Metaparadigm: A soft systemic MIS design approach

Journal of Medical Systems, Vol. 17, No. 1, 1993

Metaparadigm: A Soft Systemic MIS Design Approach Peter Kokol

The aim of this paper is to introduce the soft systemic approach to medical information system design, called metaparadigm. The metaparadigm is a metadesign paradigm. Using it we first design an IS design paradigm and it is our deep belief that metaparadigms employment can enhance many conventional IS design weaknesses and result in successful information system design and use. In the paper we show the applicability of the metaparadigm in MIS design and introduce a new personal MIS design paradigm.

INTRODUCTION Software crisis is one among the most used idioms in the software community. 2'5'7'9'15'19'31'34'37 It can be easily recognized by unsuccessfulness of many information and software systems (IS) from various fields. Because of its special character, the medical area is especially affected. Undoubtedly the crisis is the consequence of many different factors. However it is our belief that the inadequate design, based on the traditional waterfall paradigm, is one of the most influential. The strict separation between specification and implementation, no support for rapid prototyping or automation, neglecting of the social, cultural, political etc. aspects of system design are some of most important weaknesses of such type of design.

New Software Design Paradigms Many new software design paradigms have been proposed as a response to the software crisis. These range from solely technological to most elaborate paradigms, based on theory improvements. 34 Operational, prototyping and transformational paradigms 2,6,12,15,17,18,20,23,25,27,36,38 are most prominent among them. They try to incite user participation in order to deliver executable code early in the design process, and automate the production of software. Despite many advantages of new software design paradigms it is arThe Faculty of Technical Sciences, University Maribor, Smetanova 17, 62000 Maribor, Slovenia. 47 0148-5598/93/0200-0047507.00/0 © 1993 Plenum Publishing Corporation

48

Kokol

gued3'16'30'31'39'45'46'47 that it is unreasonable to rely on only one paradigm, because the tools and techniques for one set of circumstances need not be appropriate for others. Various solutions based on metadesign and modeling of software design paradigms like domain analysis 16, metamodeling 22'4°'49, megaprogramming 47, system factory 45, Metaview system46, TAME 39 and Meta-system approach 8 were proposed recently in the manner to enable more flexible design in different design situations. However, the majority of above metadesign approaches have at least one of following limitations: • they are concerned only with some aspects, dimensions and phases of metadesign • they do not provide a well defined metadesign methodology • they are often to technological and neglect political, cultural and other dimensions of system design • their metadesign life cycle follows the waterfall model • they are concerned only with large systems In this paper we would like to present a new metadesign approach called metaparadigm with which we try to overcome above limitations. We first we present some fundamental definitions and introduce the metaparadigm and its use in IS design. In succeeding sections we show the employment of the metaparadigm in the MIS (medical information system) design and present a new end-user oriented MIS design paradigm (PMIS). We conclude the paper with an overview of metaparadigm's advantages and weaknesses, and propose some challenges for the future work.

WHY

TO METADESIGN?

The analysis of the state-of-the-art and our own research have shown that the information system design: • is a human activity system11; • is a process 24"4° performed in a specific design situation, according to a specific design paradigm; • is a goal-oriented decision making exploration and learning activity; • occurs in two contexts: the context within which the designer operates and the context produced by the designing design itself. Analyzing design processes, design situations, design paradigms and relations between them 31 we found that: • there are no absolute good or bad design paradigms, they perform good or bad only in specific design situations; • there are design situations for which no known design paradigm is appropriate. Congruent to above findings we contend that to enable successful software system design in any design situation we must first design an appropriate design paradigm. This activity was called the metadesign and a metadesign paradigm a metaparadigm. 3°'31 With the use of the metaparadigm we expect to achieve the following major advantages:

Metaparadigm: A Soft Systemic MIS Design Approach

49

• to employ the right design paradigms in every design situation • to design ISDPs which use will satisfy all involved parties (designers, users, management etc.) and will result in successful information systems • to enable superior knowledge eliction regarding both the design situation and the needed IS • to enable more effective and efficient management, communication and conflict resolving • to enable the design in an endless learning loop.

