Faculty of Communication

University of Canberra, Australia

‘Knowledge and power circle each other like figures on a Greek vase.’ (Zuboff 1988)

An analysis of reforms to the public sector leads to a discussion of new approaches to organisational structure and management. A fractal metaphor is offered to explain how organisational structures and information technology combine on different scales, using Considine’s model of the policy process. Participatory practices and interactivity in IT are linked, giving examples from current federal projects. Several case studies under way in Canberra of interactive technologies and internal policy development are described, and finally, a model for increasing lateral accountability through appropriate application of interactive technologies is proposed.

Many Commonwealth agencies have been early adopters of interactive technologies in their communication programs. Figures from a survey conducted by the Office of Government Information Technology (OGIT) indicate that as of May, 1995, at least 45 agencies had Web home pages and were disseminating information electronically via Public Networks. Email access for public servants has also grown rapidly, with nearly 22,000 connected. These numbers are now likely to be much higher, with at least one agency, CSIRO, successfully implementing a universal Internet access program.

As in private enterprise, these technologies have co-existed with a range of other trends which are affecting structures and management practices. Since approximately the early 1980s, researchers have been studying the implications of interactive technologies for organisations. (Zuboff, 1988) The field of computer-mediated communications, or CMC, has documented many subtle shifts that occur to work practices as interactive technologies are introduced. There is growing awareness that communication technologies are both a cause and a consequence of structure (Fulk 1993) The role of advanced computer systems in managing business flow and providing feed-back loops for faster decision making is acknowledged as essential, and is increasingly being applied to government processes. Before discussing how this is happening in the Australian Federal Government, it is useful to outline the overall framework of change to government systems, and how this affects the applications of information technology.

The increasing application of ever-broader computing systems to government may be considered part of a wider convergence between the public and private sectors, which places a high regard on ‘instrumental’ efficiency and accountability. The standards and practices of the private sector have been held up as models for the public sector, which is often considered a barrier to economic progress. (Ives 1994) The reforms to the public sector which seek to make it more business-like have been widely documented, and are global in their application (Zifcak 1994, Peters 1996).

Although there is variation and even contradiction in the way these reforms have been applied (Peters 1996), the general features have been characterised as:

adoption of business plans and revenue targets; introduction of guidelines in place of inflexible rules; a move away from centralised controls; application of market practices such as user pays; more flexible work and pay arrangements, and more emphasis on client needs (Ives 1994)

A different analysis describes four key concepts as: an emphasis on quantifiable, single purpose economically measurable outputs, neglect of the political dimensions of public organisations, a focus on integration in preference to decentralised forms of service development and delivery; and an unwarranted optimism about the potency of technical rationality under central direction. (Considine 1988) This last is simply the mutton of technological determinism dressed up as a lamb of economic rationalism.

These reforms have been criticised as post-Taylorist, and ‘offering few prospects for genuine and lasting reform’ (Considine 1988) Zifcak also concluded that these reforms had little effect in either the UK or Australia ‘beyond the sphere of operational administration’. (pg 191) Zifcak notes that part of the problem was the failure to deal with the political dimension of the problems at hand, through ‘effective presentation, analysis and evaluation of competing policy prescriptions.’ (pg 44) This does not imply, however, that reforms have no political element. Rather, it may well reflect Guy Peters’ observation that seemingly contradictory approaches can take root in the name of public sector reform precisely because ‘reform of the public sector is a political exercise that is rarely, if ever, informed unambiguously by organisation theory. ‘(Peters 1996)

An over-reliance on business models which equate citizens to consumers or clients has also been rejected by Henry Mintzberg (1996) who has called for a better balance is the range of organisational structures available for different purposes. He notes some of the myths of management, and reminds us that many activities are in the public domain precisely because of the difficulty of attaching a set of clear measurements, outcomes and responsibilities to them. The need for a diversity of structures has also been noted from an economic perspective, as over-reliance on competition places an emphasis on short-term survival, and diminishes the economic benefits available through collaboration and cooperation. (Tissdell 1997)

Particularly since the 1996 federal elections, an escalation of these trends, along with ‘down-sizing’ and a corresponding loss of job security has been evident in the Australian Public Service. The dominant structures remain strongly hierarchical, although this ‘paradoxically contradicts current private management precepts.’ (Considine 1988)

Indeed, there is a strong surge among management theorists to adopt more flexible, evolutionary approaches to organisational structures and processes. They call for the creation of ‘learning organisations’ which ‘empower’ employees and rely on an ethical commitment to goals which encompass, but are wider than, economic success or survival. (Senge 1992, Covey 1989)

Thus, although the public sector is being urged to become more like the private sector, over the longer term the private sector may ‘take up the slack’ by becoming more socially responsible. This has a host of implications, not the least for citizens rights and obligations in an age of information and data exchange. These issues are central, as described later, for information management strategies. For example, outsourcing may place increasing amounts of information crucial to the public good in the hands of private companies which are not subject to Freedom of Information or Privacy Act considerations.

