Children Support Agency Case Study
Introduction The use and potential benefit to system developers is examined by use of the Semiotic Framework method in the case study information supplied regarding the Children Support Agency (CSA).
Analysis of Semiotic Framework The framework as described by Kangassalo et al (1995) (1) refers to the work of Stamper (1987) (2) as it applies to information systems, and distinguishes the four levels of properties as empirics, syntactic, semantics, and pragmatics. This is likened to a semiotic ladder running from the physical world to the social (business) world. The semiotic framework consists of two main components, these being the human information functions and the IT platform considerations. These are both split to three sub-components.
Social World, developer actives would be: To determine how best to match the negative responses of some staff to new technology with the high expectations of others, by designing a system which takes account of both To ensure the legal aspects such as compliance with the Data Protection Act (DPA) (2) are addressed. To ensure contractual information is protected in transmission. To meet the cultural standards held by those who work in an organisation whose purpose is to support disadvantaged young people.
Issues are: Lack of computer literacy among some CSA staff, its status as a charity will probably restrict funding available for the system, feelings of protection for financial data versus lack of (apparent) concern voiced about personal data of vulnerable young people. The wish to accommodate training in IT for young people, without concern that this may lead to opportunities for any who have anti-social tendencies to affect the overall operation of the system by having access. The lack of realisation that today's young people in the age range 12 to 24, whether from a deprived or difficult family background may be conversant with the use of computers. _________________________ 1. Kangassalo et al, (1995), p 358. 2. Stamper et al, (1987), p 43-78. 3. Data Protection Act 1998.
Pragmatics, developer activities would be: To attempt resolution between conflicting attitudes in conversation which were expressed about the value of the system, and consider capital and revenue funding for the new system.
Issues are: To determine how the system would be supported, and responsibility for the support.
Semantics, developer activities would be: To attempt to model the syntactic structures, which are by nature, the technical concerns, to the semantic which concern the world are matched, in a machine-independent manner.
Issues are: Security concerns, which are people-related, with system issues, which are software dependent.
Syntactics, developer activities would be: The formalisation of documentation of the system specification, and outline the programming requirements. This is the bridge between the conceptual and the formal rules governing system development.
Issues are: The documentation may only be understood by the IT people who create the system
Empirics, developer activities would be: To estimate the number of data fields required, their volume, the speed with which they require to be transmitted, and the overall performance as perceived by the user.
Issues are: Limited information available, combined with inability of potential users to express these attributes.
Physical World developer activities would be: To analyse of existing systems, networks, hardware and software. Estimation of storage and retention of data requirements, physical condition of room housing system equipment and communications, power supply, entrance restrictions to sensitive areas, policy on removal of media from buildings, printout handling, access by young people to IT equipment.
Issues are: Replacement of existing communication links, introduction of encrypted traffic, offsite storage of backups, disaster recovery, software licences, fire detection and suppression, volumes of data transmitted and stored. To separate young people's IT equipment.