BRIEF DESCRIPTION OF THE METAPARADIGM Some Basic Definitions Before describing the metaparadigm in more details we would like first to specify some basic definitions. First and most important we wish to define the idiom paradigm. According to some basic explanation, stating that the paradigm is • a set of universally recognized scientific achievements that for a time provide a model of solutions to the community of practitioners33; or • a set of meta-theoretical assumptions about the nature of the subject of study 1°, we have defined it as a pattern for executing actual actions in the real world. Using the last denotation we claim that the IS design paradigm (ISDP) is a pattern for executing actual IS design actions in designing real-world IS applications and the metaparadigm is a pattern for executing actual design actions in designing real world-design paradigms. The IS design paradigm is hierarchically the highest entity used in the design of information systems. It defines design processes, design approaches, design methodologies, and design methods and techniques.

Metaparadigm Objectives To be functioning, the metaparadigm should be able to perform the following minimal set of activities: • formal description of design paradigms, design process and design situations; • formal evaluation and comparison of design paradigms; • metadesign of design paradigms, where metadesign is treated in a very broad sense as invention of new design paradigms, adoption of known design paradigms, composition of known design paradigms, selection between design paradigms etc.; and • learning, accumulation, and reuse of new knowledge.

Metaparadigm's Characteristics It is interesting to note that a similar crisis as in IS design, is also observed in the traditional scientific approach, which, as it seems, is not capable to solve real world

50

Kokol

problems. Thereafter some new approaches have been proposed. The most successful one among them is the soft system approach. 11 It is based on the holistic view of the world. The consequences of such a view are multidisciplinarity and pluralism. The pluralism and multisciplinarity are especially important in the new information system research field where there is a shortage of old hypothesis and a need for new ones. 37 It is also believed that the information system design and use is a process of the multidimensional change which is performed in an endless learning loop. Thereafter we decided that a metaparadigm should be pluralistic, useful, soft systemic, easy to use and recursively adaptable (recursive adaptability states that a metaparadigm must be an iterative learning system accumulating new knowledge about metadesign, software system design and related). The metaparadigm is composed out of a framework, a theory and a metadesign methodology and philosophy in its current form. We have selected these three elements because they represent the minimal set which entitles one to define and describe metaparadigms functions and characteristics. In our first attempt in selecting suitable metaparadigm's elements we have chosen the idea of a process as the framework, the Checklands Soft System Methodology (CSSM) 11'26 as the methodology and philosophy, and the process formalization (metamodeling) 22'24'39'43 as the theory. The most important arguments for the selection were: the applicability of CSSM for researching human activity systems, the similarity between definitions of a process and a human activity system, and a strong relationship between a process and its formalization. The use process of the metaparadigm is shown on Figure 1.

CSSM as the Part of the Metaparadigm When we use the CSSM (a brief description of the CSSM and its use can be found in 26'28) as a part of the metaparadigm (see Fig. 2) there is no conventional problem situation. There are just various expectations and ideas in the heads of involved customers (the future users of the information system and the design group) which paradigm design group have to take into account. The paradigm design group should also analyze and express the general situation in information system design, the specific situation in the field for which the information system should be constructed and the specific design situation. IHETAPARADICH I

(

rouse

use I

evlaluaUon ,n/en•on reuse I

design descripUsn I l I I adapt~sn insLanat3on cosLunlLzaLion... f )ISS design paradigml( redocUsn ,, 5S desLgu approach SS design, process I

use

'I

IS

Figure 1. The metaparadigrn and its use

I

Metaparadigm: A Soft Systemic MIS Design Approach

51

and expressinn of - general siLuaHon

[7.Imwovement of F~rooL definiHons

- field sitnation

liand concept, modelsI

- specific siLuaL.

|

- e,peetUons and

f

[ ]1

[

7'

ideas of involved

I ]S.Conllarison of 2 ] /

cesLoners

I IniLh 4

. . . . .

. . . . .

. . . . . . . .

I 3. ConsLruc--c-t~o-F-~oL---~

I definitions

I

16.$earchin9

/

/ I

1 REAL HORLD

. . . .