These ‘new age’ management principles have also been described, if not often implemented, for the public sector. The ten principles for transforming the public sector recommended by Osborne and Gaebler may be considered a hybrid which recognises both the need for efficiency and the unique role and accountabilities of government. Issues such as decentralisation, increased participation, better partnerships with the business sector and empowering communities also form part of innovative adaptation schemes for regional areas which have lost their traditional economic base. Thus, Jin and Stough (1996) describe an ‘agile city’ which uses advanced technologies to share information rapidly among the education, business and government sectors, in order to create a ‘learning’ metropolis.

From a different perspective, but one which informs advanced systems theorists, complexity theory and modelling of self-organising systems has been used to point out the irrationality of neo-classical economic theory and highlight the central role of information exchange in sustainable systems of all kinds. (Waldrop 1992, Allen 1996)

On one level, recognition of complexity theory can be seen in the increasing acknowledgment of the role of contradictions and paradox in management thinking. (Peters 1996) On another level, it feeds into concepts such as Senge’s (1992)‘reinforcing loops’. These resemble the ‘increasing returns’ which for a long time were considered heretical by classic economists, who wanted to believe in perfectly balanced rational markets the way the Catholic Church wanted to believe the earth was the center of the universe. (Waldrop 1992) Another direct connection comes from the literature on artificial life. Emergent, adaptive systems arise from extremely simple, bottom-up models. The incredibly complex and functional behaviour of a flock of birds, for example, can be specified with only three rules. Top-down models, on the other hand, ‘are forever running into combinations of events they don’t know how to handle, and tend to grind to a halt in a dither of indecision.’ (Waldrop 1992, pg 279)

There is not space in this paper to go into detail about the relations between complexity theory, information economics, and the emergence and diffusion of communication technologies. It is sufficient here to note that there is currently a resurgence of interest in the ideas of Joseph Schumpeter, which emphasises the evolutionary aspects of economic processes. Interdisciplinary approaches highlight the importance of diversity in biological as well as economic systems. (Foster, 1996)

These approaches demonstrate the non-adaptive aspects of uniform solutions for either organisations or economies. For example, Tissdell illustrates the elusiveness of ‘best practice’: ‘Even if all firms in an industry were to adopt the best practice techniques, this may not be optimal, because they would all replicate one another and all diversity would be lost.’ He believes the aim should be ‘responsive economic system’, where ‘preservation of institutional diversity satisfies the precautionary principle.’ He concludes that ‘Much competition and public technology policy is being guided by out of date and largely irrelevant economic theories.’ (Tissdell 1996) Modelling of fisheries and urban centers demonstrates that without the bubbling up of random ‘chaotic’ elements, these systems become entropic. (Allen 1996) In other words, vibrant, adaptive systems ‘feed’ on a diversity of information.

Work from such diverse fields may seem to hold little relevance to the application of information technology to policy development. One link is Considine’s definition of policy as a process of both innovation and communication among systems of actors.(Considine 1994) Taken together with Clayton’s findings showing innovation is more likely in a ‘learning’ organisation (Clayton 1995), we can conclude that there is a positive relation between policy innovation and systemic learning. Within Australia, the tension between managerialist vs evolutionary, learning approaches can be clearly seen in the current documents on information management in the Commonwealth. The OGIT explicitly notes the centrality of information for government:

‘Information is at the heart of all government activity -in policy making, public dissemination and feedback, taxation, purchasing, provision of benefits under government programs, and the public accountability process - all are dependent on the effective handling of information.’

And as Considine points out, ‘Management no longer can be considered external to policy as in legal-rational models of corporations and bureaucracy.’(Considine 1994, pg 232) Thus, the management structures or ‘mental models’, to use Senge’s term, which prevail will set the framework for information exchanges which are, in turn, the bedrock for policy deliberations.

The key issue here is democratic process. On a global scale, it has been argued that new communication technologies such as the Internet are a product of global corporate capitalism, and are therefore likely to exacerbate inequalities. (Lyon 1988, McChesney 1996) On the other hand, the many experiments with on-line democracy, community networks and discussion groups may illustrate the growth of electronic empowerment and virtual communities.

