Browsing by Author "Bradley, J."
Now showing 1 - 20 of 21
Results Per Page
Sort Options
Item Open Access APPLICATION OF SQL/N AND ASSOCIATION MODES TO DATA BASES CONTAINING RECURSIVE ASSOCIATIONS(1985-03-01) Bradley, J.Modes of association are defined for the recursive associations that occur in bill of materials data bases and data bases for the family tree of corporations. It is shown that a series of useful modes may be defined. Modes of association may be schema defined to enable a relational data base system capture more meaning about an association. When used with natural quantifiers and SQL/N, an upward-compatible extension of SQL, they permit construction of concise expressions for complex retrievals involving explosions and implosions of entities described in the data base. A mode of association is a set of propositions, each describing a pair of associated entities.Item Open Access APPLICATION OF SQL/N AND ASSOCIATION MODES TO DATA BASES CONTAINING RECURSIVE ASSOCIATIONS(1986-09-01) Bradley, J.Modes of association are defined for the recursive associations that occur in bill of materials data bases and data bases for the family trees of corporations. It is shown that a series of useful modes may be explicitly defined. Modes of association may be schema defined to enable a relational data base system to capture more meaning about an association. When used with natural quantifiers and SQL/N, an upward-compatible extension of SQL, these modes permit construction of concise expressions for complex retrievals involving explosions and implosions of entities described in the data base. Full explosions/implosions may be expressed in SQL/N, which is not possible with SQL. A mode of association is a set of propositions, but may be defined and implemented as a relationship relation.Item Metadata only A CONSISTENT DENOTATIONAL METHOD FOR THE BASIC AND NATURAL QUANTIFIERS(1984-12-01) Bradley, J.A consistent method of denoting the natural quantifiers is described. The conventional notation for the universal and existential quantifiers is preserved but is integrated into the broader denotational system for the natural quantifiers. The system permits quantifiers to be combined to form new quantifiers, and consequently can be used to simplify predicate calculus expressions that contain natural quantifiers.Item Open Access CONSTRUCTS FOR A NON PROCEDURAL OBJECT-ORIENTED DATABASE MANIPULATION LANGUAGE OSQL/N(1992-01-01) Bradley, J.Constructs for an object-oriented non procedural data base manipulation language called OSQL/N are presented. The major feature is that the language allows for conditional manipulation of an object together with a quantified set of its related objects. The constructs required for this were borrowed from a language called SQL/N, which is implicitly object-oriented, but which was originally developed as an attempt at a more natural relational data base language. The constructs behind SQL/N are much more suited to object-oriented data bases, and the OSQL/N language structure, as a result, allows for incorporation of the whole array of quantifiers that are commonly used in natural language, that is, the natural quantifiers.Item Open Access A CORE PROTOTYPE IMPLEMENTATION OF AN EXTENDED RELATIONAL DATABASE LANGUAGE FOR EFFICIENT MANIPULATION OF CONTAINMENT AND COMPOSITE STRUCTURES(1996-12-01) Bradley, J.; Rata, C.D.The core prototype implementation of an extended relational data base system is described. The full data model allows for restricted unnormalized relations. Essentially attributes that are sets or lists of atomic attributes are allowed but not attributes that are relations. There is support for both inheritance and composite entities. Manipulation languages are SQL and a separate new SQL language subset for efficient manipulation of containment structures. The language subset has a predicate calculus expression structure, but allows the use of genitive relations to model containment relationships and facilitate application of natural quantifiers to containment sets. A genitive relation is a relation that is the equivalent to the genitive case grammatical construct in natural language. In the query processor a query decomposition technique was employed. Declarative language expressions involving naturally quantified genitive relation clauses for one-to-many relationships, both composite and non composite, both recursive and nonrecursive, and for subset genitive relations, can be reduced. The natural quantifiers are the quantifiers of natural language and the prototype implements 14 basic types including universal and existential quantifiers. The query processor translates declarative language expressions to Extended Relational Algebra (ERA) routines. ERA differs from conventional relational algebra in that it has a possibility join operation, as well as two operations involving quantifiers, namely group select and subgroup select operations.Item Open Access DESIGN AND IMPLEMENTATION OF COOL, A DECLARATIVE, GENITIVE RELATIONAL, NATURAL QUANTIFIER LANGUAGE FOR OBJECT-RELATIONAL DATABASES(1995-08-01) Bradley, J.; Rata, C.D.The structure and implementation of a new kind of declarative language called COOL for object-relational data base manipulation is described and analysed. COOL is motivated by the goal of an easy-to-use object-relational declarative language that could be the basis for a language for both oral database interrogation and for manipulation of very complex objects. Fundamentally COOL has a predicate calculus expression structure, but allows the use of genitive relations and natural quantifiers, which renders the COOL expression structure remarkably close to that of English language expression structure. A genitive relation is a relation containing the equivalent of a genitive case in natural language. So far only the genitive relations for one-to-many relationships, both composite and non composite, both recursive and non recursive, have been implemented. The natural quantifiers are the quantifiers of natural language; 14 basic types were implemented. COOL expressions are translated in two steps, firstly to Extended Relational Algebra (ERA) routines, and secondly from the ERA routines into SQL for processing by a relational data base system.Item Open Access EN:M FRACTAL TIME FUNCTIONS, GRODECS AND GRODEC STACK MACHINES(1995-12-01) Bradley, J.This paper is concerned (a) with a family of fractal growth-decay time functions called En:m functions, (b) with fundamental growth-decay storage and switching components called grodecs, that have the switching properties of neurons, and (c) with machines constructed from stacks of grodecs, called grodec-stack machines. An En:M function consists of a series of replicating and endlessly subdivided alternating growth and decay segments. Grodecs could be used to build computing machines that work like biological information processing organs but that aspect of grodec machines is not analysed. The grodec stack machines presented and analysed exhibit time fractal growth-decay En:m function behaviour. Left-sink grodec-stack machines can exhibit En:1 function behaviour where only growth segments subdivide, and right sink grodec stack machines can exhibit negative En:1 function behaviour where only decay segments subdivide. These two types of machines can be combined to form balanced grodec-stack machines with En:m function behaviour - where both growth and decay segments subdivide. It is shown that only resonating En:m functions can be generated by such machines. The behaviour of grodec stack machines can be unfathomably complex; they can exhibit sensitive dependence on initial conditions, and therefore also chaotic behaviour.Item Open Access EXISTENCE OF AN EXACT SELF-AFFINE TIME FUNCTION WITH A RANDOM-WALK SCALING PROPERTY(1991-07-01) Bradley, J.An exact self-affine time function, unlike a fractal in two-dimensional space, replicates exactly when scaled by differing ratios in the amplitude and time axes. A statistical self-affine time function replicates only statistically when scaled by differing ratios in the amplitude and time axis, the best known example being a random walk, where the time scaling factor is the square of the amplitude scaling factor. The existence of at least four exact self-affine time functions, called Elliot or E(t) functions, that allow for infinite number of exact replications of 12345abc structures, has been demonstrated. These E(t) functions are defined algorithmically and have no derivitive anywhere. One of these E(t) functions has the unexpected property of scaling like a random walk. This leads to speculation that it might be possible to construct a statistically self-affine E(t) function that would be a random walk forgery.Item Open Access AN EXTENDED RELATIONAL DATA MODEL AND SQL SUBSET WITH SUPPORT FOR CONTAINMENT, COMPOSITE STRUCTURES AND INHERITANCE(1996-12-01) Bradley, J.An extended relational data base model and SQL subset is described. The data model is fully relational. It allows unnormalized relations only where attributes are sets or lists of atomic attributes. There is support for both inheritance and composite entities. A composite structre can be stored by decomposition into sets of tuples stores in relations but can also be stored virtually as a view. Composite structures, including OOPL objects, can be transferred directly between database system and program variables. Manipulation languages are an extended SQL