BUSS 951 Critical Issues in Information Systems Lecture 8 Systems for Organisations 2: Communicative Technologies Clarke, R. J (2001) L 951 -08: 1
Notices (1) General n In the seminar today we review Arguments to help with Assignment 2 n Make sure that you have a copy of the handout on Stamper’s staircase for Assignment 2 n Make sure you have a copy of Assignment 2 n BUSS 951 is supported by a website (available from Tomorrow), where you can find out the latest Notices and get Lecture Notes, Tutorial Sheets, Assignments etc www. uow. edu. au/~rclarke/buss 951. htm Clarke, R. J (2001) L 951 -08: 2
Notices (2) Readings for this week Because Assignment 2 is due this coming week and it will be useful to discuss Arguments in the seminar this week, we will defer discussion of the readings until the seminar of Week 9: 1. 2. Yu, E. (1998) “Why agent-oriented requirments engineering” Reading 6 Yu, E. S. K and J. Mylopoulos (1994) “From E-R’ to AR’- Modelling Strategic Actor Relationshiups for Business Process Reengineering” Reading 7 Clarke, R. J (2001) L 951 -08: 3
Agenda (1) n Discuss some problems with traditional systems analysis views of work in offices n Promote a view which looks at office work in terms of action and human communication (similar to a Systems Auditors View of an IS) n Introduce the ideas behind Action Workflow (one type of LAP approach) Clarke, R. J (2001) L 951 -08: 4
Agenda (2) n Language Action Perspectives are very important within the Information Systems discipline because they demonstrate the possibility of producing effective and efficient systems without the use of Shannon and Weaver n these approaches also exemplify one of the ways in which a knowledge of human communication can be used to actually analyse, design and implement information systems Clarke, R. J (2001) L 951 -08: 5
Agenda (3) n LAP approaches are based on an entirely different theoretical basis to traditional information systemsbecause IS is a design practice and not a sciencethis means that many approaches to the design of IS are possible n We consider only two particular models that have emerged within the Language Action Perspective (LAP): n Action Workflow- one of a range of communicative IS models developed by Prof. Goran Goldkuhl and the VITS research group (Information Systems and Work Contexts at Linkopings University, Sweden) and n Prof. Jan Dietz and the DEMO group at Delft University, The Netherlands Clarke, R. J (2001) L 951 -08: 6
Views of Systems: Analyst -vs- Auditor Clarke, R. J (2001) L 951 -08: 7
Analyst’s View (1) n systems analysts are responsible for the analysis of a business system to access its suitability for computer application n analysts may also design the necessary computer system (referred to analyst/designers) Clarke, R. J (2001) L 951 -08: 8
Auditor’s View (1) n provides an independent view of the system and determines: n whether the system is functioning as it should, and n whether the accounting records stored in a computer accurately reflect the company’s financial condition Clarke, R. J (2001) L 951 -08: 9
Auditor’s View (2) n two approaches to auditing a system: n auditing around the computer- involves examining system inputs and outputs but not the processing n auditing through the computer- involves using the computer itself to examine the systems data and audit trail n a systems auditor’s view is different to a systems analyst’s view! Clarke, R. J (2001) L 951 -08: 10
Contrasting Views of Systems: Analyst -vs- Auditor n the systems auditor’s view is like a pedestrian- they follow transactions through the system n the systems analyst’s is like a birdthey view the system from above only descending when they need to examine a subsystem in detail Clarke, R. J (2001) L 951 -08: 11
Contrasting Views of Systems: Relationship to OA n arguably, an auditor’s view of systems is very useful when analysing OA system n rather than trying to specify data stores, dataflows, processes, external entities relating to office documents n we could see what elements are required to describe a document, where they come from, how they are used etc. . . Clarke, R. J (2001) L 951 -08: 12
Contrasting Views of Systems: Relationship to OA n we shall now consider one of the major approaches used to analyse OA systems n the Language Action Perspective (LAP) is a general approach to specifying and developing systems (including OA) developed in Scandinavia Clarke, R. J (2001) L 951 -08: 13