Which model holds sway in Commonwealth strategies towards IT? The short answer may be ‘both.’ But we may need a ‘fractal’ metaphor to help us understand how. Fractal patterns are the same on all scales. Examples from the physical world include snowflakes and coastlines. Human organisational structures may also ‘self-organise’ into fractal patterns, given the right circumstances. I maintain that the convergent forms of computer technology currently in wide use, along with similar models for corporate and government structures, create a tendency towards similar patterns of outcomes, conflicts, and possibilities. Exploration of examples which ‘break the pattern’ can both clarify the overall trend and suggest alternatives paths.

Thus, autonomy vs top down control may be an issue within a state agency (Considine, 1992) which is echoed in Commonwealth-State relations, and then in international fora. In relation to IT, the platform for such conflict may be seen as an individual’s wish to access colleagues via external email, an agency’s claims to maintain an independent payroll processing system, or in global resistance to domination of Internet content and costs (Krause 1996).

At last year’s Internet Conference, Vint Cerf, one of the founders of the Internet, commented that ‘democracy doesn’t scale’. Leading political theorist Guy Peters made a similar remark at an ANU lecture, to the effect that ‘dialogic democracy’ is very difficult in a mass society. This is interesting, given that commodity prices, marketing figures and banking data can, and is being collected, collated and acted upon with lightning speed and efficiency all over the world. An explanation may be found in the political economies that direct outcomes of organisations, regardless of scale.

The crucial influence of organisational and social structures on applications of IT is corroborated by the literature on CMC. These resemble the political economies described by Considine. (1994, pg 10) Early hopes that email or electronic groupware would, of themselves, foster more casual, democratic relationships between staff and supervisors have been repeatedly tested and discarded by researchers. (Mantovani 1994, Nunamaker 1991) While the asynchronous, uniform nature of these technologies eliminates some social cues (Sproull and Kiesler 1986), existing social structures will dominate the adoption patterns and uses for these technologies (Fulk 1987, Perin 1991). Rather, the self-replicating nature of internal relationships means collaborative effects tend to occur laterally, rather than breaking through vertical structures. (Perin 1991) Clement (1994) found that ‘empowerment’ was used in conflicting ways, and that often the ability to ‘act to create value’ was the result of computing innovations in leaner organisations whose focus was improvements to function and efficiency. He contrasts this with ‘democratic empowerment’ which would involve lower-level staff in discussions about the conditions of work and setting goals.

This ‘instrumental’ use of IT for doing more of what was already being done, rather than exploring new ways of working, has many roots in organisational history and internal power structures. It reflects an under-investment in the ‘soft infrastructure’ of training and a reluctance to deal with the underlying implicit threats to managerial power and the forces of counter-implementation. (Benjamin and Levinson 1993, (Keen) Such conflicts can be traced back to the beginning of the industrial era, as richly described in Zuboff’s powerful work on the introduction of computerisation to a set of firms in the 1980s.

In ‘The Age of the Smart Machine’ (1988) she carefully illuminates the distinction between automation and ‘informating’, which uses technology as a tool for learning, growing, and reskilling, rather than deskilling and increasing ‘panoptic control’. Because the computerised process goes beyond mere automation in creating a new stream of information, which ‘provides a deeper level of transparency to activities’ (pg 9)

Because this capacity is hierarchically integrated (pg 11), ‘Managers can choose to exploit the emergent informating capacity and explore the organisational innovations required to sustain and develop it. Alternatively, they can choose to ignore or suppress the informating process.’ (pg 11) These choices ultimately concern ‘the conception and distribution of knowledge in the workplace.’ (pg 5)

Again, there is not space to elaborate on these concepts, which have influenced many researchers of technology and managerial theory. Rather, their relation to the application of IT to the public sector, and to theories of the policy process, is our focus here.

Canadian Peter Heimler describes in some detail how the ‘productivity paradox’ takes hold in hierarchies that implement IT without making the necessary shift in ‘mental models’. Questioning the ultimate benefits resulting from the introduction of computerisation is not new (Attewell 1996), and is possibly growing, as the dependency on these technologies increases. Also, the basis on which predictions of economic benefit are made may reflect a bias towards the commodification of information and a deliberate neglect of social issues. (Preissl 1997)

Heimler argues that IT cannot improve productivity in current government hierarchies, because the sheer volume and velocity of information overwhelms the decision makers at the top. Thus, new IT systems tend to bog down because staff and clients haven’t been involved in legitimate consultation, and bottom-up communication is suppressed. Solutions then tend to mirror beliefs of those at the top, and as the crisis increases, the ‘need to be seen to be doing something outweighs the need to be thinking about things’. Despite corporate plans, Heimler says government fails to act as a mediator, as the ‘bureaucratic model does not focus on value, creativity, vision, problem solving and client service.’

