Contact us Heritage collections Image license terms
HOME ACL ACD ICF SUS DCS G&A STARLINK Literature
Further reading □ Foreword □ Contents □ Preface □ Prologue □ Acknowledgements □ Participants □ 1. Introduction □ A. Guedj □ B. Hopgood □ C. Crestin □ D. Warman □ E. Sabin □ F. Encarnacao □ G. Dunn □ H. Bono □ I. Newell □ J. Foley □ K. Foley □ L. Sancha □ M. Sancha □ N. Sancha □ 2. Working documents □ Current position □ Graphics primitives □ Core □ Attributes □ Structure □ Methodology: Structure □ Design □ Input □ Transformations □ Formal Specification □ Conceptual Framework □ IFIP Report □ Recommendations □ Future
C&A INF CCD CISD Archives Contact us Heritage archives Image license terms

Search

   
⟶ ACD ⟶ Literature ⟶ Books ⟶ Methodology in Computer Graphics
⟶ ACD ⟶ Literature ⟶ Books ⟶ Methodology in Computer Graphics
ACL ACD C&A INF CCD CISD Archives
Further reading ⇓

□ Foreword □ Contents □ Preface □ Prologue □ Acknowledgements □ Participants □ 1. Introduction □ A. Guedj □ B. Hopgood □ C. Crestin □ D. Warman □ E. Sabin □ F. Encarnacao □ G. Dunn □ H. Bono □ I. Newell □ J. Foley □ K. Foley □ L. Sancha □ M. Sancha □ N. Sancha □ 2. Working documents □ Current position □ Graphics primitives □ Core □ Attributes □ Structure □ Methodology: Structure □ Design □ Input □ Transformations □ Formal Specification □ Conceptual Framework □ IFIP Report □ Recommendations □ Future

Some Methodological Remarks

R.A. Guedj

Some Methodological Remarks for the IFIP W.G. 5.2 Workshop on Methodology in Computer Graphics

1. Introduction

The need for some standardization in computer graphics is felt by most of the people involved, designers, implementers, users of different sorts, and by institutions. There are several advantages in favor of standardization. Essentially standards can reduce costs by several mechanisms, as are described in (1). Due in part to the lack of consensus on anything proposed so far, some of us fear that a standardization is premature. Yet it is natural to ask the question: Is it possible to produce some acceptable set of recommendations?

The prerequisite for such a production is without doubt a clear understanding of a workable set of essential concepts. Hence, the path through methodology, a workshop on methodology in computer graphics. The Workshop has a methodological aim, the scope of which must be clearly grasped for our ambition has two sides: methodology of knowledge and methodology of action. We would like more to educate than teach- Therefore, we do not want the Workshop to merely lead to a state of the art report at the theoretical level, as this would just maintain the two well-tooted ideas: that the discipline should be left to computer graphics experts and that, being experts, they would also be neutral. Instead, we would like to produce a set of guidelines of action at a practical level.

A look at the present world of computer graphics systems will not be enough, for knowing and being able to act are very much linked. At a practical level, a systemic approach is action oriented.

The methodology we are looking for will allow us to gather and organize the body of knowledge in order to act more efficiently.

Let us outline the purpose of the Workshop and give some guidelines and commandments.

2. Purpose of the Workshop

  1. Get a clear understanding of a set of essential and fundamental concepts in computer graphics.
  2. Bring into light various areas of the whole discipline; identify those which need more research, those which have achieved a reasonably stable level of understanding and those which are in-between.
  3. Where possible, outline some acceptable set of recommendations. The recommendations might be at different levels of confidence and generality and at different levels of specification.
  4. Disseminate the work produced from the Workshop to the community at large through the appropriate channels. At a practical level, the purpose of the Workshop is to produce a document.

3. Commandments and Guidelines

I would like to make a plea for a systemic approach. I will assume that everyone is familiar with the notion of a system, a notion that I am not going to define. Generally speaking, a systemic approach is to be preferred to an analytic approach. The two approaches could be contrasted in the following (2):

Analytic approach

  1. Proceeds by isolation : focusing on elements.
  2. Looks at the nature of interactions.
  3. Relies on the precision of details.
  4. Modifies one variable at a time.
  5. Good knowledge of details but goals not sharply defined.
  6. Efficient approach when interactions are weak and direct.