Contrasting Views of Systems: Relationship to OA n one type of LAP methodology is called Action Workflow n Action Workflow supports both a systems auditors view as well as the traditional systems view of an IS or OA system Clarke, R. J (2001) L 951 -08: 14
Language Action Perspective Clarke, R. J (2001) L 951 -08: 15
Language Action Perspective n the Language Action Perspective (LAP) is a theoretical orientation for studying modeling, design, implementation and usage of information systems in organisational contexts. n pioneering work was done by Flores and Winograd (see Reader). Clarke, R. J (2001) L 951 -08: 16
Language Action Perspective n LAP is based on an action view on language and communication, emphasising what people DO while communicating. n It has its theoretical roots in speech act theory from the Philosophy of Language developed by Austin (1955/1962) and Searle (1966) and communication action theory (Habermas) Clarke, R. J (2001) L 951 -08: 17
Language Action Perspective n since 1980 there has been a growing interest in LAP among scholars in information systems and computer science. n there are now several frameworks and methods for communication modelling for example: Action Workflow, DEMO, SAMPO and BAT. Clarke, R. J (2001) L 951 -08: 18
Language Action Perspective n Action Workflow (Goldkuhl 1996) is an approach which uses the Language Action Perspective or LAP (Dignum et al 1996) n LAP approaches in general emphasise the importance of human communication in understanding workpractices and information systems Prof. Goran Goldkuhl Clarke, R. J (2001) L 951 -08: 19
Case Study of ALABS: Loan/Return at the MCL Clarke, R. J (2001) L 951 -08: 20
Case Study of LAP n we exemplify (show) how LAP and Action Workflow can be used to describe systems by using a case study n the case study is of a system called ALABS (Automated Library And Borrowing System) which used to exist at the Microcomputer Laboratories Clarke, R. J (2001) L 951 -08: 21
Case Study of LAP n ALABS enabled students to borrow software (disks) which were then used in the Laboratories (the system was developed before networks became commonly available)- we will analyse the Student Loan Workpractice n the workpractice was first analysed by the Rodney Clarke but this analysis is from the work of Christofer Tolis Clarke, R. J (2001) L 951 -08: 22
Case Study of LAP n Students request the loan of valuable items including software, manuals, and/or hardware stored at the MCL. n Loan requests are handled by a Laboratories Staff Member who records the loan using the ALABS Student Loan feature Clarke, R. J (2001) L 951 -08: 23
Case Study of LAP n the loan must be recorded in order to not compromise the integrity of the holdings nor to infringe the licensing arrangements Clarke, R. J (2001) L 951 -08: 24
Language Action Perspective Theoretical Approach n inspired by the work of Winograd and Flores (1986). n their model describing “the basic conversation for action” (Winograd & Flores, 1986, p. 65) uses Speech Act Theory Clarke, R. J (2001) L 951 -08: 25
Language Action Perspective Theoretical Approach Clarke, R. J (2001) L 951 -08: 26
Action Workflow (1) General Approach n Action Workflow is a way of describing interaction between roles in an organisation- viewed in terms of commitment. n Definition of all constructs: Loop consisting of four phases: preparation, negotiation, performance, and acceptance. n Two roles: customer and performer. Clarke, R. J (2001) L 951 -08: 27
Action Workflow (2) General Approach n further developed into a general workflow loop ( Medina-Mora et al. 1992). n see the following diagram. . . Clarke, R. J (2001) L 951 -08: 28
Action Workflow (3) Basic Loop Clarke, R. J (2001) L 951 -08: 29
Action Workflow (4) Applied to the Case Study n can apply this to the case study to create a simple loop that describes: n student as customer n labstaff as performer n the resulting map describes the two roles involved (student and labstaff) and the workflow at hand (loan material). Clarke, R. J (2001) L 951 -08: 30
Action Workflow (5) Applied to the Case Study Clarke, R. J (2001) L 951 -08: 31
Action Workflow (6) Applied to the Case Study n this loop doesn’t say anything more specific about the case. What does say is that the interaction should be able to be understood in terms of the four phases. n let’s have a look at each of the four phases in turn– and relate them to an actual interaction from a transcript. . . Clarke, R. J (2001) L 951 -08: 32