One outcome of this bleak scenario is that ‘policy and program responses tend to look and feel the same across a range of portfolios’. Drawing on the literature of learning organisations, he suggests that one way out would be to truly devolve decision-making from the over-loaded top, and look for more lateral forms of accountability. However, given that such approaches involving team building and bottom-up decision making are ‘knowledge driven’ (rather than power-driven), more often the outcome is to ‘further strengthen the centre of government, adding more and yet more politicians, bureaucrats, experts and computers in the desperate hope of outrunning the acceleration of complexity.’

Heimler’s analysis bears a striking similarity to Mark Considine’s (1988) critique of the corporate management framework, which disguises tighter central control with devolved control over resources, but not policy. The contributions of all but the executive are seen as primarily and essentially technical, as agencies are forced into ever greater integration, uniformity and consistency among government services and activities.

If this is the prevalent mind-set, then the relatively modest recognition of the potential for electronic consultation is no surprise. As one government official succinctly put it: ‘the technology is lateral, but the structures are vertical’. Other anecdotes are more revealing: when offered an electronic list for wide discussion of a particular policy issue, the shocked reply from an executive officer was ‘You mean we wouldn’t be able to control who we communicate with?’

Alternative ways of working which challenge such paradigms are difficult to introduce, and are thus rarely costed or measured. Costs to the Commonwealth in 1993/94 for IT to support service delivery alone was estimated at $1.4 billion, with possibly several hundred million dollars in additional costs per year due to ineffective use of information management tools. Methods and technologies which build on existing patterns, offering more speed or less time, are more welcome. As we shall see later, improvements to client services are often the major focus of government IT efforts, so much so that other possibilities of ‘informating’ are considered tangential, with a few notable exceptions. As one senior officer involved with the area observed: ‘The service delivery people have hijacked the IT agenda.’

A multi-dimensional model of the policy process, such as put forward by Considine (1994), offers a means of linking current activities involving IT to wider policy formulation in all their complexity, diversity and sometimes, contradictions. Seeing policy as ‘the continuing work done by groups of policy actors who use available public institutions to articulate and express the things they value.’ (pg 4), he describes the rich web of actors, dependencies, historical and cultural influences as well as the mutability of organisations and values which combine to create ‘policy’.

Along with Heimler, Considine (1994) emphasises the importance of participation, and its potential to create social capital from which ‘all central democratic objectives spring, including legitimacy, co-operation and innovation.’ (pg 130) Returning to our fractal metaphor, participation may be a crucial value on every scale. Heimler discusses it primarily in relation to the internal decisions which affect application of new IT systems, and their flow-on to the wider community. This internal policy level is the subject of the two case studies in this paper. Considine’s model applies broadly to several layers, although he focuses his examples on state and national levels. Other writers express concerns that international corporations are making the decisions which will affect access to information and participation in decision making well into the future, and they are not asking for alternative opinions. (Mander 1997 and Preissl 1997)

The overlooking of important alternative strategies is one obvious drawback of non-participatory policy formulation (Considine, 1994, pg 134) However, die-hard managerialists would clearly see this as a reason not to consult. The link between scales is important, as individuals are policy actors on several levels. Dick Sclove, of the Loka Institute, suggests that greater democratisation of the workplace, along with a better integration of community and work will give people the opportunity to learn about cooperative and participatory methods of decision making.

It is ironic but not trivial that people need to learn about such fundamental processes; Senge found that long experience of compliance left people bewildered and even suspicious when placed in situations that called for commitment, much less vision (Senge, 1992, pg 219). It also seems to be the case that while intentions to become market-driven and client-oriented are accepted without qualm, mention of improving democratic process or participation is considered ‘ideologically driven.’ Evidence such as that provided by Considine (1992) that lateral and collaborative working structures can produce observable efficiencies does not seem to reach a wide or receptive audience.

The aspect of IT which shows greatest potential for encouraging participation is interactivity. This can be either individual, group, or massed. If the ‘dumb terminal’ typifies a passive and powerless relation to a technology and its masters, then the World Wide Web is the model of unbounded, open-ended communication, where producers and receivers of information entwine freely.

