Browse Prior Art Database

Original Publication Date: 1975-Dec-31
Included in the Prior Art Database: 2005-Sep-15

Publishing Venue

Software Patent Institute

Related People

Harvard H. Holmes: AUTHOR [+3]


Schematic diagrams form a natural medium of communication in a wide range of problem areas. In this thesis, we will describe a comprehensive approach to problem solving using schematic diagrams as the interface between man and computer. Past efforts at computer aided design have been hampered by an approach which combined the man-machine problem-description interface with the problem analysis portion of the system. In this thesis, we set forth a methodology which separates these two aspects of computer aided design. By recognizing those topological properties of schematic diagrams that are common to a wide variety of disciplines, GMS is able to provide a single man-machine problem description interface for use in a wide variety of problem solving disciplines. In addition, GMS includes intermediate data structures and facilities that form a natural interface and starting point for the creation of additional analysis capabilities are described. The problem-definition interface supports two main activities: the creation of elements, and the interconnection of these elements to form diagrammatic models. These elements are the basic building blocks for creating models. In the past, the elements have been embedded rather deeply in the software. Thus, the creation and description of elements was done by the system designer or, at best, was relegated to a separate phase which required substantial familiarity with the software. The main difficulty in creating a new element was to communicate to the analysis portion of a system the exact meaning of the element. Our approach, on the other hand, makes it easy for a user to describe the meaning (semantics) of a new element in a natural way. Tf the new element is a primitive (containing no other elements as components), the semantic description is given in analytic form (e.g. a formula), or in empirical form (such as a table of numbers). If the new element is composite (a combination of previously created elements), its semantics are defined implicitly by the semantics of the component elements along with the topology of the interconnections. A convenient representation of the topology of the interconnections is given by a routine which traces lines in a drawing and recognizes nets of joined lines. A translator is described which produces explicit semantics of composite elements from the semantics of the component elements and the topology. A complete prototype implementation for this part of the system is described. The problem-soiving portion of this methodology allows a wide variety of analysis techniques which can be flexibly combined to solve a particular problem at hand. This approach recognizes that no workable scheme for automatically constructing computer programs has been developed. Nevertheless, several improvements in present techniques for constructing programs can be made to prepare for such schemes. These improvements include flexible data structuring facilities for programming languages and generalized user interfaces. An initial set of data structures and analysis functions for constructing programs is described. Another aspect of our work is to demonstrate how existing problem solving systems can be extended using the graphics problem description interface. These extensions serve to tailor analysis packages to a spectrum of disciplines where such analysis techniques are appropriate. This gives the user the impression that he is using an analysis routine which has been specifically constructed for his problem.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 2% of the total text.

Page 1 of 71



Harvard H. Holmes

November 1975

Prepared for the U.S. Energy Research and Development Administration under Contract W- 7405-ENG-48



An Int.roduction to Diagrammatic Modeling I

1.1 Diagrammatic Modeling Defined and Illustrated 3 1.1.1 Use of Diagrams for Modeling 3
1.1.2 Earlier Modeling Programs

1,2 A Conceptual Framework for Interactive CAD 11 1.2.1 The Art of Design 12 1.2.2 Goals of CAD Systems 15 1.2.3 Need for Separation of Tasks 17 1.2.4 The Graphics Facility 1.2.5 The Analysis Facility 20

1.3 An Idealized Modeling System 23 1.3.1 Defining a Primitive Element 23 1.3.2 Creating a Composite Element 26 1.3.3 Graphics Features 29 1.3.4 Topological Analysis of Composite Elements 31 1.3.5 The Role of the Translator 32

1.4 Examples of Modeling 34 1.4.1 PERT Diagram Example 34 .1.4.2 Circuit Diagram Example 39 1.4.3 Digital Logic Example 45 1.4.4 Compartmental Modeling 54

2. A Graphics Modeling System 70 2.1 Software Organization (and Operation) 70 2.1.1 GMS Information Flow 72 2.1.2 Graphic Primitives 74 2.2 Data Structures and Storage 76 2.2.1 Types of Blocks 76 2.2.2 Data Structures for Graphics and Derived Blocks 78 2.2.3 Data Structure for Text Blocks 93 2.2.4 Filing Module 84 2.3 The Prototype Graphics Editor 86 2.3.1 Data Structures Used by the Graphics Editor 86 2.3.2 Implementation of Graphics Editor Commands 88 2.3.3 The Software Graphics Interpretor 91 2.4 Text Editor 93 2.5 Analyzing the Topology 95 2. 5.1 Data Structure 9 5 2.5.2 The Topological Analysis Process 99 2.5.3 Other Data Structures 101 2.6 The Translator 102 2.6.1 Notation 103 2.6.2 Overall Operation of the Translator 104


I an indebted to my thesis committee, Professor H. B. Baskin, Professor D. Ferrari and Dr. L. P. Meissner, for their encouragement and guidance, to D. M. Austin for many helpful suggestions and collaboration in the programming of the prototype, and to the users at Lawrence Berkeley Laboratory for their suggestions and motivation, specifically Ivan Wood and Horace Warnock of Electrical. Engineering Drafting, and Michiyuki Nakamura, Richard LaPierre, John Mendes, Don

University of California at Berkeley Page 1 Dec 31, 1975

Page 2 of 71


Evans and Frank Neu of Electrical Engineering Research and Development. I am also indebted to James Baker and Carl Quong, Math and Computing, LBL, for support and encouragement; and to Virginia Franks for preparing the manuscript.

Finally, this project would never have been completed without the encouragement and support of my wife Susan.

This work was supported by the Energy Research and Development Administration under contract W-7405-eng-48. GRAPHICS MODELING TECHNIQUES IN COMPUTER AIDED DESIGN

Harvard H. Holmes

Lawrence Berkeley Laboratory Berkeley, California


Schematic diagrams form a natura...