Action Workflow (7) Analysing an Actual Transcript Clarke, R. J (2001) L 951 -08: 33
Action Workflow (8) Analysing an Actual Transcript n Look for the (speech) act that advances the loop into the next phase. n Preparation: Ends with student making a request (in this specific case, it’s actually the labstaff that specifies the student’s request!) n Negotiation: Ends with labstaff agreeing. Clarke, R. J (2001) L 951 -08: 34
Action Workflow (9) Analysing an Actual Transcript n Performance: End with reports completion: n Satisfaction: Ends with student declaring satisfaction n Having gone through the details of the loop, what are the possibilities of extending the map in order to show more of the specific details of the interaction in the case? Clarke, R. J (2001) L 951 -08: 35
Action Workflow (10) Analysing an Actual Transcript n Connections between different loops: “Child” workflows expand on a certain workflow quadrant, further detailing it. n the workflow can simultaneously be described on several levels of detail in the same diagram. Clarke, R. J (2001) L 951 -08: 36
Action Workflow (11) Analysing an Actual Transcript n each workflow involves the interaction between two (human) parties n this limits the possible expansion of workflow quadrants (this is about as detailed as it can get). Clarke, R. J (2001) L 951 -08: 37
Action Workflow (12) Analysing an Actual Transcript n Names of workflows given fram customer’s point of view (“Can I please. . . ”). n Depending on the situation, it can also be read from the performer’s point of view (“Wouldn’t you like to. . . ”), cf. the following point. Clarke, R. J (2001) L 951 -08: 38
Action Workflow (13) Analysing an Actual Transcript n Sometimes the performer initiates a workflow by making an offer (e. g. in the workflow “Get labpass and material”, the student can be seen to receive an offer to get the labpass and the material). n The map discussed so far, was based on a sole interaction Clarke, R. J (2001) L 951 -08: 39
Action Workflow (14) Analysing an Actual Transcript Clarke, R. J (2001) L 951 -08: 40
Parent Workflows (1) n How does the Action. Workflow providing help in describing the larger environment of a workflow? n With the Action. Workflow approach, the larger environment of a workflow is simply a larger workflow, where the first one is a part. Clarke, R. J (2001) L 951 -08: 41
Parent Workflows (2) Clarke, R. J (2001) L 951 -08: 42
Parent Workflows (3) n Note that the description of the larger context requires a decision on which of the two parties point-of-view to use n Note that the larger picture is quite different for the student and for the labstaff member Clarke, R. J (2001) L 951 -08: 43
Parent Workflows (4) n For the student, the loan of material from the labstaff is only a small part of doing an assignment. n In the parent workflow, it is the lecturer who is the customer whereas the student is the performer. Clarke, R. J (2001) L 951 -08: 44
Parent Workflows (5) n After the teacher has prepared and given out the assignment (preparation phase), there might be some discussion before reaching an agreement on what the student is to do (negotiation phase). Clarke, R. J (2001) L 951 -08: 45
Parent Workflows (6) n Based on the agreement, the student goes on to actually do the assignment in order to give the results to the lecturer (performance phase). n Finally, the lecturer evaluates the result, hopefully satisfied with it (acceptance phase). Clarke, R. J (2001) L 951 -08: 46
DEMO Introduction Clarke, R. J (2001) L 951 -08: 47
DEMO Introduction (1) n In another supplement we describe the LAP in Office Automation- a way of analysing OA and other systems which does not use standard data or information techniques n this supplement explores another language based technique called DEMO (developed in the Netherlands) Clarke, R. J (2001) L 951 -08: 48
DEMO Introduction (2) n DEMO is conceptually placed between LAP (Action Workflow) and traditional datalogical approaches n we review a case study of the use of DEMO to develop an actual system- the implementation of a business process for the delivery of Leased Lines at the Dutch phone company PPT Telecom Clarke, R. J (2001) L 951 -08: 49
DEMO Introduction (3) n DEMO = Dynamic Essential Modelling of Organisations n motivation for DEMO- that IS should have a theory of organisations (and also of discrete dynamic systems in general) n uses an idea of information that people would understand (information as informative) Clarke, R. J (2001) L 951 -08: 50