Thus efforts to use interactive technologies for genuine empowerment are valid starting points for the discussion of wider participatory policy efforts. A range of initiatives in this direction have sprung up within the Australian Public Service over the past several years, with the wide-spread introduction and dramatic growth in the use of interactive technologies in the Commonwealth. On both internal and external scales, a rich complexity emerges. Many factors affect the modest stirring of electronic democracy within the Australian Public Service, including security concerns, existing structures, the novelty of the technologies, and drives towards cost-savings. (Geiselhart 1996)

Several of these initiatives offer insights into the possibilities for the uses of IT in policy work. The Community Information Network (CIN), established by the Department of Social Security, was an ambitious project which intended to encourage development of local community content. The public access network, via CIN sponsored PCs, was withdrawn last October, leaving an interactive Web site with discussions for specialty groups, and information about DSS entitlements and policies. There are separate ‘rooms’ for seniors, youth, women, the disabled, etc. It looks like it continues to be managed by dedicated team of community-oriented computer enthusiasts, who invite comment on the various web pages, partly for evaluation purposes. It is not clear whether comments on policies, rather than comments on web presentation, flow through to policy areas. However, the loss of the public access network definitely diminishes the possibilities for local interactivity, at a time when general directions for Australia’s information channels is evolving rapidly.

The Education Network Australia, (EdNA) was always conceived as a de-centralised network, partly reflecting Commonwealth/State demarcation issues. It intends to operate several dozen mailing lists, newsgroups and online discussions on a variety of education-related topics. These, like the CIN discussions, are examples which would cost government very little, while providing a seed-bed for participation. It is still early days for such experiments, as Australia probably does not yet have a critical mass of users among the most crucial groups. Access and training present sizeable hurdles, and there is little talk of a universal access policy, comparable to our telephone standards.

Departments and agencies have Web pages, with few exceptions. Many provide valuable links to international sites, and give overseas Internet ‘visitors’ a look in at our policies and programs. Increasingly, forms and personal information will become accessible, and electronic commerce with the Commonwealth will become widespread. The number of draft documents for comment, press releases, and publications on-line increases almost daily.

Both the benefits and the limitations of such initiatives can be seen in the PUBSEC mailing list, sponsored by the Department of Administrative Services. It offers a forum for the discussion of public sector issues generally, and postings include much useful information, as well as a few sighs and brickbats, gently monitored. The initiator of the list, Denis Strangman, had a clear view of the potential for such discussion to contribute to policy development, particularly with a mix of international bureaucrats, private sector subscribers and academics. (Strangman 1996) An evaluation of PUBSEC reflected the benefits of such a collaborative setting, and after some hesitation, DAS has agreed to continue its support, at least until other arrangements can be made.

PUBSEC illustrates the difficulty that many participatory initiatives meet in the public sector. Because their benefits are long term, vague, and not easily quantified, they find a cool reception from senior managers. Officers with the most to contribute are often passive ‘lurkers’, with PUBSEC subscribers in the Department of Finance occasionally prodded for their opinions. The pattern found in CMC of horizontal collaboration (Perin 1991) tends to be repeated, with members clustering at the middle management level. Flow-on to the subscribers’ home agency is uncertain, implicit and tangential, rather than intentionally exploited. Officers tend to stick with factual exchanges, noting that they haven’t time for philosophy. Prominent disclaimers distance their statements from the official position of their agencies, although, as one Finance officer observed, ‘I don’t believe there is any such thing as an off the record conversation.’ The ‘textualising’ of communication on interactive systems (Zuboff 1988) encourages a cautious approach. Such lists also tend to have a life cycle, with ebbs and flows of interest and intensity.

Thus, open and encouraging of participation as it seems, a list like PUBSEC is unlikely to have a formal role in policy development. Structural and cultural norms and expectations ensure that such effects are dampened from within, unless informed and motivated influences higher up the ladder recognise the potential such forums offer.

Many other projects are underway at state and other levels which overtly seek the leverage and efficiencies which sophisticated use of interactive technologies can provide. A project to develop software specifically for policy development is underway in the Department of Defence, but faces uncertain funding.

Overall, the initiation of interactive projects in the Commonwealth fits the policy model proposed by Considine. Players on different levels make proposals, and some get up and running, for a variety of reasons. The desire of agencies to be part of the Internet revolution, or the pressures to become more cost-effective, lead to a variety of programs, some of which survive, and others which collapse. Dependencies and legitimation generally bow towards upper structures, leading to an emphasis developing measurable efficiencies in centralised, large-scale client - government projects. Citizen-to citizen communication, or development of citizen-government interaction in the fullest sense is given a much lower priority.

But even at the highest levels of policy formulation for uses of government IT, a diversity of views injects life into the process. Over the past year, there has been some movement towards recognition of the importance of the potential of strong interactivity, and its potential in policy work.