I 4. Const.rucLion of

' [5~aceptual nndels

]

,A

4a.Conyarison I ]4b.DescrlpLion niLh a

eit~ a forml I I~tamdel. evaluation model

and comparison eith other mdels

Figure 2. The use of the CSSM in the context of the metaparadigm

CASE STUDY: A PARADIGM FOR PERSONAL MIS DESIGN After employment of some more or less experimental design paradigms 23'25'26'29'31 we successfully used the metaparadigm in different medical design situations. One of the most interesting applications was the design of a paradigm for personal medical information system design (PMIS). Every PMIS should support the every day work of a single physician in a pediatric department. According to the great number of physicians in the department and their individual requirements and needs we have to design a different system (PMIS) for each of them. Furthermore, according to our theory we have first to design a different design paradigm, too. A preliminary analysis of users has revealed two types of them: • users which want to design their personal MIS by themselves; and • users which want to hire a design team to perform the design mission. In present example we want to present the later case. To simply our task we have first collected common characteristics of different paradigms of the above type and designed a general PMIS design paradigm (GISDPp~Is). GISDPpMIs has then been used as the framework for the design of individual paradigms, and will be used also in our future applications.

Analysis and Expression of the Situation General Situation. To enable successful IS design in any design situation it is urgent to understand what are the most influential causes of the present software crisis. Using the reduced Ives f r a m e w o r k 27 w e have divided the causes into seven classes presented below: 1. external environment causes are a consequence of the general problems of "traditional science methods" like complexity of the real world, unavailability of experimentation objects, special role of humans with theirs freedom of choice and management problems;

52

Kokol

2. organizational cause arise because prevailing social arrangements inhibit creative planning and successful change; 3. causes tied with the nature of IS arise because IS are more complex according to their size than any other artifacts, much of their complexity is arbitrary forced by institutions and systems to which an IS must conform, the IS is invisible and it is not subjected by physical lows; 4. user environment causes arise because the user don't know exactly what he wants, he can not describe that or the new IS changes his environment and thereafter his requirements; 5. IS development environment causes arise because of the imperfect foresight of design group members; 6. causes tied with IS development process are the strong separation between specification and implementation, physical limitations, semantic gap between users and developers; methodologies lack synergy with other IS research areas, they have inadequate conceptual base, they have limited theoretical foundations, they are unware of philosophical underpinnings of system development and they are based on the monolithical language paradigm; 7. causes tied with the us process arise because of shortage and inappropriateness of empirical studies, ill defined metrics etc. Specific Situation in Medical Informatics. Our recent MIS design literature analysis 31 performed on about 1000 papers from various sources like conferences, journals and newsletters, has shown that despite many published papers, only 14% contain more detailed MIS design description. Among them, most present the use of simple techniques and methods, and only a few the use of more elaborate approaches and paradigms. 40% of authors have used specific medical design methods, and only one third of them have developed their own ones. The authors have mainly used old structured techniques, ERA approach, organization modeling, prototyping and also the CSSM. 13,21,42,43,44,48 Among recognized specific medical methods MUMPS 3s and HELP 41 were used in most applications. The most interesting new design methods were KALEID 1 and PRIST 14. According to the above literature overview we can state that only a small part of medical informatics researchers regards the MIS design as a special field and realize its complexity and specificity. These few of them propose original and very interesting solutions, but which are unfortunately to specific to be used in general manner. Specific Design Situation. All PMIS design situations were characterized by: • the wide spectrum of physician needs which vary by speciality, practice patterns, research requirements, personality, environment, culture, education etc.; • great semantic gap between users and designers; • the conservativeness of many physicians; • lack of time--the physicians are to busy with their own work to "waste" time with the participation in the design process; • lack of computer knowledge; • the medical users understand the PMIS design process as typical engineering work in which unambiguous requirements written in natural language are simply transformed into the final PMIS (the final PMIS is really final, it doesn't need any maintenance). We have used the software process model notation, called BSM (Figure 3.) to describe

Metaparadigm: A Soft Systemic MIS Design Approach

53

Hain parL

-~artifaet

agenL:

A A

!attribttt

role

activit 9

I_.~_FinforeaLion flo~ Al

macro acLivity

Hacro acLivibj

-~artifacL from the main parL or hierarhicaly higher macro acLivity

Figure 3. The BSMmetamodel

the users view of the PMIS design process (Figure 4.). • the design group understand the PMIS design process as an endless learning process (Figure 5a). They want first, with the great involvement of users, build a prototype and adapt it until the physician is satisfied. Reuse of routines is used in the subprocess of building the prototype (Figure 5b). During the implementation the design group improves the prototype's performances to meet also the nonfunctional requirements. It is recognized that the end-product (final PMIS) is never finally finished and needs continuous maintenance. The main differences between the users and design group understanding of the PMIS design process are identified in the Table 1.

The Root Definition and the Conceptual Model According to the situation expressed above the following root definition of the PMIS design paradigm was defined and agreed from both users and designers: A paradigm defining an iterative process performed by a design group with the participation of a physician, concerned with learning requirements and needs, education, data

LesL + a apt.ion

STATEHEHT OF inLervieus HEEDSAHD

entr ................ da'~9 >I

DATA

Figure 4. The physicians view of the PMIS design process

54

Kokol

e×ercisin 9

mainLenance

,i

STATEHEHTOF *inLervie~s ) PROTOTYPEOF ieproving HEED prototype PHIS performances'/ PflI5 coostruc~oo~

ieproveeent

i

i]~iiil'eih~ii~z~

~

i

~LL3

i

! ~]iiiil'~ii~ii[z~ ~,ied in4GL i

iedinqGL i ....................

i .................... : I]eveF

ifinished

i

n Pro~oy~ construction G ~prove~eot )

B inclusion

REPOSITODYI oF DOUTIHES[ ~ selecLion USEFUL

IDounHts

]q I~

ladapLion I

HEEDSAHD programnin9 , HEH RE~IREHEHTS " ROUTIHES

ADAPTED

[

ROUIIHES

1

I inLegraLion '

'gOTO/Y'~OFI PMIS

Figure 5. The design group view of the PMIS design process: (A) the main part and (B) the macro activity.

entry, requirements implementation and maintenance, so that the process results in a successful PMIS (successful in terms of user satisfaction4). Using the above definition as a basis we constructed an appropriate conceptual model and expressed it with the BSM (Fig. 6). The model is the composition of both previous models (Fig 4. and Fig 5.) with the added education activity which should support the Table 1. The Main Differences Between the Users and Design Group Understanding of the PMIS Design Feature

Design group

Physician

Process model

Iterative users and design group

Sequential design group

Agents of design Data-entry

Integrated

A separate activity performed by design group informal (natural language)

Type of specification

Operational (prototype)

Metaparadi~m: A Soft Systemic MIS Design Approach

0xereising

. s ,tt,taTorl'l"t' "l's[] ,I,eomv,t ,tta

] pro+o~y~ ' [ P , I S

oF

55

JI,

mainlenanee

PHis lOOC0't' J 'I*,[ 100rf0,,ao-s i,provi,,g

a0:a,o,tatl0,:

conslrucdfan~ ,~ ! , .~-.--.i ........r'" [l'~"~n'e;;ei="+ iaproveuent It ~li~ple~enL~ ~ im,Pz.eme,n.t- ,i..[..i...n.i...s..h..e..di ied in 4GL i ~ -.~..z..n....q.~..

