Critical Success Factors for a Collaborative Database in a Large, Geographically Dispersed Organization

Owen Ambur, University of Maryland University College, May 9, 1996


Abstract

Based upon a review of the literature, success factors for a collaborative database in a large, geographically dispersed organization include: commitment and support from top management; a workforce that is ready, willing, and able to implement and use such technology; and a clearly defined business process that will benefit from on-line collaboration. Technical support requirements are heavy. The most successful collaborations begin with a draft document and end when it is in final form. General purpose discussion is not likely to justify a collaborative database, even in the event that employees are inclined to use it for that purpose. Indeed, a discussion database may aggravate the problem of information overload. The most critical factor to be addressed prior to implementation of a collaborative database is whether it is the best tool to be applied to the most important business processes of the organization.


Introduction

Teamwork is as old as mankind, and studies and discussions of how best to facilitate collaboration, teamwork, and communications abound. In the relatively short existence of "personal" computers in the workplace, however, an even newer phenomenon is the rise of software explicitly designed to foster group activities.(1) As a class, such software is commonly called "groupware".

Baeker (1993) defined groupware as "multi-user software that supports computer supported cooperative work … such as problem solving or communication, carried out by a group of collaborating individuals." He suggested that "together, this technology and concept promise to revolutionize the use of computers." Not only does groupware hold the potential to revolutionize the way that people use computers, but at a more fundamental level, it can also change the way that people organize their affairs and relate to each other. Therein lies not only the potential but also the problem, as highlighted by Coleman (1995) in his overview of groupware:

… groupware is not just technology… [it] is collaborative technology… it impacts the way that people communicate with each other … the way that people work and eventually the structure of the organization… The difficulty that most organizations encounter with groupware is not with the technology…, but with the relationship between the technology and the people in the organization…
Khoshafian and Buckiewicz (1995, pp. 4-6) classify groupware in three main categories: 1) document and forms-based, 2) transaction-based high-volume information management, and 3) organizational communications. Among the applications included in the third category are databases used for collaborative purposes, i.e., on-line discussions and document drafting that might otherwise take place orally by phone or in one-on-one or group meetings.(2) Khoshafian and Buckiewicz acknowledge considerable overlap among the categories.

An alternate and perhaps clearer classification might address the structure and formality of the information to be processed:

Highly structured (and thus "formal" by definition) -- forms automation and databases, including electronic "meetings" that warrant such treatment;
Relatively unstructured but formal -- document management, including not only text, but also images, video, audio that warrant sharing, record-keeping, and periodic retrieval;(3) and
Unstructured (thus "informal" by definition) -- E-mail and other on-line collaborations that are transitory in nature and not focused on formal document, forms, or database management applications.(4)
By this classification, discussion databases would fall neatly and solely in the first category -- highly structured/formal -- thereby drawing a clear distinction from E-mail while at the same time avoiding confusion and overlap with document management and processing.(5) The actual purpose of discussion databases -- to supplement or substitute for meetings - would be clearly highlighted.

Whether databases can or should be used in lieu of meetings is debatable, but in the early 1980's de Sola Pool set forth some of the advantages and disadvantages of on-line collaboration that are not shared by face-to-face meetings:

Participants can take part from any place where they have access to a terminal. There is no requirement to have all participants logged into the computer simultaneously; they may participate at a time of their own choosing. This temporal freedom is especially useful … where participants are in different time zones. The absence of time pressure afforded by computer conferences permits thought-out answers. Conferencing programs exist that can poll participants on their views or votes. The absence of physical interaction also reduces the social pressure toward "group-think." It also eliminates obstructionism by one member who monopolizes the meeting time by frequent and lengthy contributions. In computer conferences, other members can skip over such material and discourse among themselves. Participants can skip those discussions not of interest to them and read only contributions that are. Also, in a computer conference, everyone can speak at once. (p. 87)
There are of course disadvantages also, most notably the inability to perceive the physical reactions of fellow participants. The tone of voice, grimaces, twitches, arched eyebrows, and other body language of conference participants may convey more meaningful information than their words. (p. 88)