DEMO Introduction (4) n is used in Business Process Redesign and Reengineering contexts- is also useful in OA applications n shares theoretical foundations with other LAP methods such as Action Workflow (especially the use of Transaction based approaches which use Speech Act theory) Clarke, R. J (2001) L 951 -08: 51
DEMO Introduction (5) n a CASE tool has been developed for this methodology (CASE= Computer Aided Software Engineering) n the CASE tool was developed using a customisable version of Excelerator II) n the consequence of having a CASE tool is that prototypes or working models of systems can be rapidly implemented Clarke, R. J (2001) L 951 -08: 52
DEMO Introduction (6) n the CASE tool for DEMO called DEMOCRAT (of course!) enables the development of four types of diagrams used to specify a system n the information in this lecture is based on a paper by: Dietz, J. L. G. ; van der Rijst, N. B. J. and F. L. H. Stollman (1996) “The Specification and Implementation of a DEMO Supporting Case-Tool” unpublished Clarke, R. J (2001) L 951 -08: 53
DEMO Theory Clarke, R. J (2001) L 951 -08: 54
DEMO Theory (1) n a major theoretical foundations of DEMO involve treating organisations as social systems n organisations are treated as sets of socially constructed individual referred to as subjects n subjects influence each others behaviour through communication (ie. language not data) Clarke, R. J (2001) L 951 -08: 55
DEMO Theory (2) n the term actor is also used to abstract from particular individuals and to concentrate on the behaviour enacted by them n an actor is a particular function or activity to be performed by a subject in an organisations are considered as systems of communicating actors- subject system Clarke, R. J (2001) L 951 -08: 56
DEMO Theory (3) n the actors in an organisation communicate about some worldcalled an object world n subject systems have object worlds n subject systems and their corresponding object worlds are at every moment in a particular state Clarke, R. J (2001) L 951 -08: 57
DEMO Theory (4) n the state of the subject system represents the progress made in performing activities n the state of the object world represents the results of these activities Clarke, R. J (2001) L 951 -08: 58
DEMO Theory (5) n within traditional IS there is a distinction between the documental level in an organisation and the information level n the documental level: the organisation is viewed as a system of actors that produce, store, transport, and destroy documents n the informational level: abstracts away from the substance to focus on the semantics Clarke, R. J (2001) L 951 -08: 59
DEMO Theory (6) n DEMO abstracts one level further by focussing on the pragmatic meanings of these messages, that is their role in carrying on the business activities- what is called the essential level n philosophy of language can be used to theorise this abstraction. . . Clarke, R. J (2001) L 951 -08: 60
DEMO Theory (7) n the pragmatic unit of human communication in DEMO is called the conversation n more correctly it is called the performative conversation n these result in an actual change of state in either the subject system or the object world Clarke, R. J (2001) L 951 -08: 61
DEMO Theory (8) n performative conversations are classified into two kinds: n actagenic conversations: which result in agreements about future actions (agenda of the subject system) n factagenic conversations: which result in establishing facts in the object world Clarke, R. J (2001) L 951 -08: 62
DEMO Theory (9) n performative conversations enable original new things to happen in organisations n therefore they are considered to represent the ‘essence’ of the organisation n essential actions formed by actagenic and factagenic conversations are used to create the essential model of the organisation Clarke, R. J (2001) L 951 -08: 63
DEMO Theory (10) n essential conversations and actions are performed by responsible, authorised subjects n other activities could be performed by artefacts- especially those actions that are purely informational: n reproduction actions using existing information (database) n derivation actions of other information (processing) Clarke, R. J (2001) L 951 -08: 64