U

Figure 6. The final model of the PMIS design paradigm

motivation of the physicians and should also enable more effective communications between the physician and the design group. CONCLUSION The aim of this paper was to introduce the metaparadigm and present its applicability in MIS design. It is our deep belief that using the metaparadigm can enhance many MIS design weaknesses and as a consequence result in successful medical software systems design and use. The greatest advantages of the metaparadigm are: pluralism, the strong theoretical base, it can be used in any design situation, it can be tailored for personal needs, and it is systemic, easy to use, learn and teach. Its greatest weakness is relative immaturity. In future work we must first empirically research and test its use in many various design situations. Next we have to construct some computerized tools which should support metaparadigms basic functions and finally we need to research the possibility to extend, adapt and improve its theoretical base. REFERENCES 1. Abul-Huda, B.A.H., et al., Handling multimedia objects in medicine using KALEID. In MIE 88: Lecture Notes in Medical Informatics 35. Hansen, R. et al. (eds.). Springer Verlag, Berlin 1988, pp. 697-701. 2. Agresti, W.W., New Paradigms for Software Development, IEEE CS Press, New York, 1986. 3. Avison, D.E., and Wood-Harper, A.T., Multiview An Exploration in information systems development. Austr. Comp. J. 18:174-179, 1986. 4. Bailey, J.E., Computer based system to measure and analyze computer users attitudes. Policy Inform. 11:29-37, 1987. 5. Basili, V., The future engineering of software: A Management Perspective, IEEE Comp. 24:90-96, 1991. 6. Boehm, B.W., Gray, T.E., and Seewaldt, T., 1984. Prototyping versus specifying: a multiproject experiment. IEEE Trans. Soft Eng. 10:290-302, 1984. 7. Boland, R.J., and Hirschheim, R.A. (eds). Critical Issues in Information System Research, John Wiley, England, 1987. 8. Boloix, G., Sorensen, P.G., and Tremblay, J.P., Transformations using a meta-system approach to software development, Software Engineering J. 7:425-437, 1992. 9. Brooks, P.F., No silver bullet: essence and accidents of software engineering. IEEE Computer 20:10-19, 1987.

56

Kokol

10. Burrell, G., and Morgan, G., Sociological Paradigms and Organizational Analysis, Heineman, London, 1979. 11. Checkland, P., Systems Thinking, System Practice, John Wiley and Sons, Chichester, 1981. 12. Connors, D.T., Software Development Methodologies and Traditional and Modern Information Systems. SEN 17-2:43-49, 1992. 13. Coward, P., The use of joint application design (JAD) to determine user requirements for nursing information systems (NIS). Nursing Informatics 91: Lecture Notes in Medical lnformatics 42. Honwenga, E.J.S., et al. (eds.). Springer Verlag, Berlin, 1991, 208-209. 14. Cristiani, P., and Larizza, C., PRIST: A fourth-generation tool for medical information systems. International Journal of Biomedical Computing, 25:137-150, 1990. 15. Davis, A.M., Operational prototyping: A new development approach, IEEE Software 9:70-78, 1992. 16. Diaz, R.P.L., Domain analysis: An introduction. ACM Software Engineering Notes, 15(2):47-54, 1990. 17. Fuchs, N.E., Specifications are (preferably) executable. Software Engineering Journal 7:323-333, 1992. 18. Gomaa, H., and Scott, D.B.H., Prototyping as a tool in the specification of user requirements. 5th ICSE, IEEE CS Press, 1981. 19. Hayes, I.J., and Jones, C.B., Specifications are not (necessarily) executable. Software Engineering J. 4:330-338, 1989. 20. Hilal, D.K., and Soltan, H., The prototype or not to prototype? That is the question. Software Engineering J. 7:388-392, 1992. 21. Jacobs, P., A methodology for successful MIS projects. SCAMC 88. Greenes, R.A. (ed.), IEEE CS Press, New York, 1987, pp. 837-842. 22. Katayama, T., Proceedings of the 6th Int. Software Process Workshop, IEEE CS Press, 1991. 23. Kokol, P., Some notes on developing reliable medical software. MEDINFO '89, Barber, B. et al. (eds.). North-Holland, Amsterdam, 1989, pp. 294-29. 24. Kokol, P., Formalization of the software development process using meta-models. ACM Software Engineering Notes, 14(4): 118-123, 1989. 25. Kokol, P., Stiglic, B., Zumer, V., and Novak, B., Software crisis and new development paradigms or how to design reliable medical software. CBMS 90. Nagle, T. (ed.) IEEE CS Press, New York, 1990, pp. 137-144. 26. Kokol, P., Structured Spreadsheet Modeling in Medical Decision Making and Research. J. Medical Syst. 14:107-118, 1990. 27. Kokol, P., and Novak, B., The Application of Extended Operational Paradigm in Designing Electrical Engineering Software. Advances in Electrical Engineering Software. Silvester, P.P. (ed.), CMP & Springer Verlag, Berlin, 1990, 251-262. 28. Kokol, P., New Evaluation Framework for assessing the Reliability of Engineering Software Systems Design Paradigms. In RRES'91. Brebbia, C.A., Ferrante, A.J. (Eds.), Elsevier, Amsterdam, 1991, pp. 173-185. 29. Kokol, P., A New Microcomputer Software System valuation Paradigm: The Medical Perspective. Journal of Medical Systems 15:269-274, 1991. 30. Kokol, P., Stglic, B., and Zumer, V., Software Design Tools Evaluation in Context of a Metaparadigrn, Proceedings ofAQSDT'92, Nahouraii, E. (ed.), New Orleans, IEEE CS Press, pp. 34-41, 1992a. 31. Kokol, P., Use of the Metaparadigm in IS Design, Ph.D. Thesis. University Maribor, Maribor, 1992b. 32. Kokol, P., et al., The application of structured spreadsheet modeling in medicine. Cybernetics Syst. 23:295-297, 1992. 33. Kuhn, T., The Structure of Scientific Revolutions (second edition), University of Chicago Press, Chicago, 1970. 34. Lyytinen, K., A taxonomic perspective of Information Systems Development: Theoretical constructs and recommendations. Critical Issues in Information System Research, John Wiley, Chicester, 1987. 35. Miller, R.E. et al., Development and operation of a MUMPS laboratory information system: a decade's experience. SCAMC 88, Greenes, R.A. (ed.), IEEE CS Press, New York, 1988, pp. 654-658. 36. Mumford, E., Participative system design. Struct. Method System Objectives Solu. 1:5-19, 1981. 37. Mumford, E., Herscheim, R., Fitzgeterlad, G., and Wood-Harper, T., Research Methods in Information Systems, North-Holland, Amsterdam. 1986. 38. Nota, G., and Pacini, G., Querying of Executable Software Specifications. IEEE Trans Soft Eng 18( 1992)705-716.