The mission statement for the OGIT focuses on the instrumental value of these technologies:

‘To add significant value to agencies' application of Information Management & Communications in improving program delivery and administration’

Likewise, the vision for the Framework & Strategies for Information Technology in the Commonwealth of Australia - Exposure Draft, December 1995, has an orientation towards achievable economies and efficiencies:

‘The Commonwealth will be a world leader in government administration and in the cost-effective provision of affordable, equitable and accessible Australian government information and services.

Australians will have seamless access - that is, through common interfaces - to a range of government information and services appropriate to the client group, wherever it is cost-effective to do so.’

What is missing is Zuboff’s ‘informating’ dimension, an explicit intention to improve the fundamental processes of government through citizen’s participation. This is a common oversight in government documents of this kind. The UK Green Paper makes it clear from the title: ‘government.direct: A prospectus for the Electronic Delivery of Government Services’, that it is offering a business model. Its intention is to

‘change fundamentally and for the better the way that government provides services to citizens and businesses. Services will be more accessible, more convenient, easier to use, quicker in response and less costly to the taxpayer.’

The examples given are drawn from business, and the overall pattern very much resembles what one would expect from private enterprise, but no more. There is much mention of ‘rationality’ and cost savings. There is some acknowledgment of the potential for participation in the policy process (via email) and of citizen’s rights:

‘In addition to improvements in service delivery, the proposals in this Green Paper will help citizens to involve themselves more in the democratic process. Both Citizen’s Charter information and basic statutory information will be widely accessible electronically.’(pg 31)

A draft Community Information Strategy for the ACT listed key information which should be available electronically, but omitted urban planning issues and participation from this inventory.

A happy contrast is provided by a recent Australian Government document on ‘Management of Government Information as a National Strategic Resource’. Here, too, the title is an accurate guide to the tone and intent of the document; in this case to provide a thoughtful and comprehensive outline of the issues involved in using government information collectively for broad national objectives.

While the terms of reference for the Information Management Steering Committee (IMSC) are modest (pg 1), the document shows a pervasive awareness of the need to address social as well as economic issues, and the potential to create a framework for individuals to interact richly with government, rather than simply as clients for government services: ‘it will no longer be acceptable for government to consider clients as transactions waiting in a queue. Australians will become better informed individuals….’ (pg 10)

The need to integrate business plans and achieve cost-effectiveness through information management is an important component, but this is not viewed in isolation. Technical and legal aspects of information management are placed in a values framework, informed by theoretical understanding and discussion of information society concepts and prominent overseas projects, such as North Star in Minnesota. There is no backing away from recognition that the Australian Public Service will need to change from within to meet these challenges (pg 9).

A section on Electronic Democracy goes well beyond the platitude offered in the UK paper, and suggests that the Internet can assist by allowing earlier and wider input into policy processes. This is an example of the ‘informating’ aspect of information management. It also offers the possibility of greater lateral accountability, as several writers have recommended. (Au Coin 1991, Considine 1994)

However, it would be premature to assume that a benevolent or even a uniform approach to government information will prevail. The number of actors on the information management stage is large, and they are all still learning their lines. (Illustration 1) Considering the scope of the issues covered, diversity and conflict of views is to be expected. Approximately a dozen players are already involved in electronic commerce. An officer with an interest in the IMSC report said that electronic democracy issues were deliberately treated lightly, so as not to stir up a backlash. Another IT officer said compliance with OGIT’s plans to outsource IT operations was ‘theoretical’ possibility.

This playing field fits the model proposed by Considine (1994), and also the fractal metaphor presented earlier. Commonwealth officers and agencies have dependencies and relationships of differing potency with their ministers, and with overseas colleagues, through such collaborative projects as the Global Information Initiative.

Having given an overview of the management and government background which positions my overall research, I will now zoom in on a smaller scale, and the two agencies where my current case studies are underway.

Both studies center on the use of internal bulletin boards in corporate processes. However, they reflect different corporate approaches to information management and participation. The first agency uses one all purpose newsgroup to discuss management issues. While still in the preliminary stages, data so far from two months of postings indicates a disgruntled staff with strong divides between the executive and lower levels. Numerous messages mention the need for greater trust and formalised structures for participation in corporate change. Fewer than ten per cent of staff read the postings, with very little involvement from the higher levels. There was little change in the six months between sets of postings, except perhaps for an increase in cynicism.