System requirements specification Hass et al (2007) (4) explains requirements analysis and specification as the activities involved in assessing information gathered from the organisation involved regarding the business need and scope of the desired solution. The specification is the representation of the requirements in the form of diagrams and structured text documents. Tan (2005) (5) describes the use of a rich picture as ‘a structural' as opposed to a ‘pictorial' description. It allows the practitioners to use any form of familiar symbols to depict activities and events, plus taking into consideration, conflicting views. The definition of a use case (Seffah et al 1999) (6) is a simplified, abstract, generalised use case that captures the intentions of a user in a technology and implementation independent manner. Use case modelling is today one of the most widely used software engineering techniques to specify user requirements. Dittrich et al, (2002) (7) suggest a new approach to development they term ‘systems development as networking'. They go on to suggest the key questions to ask is ‘How do systems developers recruit and mobilise enough allies to forge a network that will bring out and support the use of the system'. Unified Modelling Language (UML) is described by Arrington and Rayhan (2003) (8) as a method, employing use case and activity diagrams for requirements gathering. They state that use case serves as a view of the overall use of the system, allowing both developers and stakeholders to navigate the documented requirements. _________________________ 4. Tan (2005), p67. 5. Seffah et al (1999), p21 6. Dittrich et al, (2002), p 311. 7. Arrington and Rayhan, (2003), p28. 8. Hass et al, (2007), p 4. Rich Picture People Activities Current system Future system Use Case Diagram See Appendix A
Primary Scenario The likely outcome when the project specification is delivered is that the funding body will agree to the bid, but subject to some changes, which will reduce the overall cost. This will involve a degree of compromise in the design of the new system. Suggestions may be made to re-enter the Excel data and to delay the phasing out of the financial system. This would mean a phased project with an all-encompassing solution left to a later stage. The impact may be additional effort on the part of CSA staff. The system needs to be delivered in phases, with core functionality first. The successful delivery of core components will assist acceptance. A key component is the security of information stored and transmitted, as much of it is of a sensitive, personal nature. The protection of information will require conforming to the requirements of the DPA (1). Due to the number of area offices, with few staff, the data repository will require to be centralised, probably at HQ. This is for simplification of backups, which will require to be weekly full, stored offsite, and daily incremental with the last day stored on site. Communications between HQ and branches requires to be encrypted, and e-mail will require protected Internet access. Anti-virus, anti-spyware, anti-phishing and spam filtering software will be required, and a firewall introduced between the Internet-facing component and the main system. Rigid field input will be required to avoid erroneous numbers or characters. Menus will be restricted to selected functions and denied to others, and Admin level (privileged) control will be able to access all menus. The training for IT clients will need to be on a separate network segment from the main systems. Compatibility between the existing financial system and the new system will need to be established, and the system will require the capability to import Excel data. The system will be required to replace the functionality of the Excel data.
Developer questions to CSA staff: How much funding is available for the proposed system and who are the stakeholders? What facilities for computer systems exist at HQ: power, space, fire suppression, telecoms, operating staff, storage required, and records retained? Who will support the new system when delivered? What configuration does the finance system have: hardware, operating system, application software, network links, storage, number of users, support? Will staff time be available for training? Will only CSA staff use the new system and will they use it from home? Will there be allocation of CSA staff for user acceptance?
Discussion of requirements analysis tools The usefulness of the semiotic framework is that it offers the system developer an insight into the attitudes and feelings of people who will use the proposed system. This aids the developer, in that he/she is more likely to pay attention to the human-computer interface (HCI) aspects of the system. This should, if properly delivered, make the new system easier to use, and consequently, be received with more enthusiasm, than might otherwise be the case. A key message that core aspects should be delivered first, rather than the full functionality required, may win more converts than might otherwise be the case. Also revealed by the use of the Semiotic Framework was the attitude of some of the staff, who sees the requirement for the new system as superfluous to their ‘real' work, and consequently wish no contact with it as they are too ‘busy'. This helps the developer, as it brings home the need for the employment of techniques to make the system simple to use and not forbidding in terms of error messages to may produce due to inexperience. What the round of interviews in the case study revealed was some conflicting attitudes among CSA staff. A key example was mention of the need for protection of financial information, but the requirement to protect personal data of clients of the CSA, some of whom may have criminal records, was not mentioned. Given that failure to protect this type of information could lead to more harm to the individual than any help they may receive from the CSA, this is cause for concern, and seems to indicate that some of the CSA staff have lost sight of the organisation's mission in life. The interview process resulting in the case study report produced a lack of vital information any system developer would require to produce a workable system. Basic items were not uncovered. As an example there is no information on number of