and a separate new SQL language subset for efficient manipulation of containment structures, whose manipulation with SQL is error-prone. The language subset has a predicate calculus expression structure, but allows the use of genitive relations to model containment relationships and facilitate application of natural quantifiers to containment sets. Range variables are allowed and can range over genitive relations as well as base tables.Item Open Access AN EXTENDED SQL AND QBE WITH SEAMLESS INCORPORATION OF GENITIVE RELATION AND NATURAL QUANTIFIER CONSTRUCTS(1996-12-01) Bradley, J.An extended SQL, called E-SQL, is described, for use with an extended relational data model. The data model allows for restricted unnormalized relations; sets or lists of atomic attributes are allowed but not attributes that are relations, and there is support for both inheritance and composite entities. E-SQL permits the constructs of conventional SQL, but also permits use of genitive relations to model containment relationships and facilitate application of natural quantifiers to containment sets. A genitive relation is a relation that is the equivalent to the genitive case grammatical construct in natural language. The genitive relation construct and the conventional constructs of SQL can be mixed in a seamless manner. For example, the condition within a conventional SQL-IN-construct could contain a condition that is a quantified genitive relational condition, and vice versa. Use of conditions that involve comparison of a quantified genitive relation with a quantified base table is also permitted. E-SQL can also manipulate a set of composite entity instances modelled as a group of relations. E-SQL is essentially a seamless merger of conventional SQL and an experimental genitive relation language, called COOL, designed for efficient manipulation of containment sets. It is shown that an Extended Relational Algebra (ERA) can be used to reduce E-SQL expressions. E-SQL expressions are much more concise and much less error-prone than conventional SQL expressions. It is also shown how the quantified genitive relation concept can be used in an extension of the graphical query language QBE that greatly increases both its retrieval power and simplicity of use.Item Metadata only FORMULA METHOD OF COMPUTING FBA-BLOCK REQUIREMENTS ON IBM FIXED-BLOCK-ARCHITECTURE(1984-12-01) Bradley, J.A formula has been derived tht permits computation of the number of FBA-blocks required for a file stored on an IBM disk that uses fixed-block architecture. Previously only an involved algorithm, released by IBM, was available for this task. Both formula and the algorithm involve the use of control intervals. Control intervals are of fundamental importance with IBM file systems since they permit disk independent program manipulation of large computer files and data bases.Item Open Access A FUNDAMENTAL CLASSIFICATION OF ASSOCIATIONS IN RELATIONAL DATABASES(1985-06-01) Bradley, J.A fundamental classification is proposed for the associations that can occur beteween any two relations in a relational data base. Associations are first classified into primitive and composite associations. Essentially an association between relations A and B is primitive if both A and B contain attributes drawn on a common domain. A composite association can be decomposed into a chain of primitive associations. The most important primitive associations are the functional, multivalued and equivalence associations. The most important composite associations are the cascade, matrix, and cascade equivalence associations. Most primitive and composite associations have cyclic versions.Item Open Access A FUNDAMENTAL CLASSIFICATION OF RELATIONSHIPS IN RELATIONAL DATABASES(1987-05-01) Bradley, J.A fundamental classification is proposed for the relationships that can occur between any two relations in a relational data base. Relationships are first classified into primitive and composite relationships. Essentially a relationship between relations A and B is primitive if both A and B contain attributes drawn on a common domain. A composite relationship can be decomposed into a chain of primitively related relations. The most important primitive relationships are the simple one-to-many, co-relationships, and cyclic one-to-many. The most important composite relationships are the composite one-to-many, matrix, and cyclic matrix relationships.Item Open Access Item Open Access Item Open Access INTENT ANALYSERS AND MODES OF ASSOCIATION IN RELATIONAL DATA BASES(1985-01-01) Bradley, J.A mode of association is a set of propositions about a given association between relations. It is schema-defined and permits the capture of the precise meaning of an association. Modes