The second, more developed case study is with the Department of Finance, where interviews have been used to supplement a survey sent via internal email to all staff. Initially, I was directly towards the Bulletin Board system (BBs) at DOF on advice that they were used for policy work. Initial liaison was with the Information Group, which was interested in evaluating the BBs as part of their efforts to improve software access for team work. After several months, the project was taken over by the Future of Finance group, which had an interest through their work on information management for DOF. Thus, the project was always intended to serve internal needs, as well as the researcher’s more abstract interest in how this particular interactive technology contributed to the agency’s internal policy process.

The Department of Finance (DOF) is a particularly interesting one to study, as it is a central coordinating agency, and as such, a key shaper of government policy. It played a central role in public sector reform during the 1980s, leading the way by devolving some functions while setting up clear accountabilities (Zifcak 1994) Also, as the controller of the government purse, it may be expected to be advanced in its use of information technology. The Department espouses a learning culture, and was working on information management plans well before the Commonwealth urged agencies to address this issue.

DOF set up a Bulletin Board system on its cc mail software about 3 years ago. There are now about 38 BBs, with varying turn-over times and levels of access. A number are used throughout the Department, accessible by all staff for information about staffing, facilities, women’s issues, and other corporate matters. Some of the Bulletin boards are for staff suggestions of improvements. Others contribute to essential policy processes, such as the preparation of budget documents. The bulletin boards interact with other technologies, such as groupware and an internal mail system, to form a complex set of information flows and arrangements, both formal and informal. Many of the patterns are tacit, and few are clear from departmental documents about IT.

As the preliminary meetings with DOF staff progressed last year, it became clear that the use of the bulletin boards for internal policy processes was very much part of wider IT use within DOF, and wider information management issues. Comments made casually by staff were rich in their implications, indicating the importance of interviews for uncovering pieces of the DOF IT puzzle.

As mentioned above, one impetus for the study was a desire, within DOF, to canvass the possibilities for implementing more sophisticated forms of interactive software, such as Intranets and groupware. Thus, key people in the organisation are seeking ways of leveraging knowledge both across and within areas, for purposes of improving performance and avoiding duplication. Some officers have indicated a desire to improve communications between the Department and other agencies as part of measures to implement a learning environment. DOF officers maintain the APS Innovations site on the World Wide Web, which has the potential for implementing one aspect of ‘policy as innovation’ (Considine, 1994) From another perspective, ‘agile cities’ are said to have an inter-sector collaborative orientation. (Jin and Stough, 1996)

Willingness to have an external researcher analyse attitudes towards an internal communication mechanism is, in itself, a substantial indicator of openness and willingness to learn. The generosity of assistance provided by staff, including during the busy period just before Budget, has also been crucial in formulating the questionnaire. In return, data based on a real life situation, rather than a laboratory set up, should prove more useful to the Department. Studies of computer-mediated communications which have relied on student experiments have produced doubtful outcomes. (Mantovani, 1994)

The methodology chosen is ethnographic, as a study of the overall corporate culture is intimately linked to the uses of IT. The research is qualitative, and emergent. That is, the shape and substance of the most important areas of study are emerging as the investigation proceeds.

A triangulation of data collection has been decided on, to ensure that each source is not skewed in relation to the whole, and to help see the interconnections between the data sets. An on-line evaluation questionnaire has been developed with the assistance of DOF staff familiar with the bulletin boards, and uses a 5 point Likert scale to assess the usefulness of the various BBs. The survey was run in October 1996 via the internal mail system.

In depth interviews are also being conducted with a range of staff, to seek additional information about attitudes towards the departmental information systems and DOF’s policy advising role. If possible, group discussions will complement the one-to-one interviews.

Documentary evidence will form the third aspect of the research methodology. A number of DOF documents, particularly those relating to the use of IT and the Future of Finance group’s advice on the policy advising role, will be studied as a background for the interviews, and as a means of checking informal and survey data against formal departmental attitudes and approaches.

Only 51 surveys were returned, out of a potential staff of over 900. There are both problems and benefits of using email surveys (Clayton 1996), and this result may partly reflect the overall information glut which respondents complained of. Replies came almost equally from staff at the Administrative Officer and Senior Officer levels, with only one Senior Executive Service respondent. No one took up the invitation to contact the researcher directly.

The BBs rated highly in their contributions to staff morale and social cohesion, having a say in corporate policies, creating a learning environment, and managing change. They were also considered to have a big role in overall work effectiveness, and as a means of speedy communication.

There were none of the pleas for increasing trust that came through so strongly in the other organisation, but rather indications of an ‘embarrassment of riches’, and requests to cut down and rationalise the number of BBs. This is likely to happen as part of the continued integration of the Intranet with pre-existing forms of electronic information management, which hopefully the survey data will assist. Several staff made useful suggestions which could easily be canvassed more broadly.

