge is already provided by the database.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... rules^8.3

The order makes difference when it comes to inference and if the inference process is to be controlled or altered.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... view^8.4

This statement is not entirely true as it will be seen later on. The statement which refers to such a generated view will look rather as a function, which is used in FROM clause

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... schema^9.1

There is an inconsistence concerning SQL. SQL allows to choose one or the other referral type, but not in all cases. For example, in case of SELECT statement, a list of attribute names has to be provided. Using their positions is not allowed. Meanwhile, in the same statement in ORDER BY clause, there is freedom of choice. A column could be referred to using the attribute name or the attribute position in the schema.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... is^9.2

Some implementations require to specify the type of a term i.e. Turbo Prolog by Borland.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...completeness^9.3

The analysis is not covered by this dissertation, however it may be provided, see [20,19]

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... Microsoft^10.1

Ironically, Open is just an empty phrase here, because ODBC remains tied up to Microsoft and it is not open at all.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... software^11.1

This requirement could have been be relaxed in favor of the previous ones

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... PostgreSQL^11.2

PostgreSQL is used for all tests and examples here.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... overhead^12.1

The actual machine did not serve any other services except for the ReDaReS system and the database.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... Project^13.1

Research Project KBN 8 T11C 019 17, see http://regulus.ia.agh.edu.pl

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.