Systemic approach

A systemic approach is not completely safe.

A systematic use of a systemic approach, (take for instance a description of models at the mere relational level), can quickly lead to a useless collection of systems.

A notion which has a too wide generality can also backfire and, instead of being prolific, can become sterile.

The uncontrolled use of analogies, metaphors and isomorphisms can also drag in interpretations which blur instead of enlighten the picture.

A few commandments or principles, acting either as a framework or to direct choices between competing designs, are tentatively set up. Some of these principles may appear to be unsound when applied too rigidly.

Cl
Proceed by enrichment of concepts instead of detailed definitions. Do not hesitate to address intuition, insight and creative thinking.
C2
Keep up a wide enough variety of concepts. Avoid the monopoly of one dominant species; this usually leads to a poor and unbalanced system.
C3
Look for invariants.
C4
Look for and identify the crossings, overlappings of areas, where concepts must be re-evaluated. Those spots are often the key points of a complex system.
C5
Make sure to clearly differentiate first in order to better integrate. There is no real union without antagonisms and conflicts. The homogeneous has a large entropy and no interest. The union of diverse entities is creative, increases the complexity, and leads to higher levels of organization. This should apply to concepts, opinions and ideas.
C6
Maintain some natural constraints. Leaving things completely loose often leads to disorganisation and is dangerous.
C7
Evolution often comes after aggression. You must be open to the external world of computer graphics and be ready to answer some hard attacks.
C8
Go to and fro several times between form and content, syntax and semantics, levels of representation and levels of functions.
C9
When necessary go into detailed description, otherwise sharpen the goals.
C10
Feel, sense, and have a due regard to the 'natural' timing. Some parts are not matured at all. Some should be fully specified. This applies to ideas, systems, as well as to people.

Conclusion

The search for a small set of basic concepts for computer graphics is taken as a challenge. As in the quest for the Holy Grail, we might never succeed in finding this set but we may have reached a better understanding of computer graphics leading to a simpler use for a wider set of people. Let us take this challenge as such and call the workshop:

G R A A L (3)

Notes:

Objectives of a Software Standard (1)

Any discussion of proposed standards should start from the specific objectives of standardization; in this case the most important objective should be cost reduction, where cost encompasses both the capital costs of hardware and software as well as the running costs of the complete application which uses computer graphics. Standards in this field can reduce costs by several mechanisms, in particular:

  1. Improved portability of software increases the market for both basic software and application programs, thus reducing the unit cost both for supply and for maintenance.
  2. Increased portability reduces the cost of transfer to new hardware, possibly allowing more cost-effective devices to be used when they become available.
  3. Standards reduce the cost of training. They increase programmer productivity by reducing the number of systems with which the programmer needs to be familiar.
  4. The use of a standard graphics system instead of a specially built one should allow a quicker return on the investment costs of application software.
  5. Compatibility between remote sites will increase communication and resource sharing between remote groups especially in the context of linked computing systems.

Area of standardization

To be widely adopted a graphics standard needs to be applicable to a wide range of graphics devices (device independent), available on a wide range of computers (computer independent) and appropriate for a wide range of applications (application independent).

References

(1) taken from report: Device Independent Graphics Design Criteria for Network Protocols and Subroutine Packages Workshop convened by the Computer Aided Design Specialist Group of the British Computer Society on the 16th and 17th of December 1974.

(2) This part is widely inspired by the chapter II of 'La Revolution Systemique' in LE HACROSCOPE by Joel de ROSNAY Ed. SEUIL, 1976 (in French).

(3) Grail (gral), n. {OF. Graal. ML. Gradalis,dish,vessel; origin uncertain.) A platter (also taken as a chalice) which according to medieval legend was used by Christ at the Last Supper, and in which Joseph of Arimathea received the last drops of Christ's blood at the cross: often called Holy Grail. Prom the New Century Dictionary of the English Language P.F. COLLIER & Son, 1927 Edition.

⇑ Top of page
© Chilton Computing and UKRI Science and Technology Facilities Council webmaster@chilton-computing.org.uk
Our thanks to UKRI Science and Technology Facilities Council for hosting this site