DEMO Theory (11) n for any organisation; at any moment their exists a hierarchy consisting of a documental model, an information model and one essential model n in principle, many documental models may create one information model etc. Clarke, R. J (2001) L 951 -08: 65
DEMO Theory (12) n choosing a documental model is what Information System (Re)engineering is about n this is part of a larger activity called Business Process Reengineering Clarke, R. J (2001) L 951 -08: 66
DEMO Theory (13) Levels of Abstraction Essential Business Process (Re)engineering Information System (Re)design Informational Information System (Re)engineering Documental Clarke, R. J (2001) L 951 -08: 67
DEMO Theory (14) Transaction Pattern n the core modelling concept is the essential transaction n considered to be the basic pattern of organisational behaviours n involves three phases: the order phase, the execution phase, and the result phase Clarke, R. J (2001) L 951 -08: 68
DEMO Theory (15) Order Phase n an agreement is reached netween actor 1 & 2 about the future execution of an action by actor 2 n this phase consists of an actagenic conversation initiated by actor 1 starting at t 1 and ending at t 2 n the result is an agendum (singular of agenda) for the execution of an objective action by actor 2 Clarke, R. J (2001) L 951 -08: 69
DEMO Theory (16) Execution Phase and Result Phase n During the Execution Phase: the objective action is executed by actor 2 somewhere between t 2 and t 3 n During the Result Phase: actor 1 and actor 2 reach agreement about the facts that have been accomplished as a result of the execution by actor 2 n factogenic conversation starting at t 3 and ending at t 4 Clarke, R. J (2001) L 951 -08: 70
Transaction Pattern Transaction Phases: Execution (E) Order (O) Actor 1 Actor 2 Result (R) Actor 2 agendum t 1 t 2 Actor 1 fact t 3 t 4 time Clarke, R. J (2001) L 951 -08: 71
DEMO Diagrams Clarke, R. J (2001) L 951 -08: 72
DEMOcrat Diagrams n several kinds of diagrams are supported with DEMOcrat, these include: n communicator diagrams n process diagrams n fact diagrams n action diagrams Clarke, R. J (2001) L 951 -08: 73
Communicator Diagram Graphical Elements Clarke, R. J (2001) L 951 -08: 74
Clarke, R. J (2001) L 951 -08: 75
Process Diagram Graphical Elements Clarke, R. J (2001) L 951 -08: 76
Clarke, R. J (2001) L 951 -08: 77
Fact Diagram Graphical Elements Clarke, R. J (2001) L 951 -08: 78
Clarke, R. J (2001) L 951 -08: 79
Action Diagrams n action diagrams are similar to flowcharts and describe processing steps, an example is provided that shows: n transaction type requested, promised or accepted n fact type available n condition n essential actions n sychroniser Clarke, R. J (2001) L 951 -08: 80
begin or end symbol transaction type requested, promised or accepted fact type available condition essential actions sychroniser Clarke, R. J (2001) L 951 -08: 81
References n Searle, J. R. (1966) Speech Acts- An Essay in the Philosophy of Language Cambridge: Cambridge University Press, pp. 23 -25; 54 -71 n Winograd, T. (1986) “A Language/Action Perspective on the Design of Co-operative Work” in Proceedings CSCW-86, Austin Texas, pp. 203 -221 n Goldkuhl, G. (1996) “Generic Business Frameworks and Action Modelling” Jonkoping International Business School and the Centre for Studies on Man, Technology and Organization (CMTO), and Department of Computer and Information Science, Linköping University, Draft 1996 -02 -07, 16 pp. Clarke, R. J (2001) L 951 -08: 82
n Goldkuhl, G. and K. Lyytinen (1982) “A Language Action View of Information Systems” SYSLAB Report No. 14; August 1982; Third International Conference on Information Systems, Ann Arbor, 13 -15 December 1982, 24 pp. n Goldkuhl, G. (1984) “Understanding Computer-Based Information Systems Through Communicative Action Analysis” Human-Infological Research Group, Department of Information Processing (HUMOR), Chalmers University of Technology, S-412 Göteborg, Sweden, Draft 1984 -12 -06, 26 pp. n Goldkuhl, G. (1993) “Contextual Activity Modelling of Information Systems” Research Report VITS, March 1993, Institutionen För Datavetenskap, Universitet Och Tekniska Högskolan, Linköping University Sweden Li. TH-IDA-R-93 -05; ISSN-0281 -4250, 12 pp. Clarke, R. J (2001) L 951 -08: 83
Acknowledgements n the author greatly acknowledges the use of material provided in a research talk developed by Jim Mc. Kee, Department of Information Systems Clarke, R. J (2001) L 951 -08: 84
Скачать презентацию BUSS 951 Critical Issues in Information Systems Lecture
69850ad78f46b701a15ed22be822c135.ppt