Metaparadigm: A Soft Systemic MIS Design Approach

57

39. Oivo, M., and Basiti, V.R., Representing software engineering models: The TAME goal oriented approach. IEEE Trans. Soft Eng. 18:886-898, 1992. 40. Osterveil, L., Software processes are programs too. ICSE 87, Riddle, W.E. (ed.), IEEE CS Press, New York, 1987, 2-13. 41. Potolicchio, S., and Hylton, J., Enhancing clinical investigation in neurology with a design information system, SCAMC 88, Greenes, R.A. (ed.), IEEE CS Press, New York, 1988, pp. 688-692. 42. Pryor, T., A development of decision support systems. Int. J. Clin. Monit. Comp. 7:137-146, 1990. 43. Rawinski, E., A soft system approach to nurse workforce planning. MIE 88: Lecture Notes in Medical Informatics 35, Hansen R., et al. (eds.), Springer Verlag, Berlin, 1987, pp. 300-304. 44. Saksater, J., KDO's concept on hospital information systems. MIE 88: Lecture Notes in Medical Informatics 35, Hansen R. et al. (eds), Springer Vedag, Berlin, 1988 pp. 454--458. 45. Scacchi, W., The USC system factory project. ACM Software Engineering Notes 14(1):61-82. 46. Sorensen, P.G., Tremblay, J.P., and McAllister, A.J. The Metaview system for many specification environment. IEEE Software 2:30-38, 1988. 47. Tracz, W., A conceptual model for megaprogramming. ACM Software Engineering Notes 16(3):36-45. 48. Wastell, D.G., Phenomenology and participation: alternative principles for information systems development in health care. M1E 88: Lecture Notes in Medical lnformatics 35, Hansen, R., et al. (eds.), Springer Verlag, Berlin, 1988, pp. 717-721. 49. Wileden, C.J., and Dowson, M. (eds.), Software Processes and Software Environments, Proc. of an Int. Workshop, Coto de Caza, 1988.