users, estimates of amount of data to be stored, how long it is to be retained, and what kind of systems are in use at the moment. The availability of capital and revenue information was not discovered, and it may well be that the funding will be dependent upon the proposed design in terms of capital and revenue costs to operate. The use of rich picture and case diagram illuminates the overall view of the required system, allowing the developer and the recipients of the system to see the whole picture and gain a better understanding of the likely finished product. It also simplifies the dependencies and collaboration required in a pictorial from which makes the ‘big picture' easier to understand. The importance of the Semiotic Framework is that it helps shed light on areas which the developer, using traditional systems development methodologies, may neglect. It concentrates the mind on the human-computer interface required, and influences the design attributes which need to be built in, in order to gain user acceptability. Taking the step-wise approach down through the levels, brings home to the systems developer, the need to start with the social needs, which focuses on the human aspirations (or not) of the proposed system. Working through the Pragmatics is very revealing of the contradicting attitudes of the potential users in conversation, and should lead to the developer making compromises between technical elegance in the design and being able to obtain a favourable reaction from at least a majority of the eventual users. The scope of the system required to be developed has not been revealed during the case study, which impedes the ability of the developer to estimate size and nature of the hardware or software required. The Syntactic level assists the design in that it forces concentration on the logical handling of data input, with system response to incorrect entry, being handled not with abrupt error messages, but more friendly advice messages and suggestions on data re-entry. This tracks back to the importance of human reaction learnt from the Social Word level. The software chosen should be influenced by the Pragmatics influence in that the choice should reflect the fact that the CSA is a charity, and both hardware and applications should be in the affordable range for an institution dependent upon charitable funding. The Empirics portion of the framework should include the estimation of required system performance, speed of telecommunications, volume of data to be stored, and response times of the system. In the CSA case study there is no information which can be used to project such requirements, so the developer would be required to utilise an educated guess, based only on the existing finance system, which could be measured, or practical experience. Some of the required information may be gathered by contact with whichever vendor delivered the existing finance system. The framework also draws attention to peripheral items, such as the Excel spreadsheet, which may well contain valuable data, not subject to strict input criteria, and possibly not backed up. The Physical World portion of the framework focuses the developer's mind on what will be acceptable to the users in terms of speed of response, the time and effort potentially to be saved, and the type of reporting of information capabilities of the system. It emphasises that there needs to be demonstrable benefits in the way of management information, and therefore capability to respond, which would otherwise have been unavailable. From a system developer's point of view, this is probably the section he/she would feel most comfortable with, as it consists of tangibles, which can be translated into MIPS, baud rated, gigabytes, and other terms which IT developers are expected to be completely conversant with. Probably the most difficult aspect of the framework for the developer is the Semantics level. The reason for this is that it tends towards the abstract, and system developers as a breed, operate mostly in a practical, exact, measureable fashion. They act as a translator between the business requirements as expressed by the stakeholders and eventual users, and the technical people who deliver code, hardware and communications to realise the stated needs. The developer has to perform a balancing act between what is sometimes conflicting requirements and technical possibilities. This required the ability to converse with, and understand, both participants in the overall project to deliver the required system. The use of the Semiotic Framework leads the developer to address these issues and attempt to develop a clear understanding of the CSA business activities, as opposed to trying to force fit them into a prejudged idea of the system. The developer may reflect that the application of the Semiotic Framework forces undue attention on the people-related aspects of system engineering, to the detriment of a design which embodies good technical practice and the necessary protective aspects required complying with any legal implications. Against this, the aim of developers to attain elegance and efficiency in design may be meaningless to the users of the system, whose main concerns are to assist in the capture of information, its ease of retrieval and the management information it can produce. In short, how it can help improve the user's work practices and make life easier for them.
References Arrington, C.T., Rayhan, S.H., (2003), Enterprise Java with ULM, Second Edition, Wiley Publishing, Inc., Indiana, USA, p28. Clarke,S., Elayne, (2003), Socio-technical and human cognition elements of information systems, Idea Group Inc, p 8. GOOD Diagram. The Data Protection Act Available from: http://www.ico.gov.uk.what_we_cover/data_protection.aspx. Hass, K.B., et al, (2007), Getting it Right, Management Concepts, p 4. Kangassalo, H.,et al, (1995), Information Modelling and Knowledge Bases, IOS Press, p 358. Seffah, A. et al, (2005), Human-Centered Software Engineering –Integrating Usability in the Software, Springer, p 21. Stamper,R.et al, (1987, Critical Issues in Information Systems research, Wiley, Chichester, p 47-78. Tan, J.K., (2005), E-health Care Information Systems, John Wiley and Sons, p 67. Tipton, H.F., Krause, M., (2007), Information Security Management Handbook, Edition 6, CRC Press, p1290-186-587