The bulletin board system seems to be well integrated with the sociological aspects of DOF corporate culture. For example, the Soap Box and Suggestion Box BBs allow broad discussion of sensitive issues such as religion, and censorship is rare. Informal protocols seem to operate which manage the flow and intensity of the postings. For example, the ‘Suggestion Box’ is actionable, meaning that someone is charged with taking the suggestions seriously, and posting a progress report on its feasibility back to the board. The appropriately named ‘Soap Box’ bulletin board, on the other hand, is more a ‘get it off your chest’ opportunity. A bit of hyperbole is tolerated. Flaming is rare, as is the action to remove postings for inappropriate language or content.

Exchange and sale of goods via an electronic ‘classified’ column and commentary on television programs and other local events is considered acceptable. This tolerance may reflect an internal culture which takes as given that Finance Officers work hard, long and well, and are therefore entitled to a few moments of electronic socialising.

Another indication of internal cohesion is the Women’s Network bulletin board, which does not exist in isolation from other Departmental activities, but is part of a thriving group which has respected input into DOF corporate policies. Their annual dinner was attended by the Secretary and other senior staff.

An approach which, consciously or otherwise, promotes an integration of social and business activities is likely to be more productive in the longer run, partly because it supports the development of internal social capital and an ‘agile’ organisation. Research has found that implementing IT-enabled change requires an equilibrium between the elements of organisational culture, technology and business process. Working with internal stakeholder needs is important for maintaining energy and commitment to ongoing change. (Benjamin and Levinson 1993) Results of the current research will be shared widely with staff. This may be taken as additional evidence of the presence within DOF of ‘individuals who adhere to a learning culture, do not fear change, and are truly concerned for the well being of those who they are trying to serve.’ (Heimler, 1996)

From initial interviews, it is clear that there is a diversity of views within DOF on new directions for IT and knowledge management. There is also acknowledgment at the highest levels that change is necessary to become an ‘info-sharing’ rather than an ‘info-hoarding’ organisation. On the other hand, the central IT area retains some aspects of a ‘control from the centre’ approach, echoing the findings of Heimler (1996).

The most difficult area for DOF, and the most challenging from an information management perspective, is their external policy advising function, which includes advising service delivery agencies on program development. They have data lines, but no interpersonal communication channels, on a wide area network to other agencies.

It is also an area where DOF’s role is often misunderstood and treated with suspicion by program departments. That is because DOF’s other, more well known policy advising role is in monitoring programs for costs and assessing each program proposal in relation to overall government priorities. Thus, there is, for DOF, no contradiction between advising a department on how best to structure a proposed program, and then to recommend against it when considering the proposal in an ‘all of government’ context. These ‘two hats’ are also not always well understood by DOF internal staff.

Recent developments within DOF, particularly in relation to information management, show that internal and external factors may have more collaborative working relationships in the future. Eleven Knowledge Networks are being set up, which relate to DOF’s most important policy issues, including Government Business Enterprises and the One-Stop Shop. They will facilitate the documentation and sharing of information and methods of working on these issues. Two of them deal with internal information issues and the Women’s Network. At least one offers self-selected membership, and invites participation from officers throughout the Department. They will attempt to bridge the gap between tacit and implicit ‘knowledge’ on these issues, which has been identified as a barrier to understanding the role and costs of knowledge (Lamberton 1997). Most of the proposed Knowledge Networks will have links to outside agencies. It is not yet clear to what extent these ‘Nets’ might invite external comment via information placed on the World Wide Web, and no specific mention was made of public or non-APS stakeholder involvement. However, these nets, as they evolve, promise to provide an innovative approach to policy development, both within and external to DOF.

It is widely recognised that for implementation of IT systems to be successful, there not only has to be agreement on what that would mean, but also the accountability for change has to be institutionalised (Benjamin and Levinson 1993). This is particularly important in the policy arena, where the potential of IT for truly opening the process has yet to be fully acknowledged or explored. Although there have been few studies of the political aspects of information systems development (Keen 1981), Considine (1992) found clear correlates between political orientation and receptiveness to alternative structures in a bureaucratic setting. The upper echelons of the public sector are drawn from a narrower social base than rest of public service, (Bunn and Ranald 1993)and slide easily into even more highly paid positions in the private sector. Such social factors cannot be regarded as incidental to policy outcomes. Interactive electronic technologies have the potential to contribute to increased lateral accountability and transparency, while enlarging the stage on which policy actors perform.