of association may be used with the relational language SQL/N, an upward compatible extension to SQL. They permit concise SQL/N expressions even when quite complex associations are involved, and allow the use of the entire range of natural quantifiers with all types of association. Because modes of association permit the user to operate at a higher level of abstraction than with SQL, subtle errors of intent are more likely. An intent analyser can be used to detect such errors in SQL/N expressions. Some intent analyser procedures are described.Item Open Access RELATIONSHIP RELATIONS AND NATURAL QUANTIFICATION IN RELATIONAL DATABASES(1987-06-01) Bradley, J.An alternative to the conventional method of constructing retrieval expressions for relational databases is proposed. This alternative uses precise definitions of relationships between relations, in the form of relationship relations, as well as quantification of subsets of a relation that are related to any tuple in the database. The relationship relation is used to precisely define how a subset of a relation is related to a tuple. This approach permits not only the existential and universal quantifiers, but also the large number of natural quantifiers that exist. Resulting set theoretic expressions, and expressions in relational languages based on this technique, usually mirror the corresponding natural language statement.Item Open Access RESONANCE IN EN: M GROWTH-DECAY FRACTAL TIME FUNCTIONS(1994-12-01) Bradley, J.There exists many kinds of fractal growth-decay En:m functions, depending on values for n and m. Equisegment En:m functions have immediate decomposition growth segments equal at all levels whereas regular En:m functions do not; both have the same decay fraction everywhere. A grodec stack machine whose pressure growth decay follows an En:m function is likely to be possible only for resonating functions. We have conclusively shown (a) that resonating equisegment E5:3 functions do not exist and (b) that a restricted family of resonating regular E5:3 functions does exist. A resonance equation for resonating regular functions is presented. It was discovered that there exists a simplest possible resonating regular E5:3 function satisfying the resonance equation. It is likely that a grodec machine for this simplest possible resonating function can be built. A controlling ratio within this function was found to be the Feigenbaum universal constant.Item Metadata only A TUTORIAL INTRODUCTION TO SQL/N, A NATURAL QUANTIFIER EXTENSION TO SQL(1984-02-01) Bradley, J.The original relational languages such as DSL Alpha used the basic universal and existential quantifiers in the traditional manner of mathematical logic, which is not the way in which they are used in natural languages. When DSL Alpha proved too difficult for use in a data base language SQL was developed. SQL uses special constructs to avoid the quantifiers of DSL Alpha but is otherwise similar. SQL is heavily supported by IBM and other vendors and is becoming the standard non procedural data base language. Although SQL is clearly easier to use than DSL Alpha, because it continues to use the equivalent of the basic quantifiers in the traditional manner of mathematical logic, for more complex retrievals either an SQL expression is difficult and contrived or is not possible. SQL/N permits the basic quantifiers of DSL Alpha but uses them in the much simpler manner of natural language. In addition it permits the use of the whole range of natural quantifiers. The result is that SQL/N expressions mirror their natural language counterparts and are simpler and more concise than SQL expressions. SQL/N is also upward compatible with SQL so that simple SQL and SQL/N expressions are identical. SQL/N can handle all known kinds of associations easily and permits the use of natural quantifiers with them. In this tutorial we deal only with the common one to many associations.Item Open Access A TWO-PATH RECURSIVE RELATIONAL DATA BASE STRUCTURE FOR MOLECULAR INFORMATION SYSTEMS (Revised Version)(1989-09-01) Bradley, J.A recursive relational data base structure, called the two path structure, is proposed for holding data about the properties and structure of molecules. The two-path data base approach solves certain severe restriction problems by providing two pathways to the structure of complex chemical entities. One pathway permits a fast descent to the atomic structure in terms of correct IUPAC carbon atom occurrence numbers. The other pathway descends through the molecular substructures, which can be referenced at any level in retrievals. The data base appears to be capable of faithfully recording the structure of all chemical entities, including proteins and enzymes.