In a similar vein, Khoshafian and Buckiewicz (p. 354) assessed the advantages and disadvantages as follows:

One of the consequences of asynchronous interactions is that relationships that are displaced in time tend to focus more on the content of the interaction and less on personal issues. For many types of interactions, this reduced intimacy is an asset; for others it is not. Indeed, the preponderance of daily business interactions revolve around "human" issues of social, political, and interpersonal concern… Psychologists and communications professionals estimate that approximately 70 percent of a communication is carried out by relatively unconscious channels such as facial expressions, body posture, and manner of speech and presentation.
Espousing a common theme among groupware analysts, Opper (1993) said, "Groupware is a way of working as much as it is an information systems process." More specifically, she cited the potential for recovering time by radically reducing cycle time - the elapsed time between the decision to take an action and the time it is completed.(6) She said such reductions result primarily from factors such as: reduced face-to-face meeting frequency, shorter meeting times, decreased travel time, less time spent waiting for responses, faster information transmission and retrieval, greater individual control over interruptions, faster decision-making, and improved organizational learning. Opper also highlighted some of the benefits of using groupware to store and categorize information so it can easily be retrieved, thereby enabling groups to use it more creatively:
When information is available to everyone who needs it projects run more smoothly, teams are more focused, and repetitive problems can be eliminated. With groupware: conversations have an audit trail; new group members get up to speed quickly and easily; everyone can access a central repository of group knowledge known as a group memory; redundant communications are reduced, and accountability is more visible and therefore performance increases.
Accountability can be threatening to some people, something akin to Big Brother watching. Moreover, the entire concept of groupware - information sharing and workgroup collaboration - may meet resistance. Sharing of information does not come naturally to many businesses, where data has been zealously guarded for decades, nor to individuals who enjoy status as recognized experts upon whom others must rely for information. Radosevich (1995) points out that the discussion database is one of the most widely employed groupware applications, but that one of the biggest challenges is getting users to participate. To help overcome that problem, Information Week (1994) suggests that organizations should provide incentives for participating in this new paradigm. Computerworld (1994) goes even farther. While suggesting that groupware is culturally well suited to Japanese companies, the magazine asserts that groupware "risks becoming a passing fad in North America."

If groupware is to avoid that fate, it must establish a real and practical purpose in American organizations. What might such a purpose be?

Noting the explosive growth of the "knowledgebase," Dallas (1995) suggests a possible, if not vital purpose -- filtering out all but that information which is most relevant at any given time. He points out there are at least three definitions of what is most relevant or important: If the purpose is to stay current, the most recently submitted information might be of greatest interest. If a particular event is of interest, it might be best to concentrate on changes to a small portion of the knowledgebase. If a particular question is asked, only information required for the answer is relevant. Dallas says, "Gleaning information from a … voluminous source is like drinking from a firehose" and he devotes an entire chapter to "finding and reading documents."(7)

In his reference to documents, Dallas propounds a more focused purpose for groupware than the general filtering of disaggregated information. For better or worse, documents represent someone's concept of how a particular subset of all of the information in the world can best be organized and presented in a meaningful fashion.(8) Thus, in and of themselves, documents represent a highly processed, value-added, pre-filtered product.(9) Moreover, citing to a study conducted by International Techvantage for Lotus, Emigh (1995) points out, "An overwhelming 91 percent of [U.S. business] managers said they require input or feedback from others when creating documents." The study also determined that managers spend 47 percent of their time "gathering input from others [and] reviewing and consolidating information." While it might reasonably be argued that groupware is a technology in search of a mission, these figures provide a clear indication of what it should be. In discussing the output of collaboration Schrage cuts to the core of the issue:

By far the most important tangible product of computer-augmented meetings is the printout. The ability to actually manufacture a document is key … In fact, if a session isn't going to generate a document …, then perhaps the meeting should not be computer augmented. Indeed, if [a meeting] can't generate a document worth distributing, perhaps it's not a meeting worth holding. (p. 206)
Making the same point in somewhat more forgiving terms, the Yankee Group (1995, in Communication, Collaboration, Coordination: The "Three Cs" of Workgroup Computing) says(10):
The most significant deficiency of collaboration, especially in traditional meetings, is the failure to capture the knowledge that is generated during the collaborative process for future reuse… As a result, the potentially valuable knowledge, insights, and experience deriving from most collaborations only serve the organization once, and are then lost forever (perhaps to be recreated time and again as new individuals face a problem that the organization has already solved). Thus, the expertise or wisdom generated by an organization is allowed to dissipate.
Collaboration without output is process without substance. That may be acceptable and even desirable in a social context, but it has little justification in a commercial or governmental organization in which time is money. To ensure that collaboration is productive, the second of Covey's (1989) seven habits for highly effective people -- to "begin with the end in mind" -- should be practiced. For knowledge workers, the output of collaboration is documentation; meaningful professional collaboration begins and ends with documentation.

Expanding on that theme, Schrage elucidates, "Collaboration is a purposive relationship. At the very heart of collaboration is the desire or need to solve a problem, create, or discover something within a set of constraints. These constraints include expertise, time, money, competition, conventional wisdom …" (pp. 36-37) Speaking of constraints, conventional wisdom might suggest that civility and accommodation are critical to successful collaboration. However, Schrage argues to the contrary:

Candor -- if not rudeness -- is at the heart of most successful collaborative relationships. To be successful, a collaboration can't afford the risk of substituting euphemisms for clarity. Collaborative relationships aren't built on rudeness; it's just that they won't let good manners get in the way of a good argument. (p. 42)
The thing that matters most is that the collaborators possess a modicum of mutual trust, the belief that they are each adding value, and a genuine desire to solve the problem at hand or create something new. (p. 43)
Schrage says, "Unfortunately, the harsh reality remains that tossing a few people into a room and nailing them to a deadline does not an ad-hocratic collaboration make." He explains that ad-hocracies involve purely human structures that are reconfigured in interesting ways. While flexible, versatile social structures are important, Schrage suggests that technical infrastructures are too. In fact, he says technical infrastructures are indispensable and that the best way to build those infrastructures is to make sure that people can easily grasp and use tools for collaboration. (pp. 59-60)

Farson (1996) concurs, "… technology invites people to 'speak' more openly and candidly than they would in ordinary meetings." While we tend to think the best communication takes place where people can see and touch each other, he says that is not necessarily so. In fact, he argues, "…face-to-face communication often introduces more 'noise' in the system and imposes more limitations on personal expression." (pp. 48-49)

Although having the proper tools is essential to doing a job well, Schrage points out that people are constantly asked to accomplish their tasks with tools that are ill conceived, ill designed, and inadequate. (p. 65) He says, "… the tools we use define who we are … a good tool becomes an extension of ourselves…" (p. 69) Since humans are social beings, it is not surprising that meetings are a tool that we use to pursue our ends.(11) However, Schrage mounts a devastating attack on the use of meetings in common practice:

People know that they have to interact; survey after survey confirms that American managers typically spend fully half their working days in meetings. Most people crave the opportunity to be and work with others. And yet there are … awful feelings of impotence and frustration [with meetings]. (p. 119)
It is easy to say that meetings should be better run, that there should be an agenda, that the meeting leader should encourage participation and deflect efforts by the more loquacious to dominate. There's a lot of blather about the importance of "being sensitive" and "being a good listener" and "being prepared." It's all so easy to say, but it's also meaningless. What's more, it misses the point. The structure of meetings in everyday life practically dictates that they will be wasteful, repetitious, and frustrating. They're antithetical to meaningful collaboration. (p. 120)
What's going on is a very primitive, almost tribal, power game based upon a perfectly reasonable misunderstanding of what communication means. [Meetings are] a linear montage of speeches, soliloquies, conversations, arguments, interjections, visual displays, gestures, and grunts all presented within a certain time frame within the same room. (p. 121)
Meeting mechanics and dynamics focus on the individual. The ecology of meetings breaks people and their environment into incompatible niches. The group is nothing more than a collection of individuals who happen to be sharing physical proximity and a common problem at a point in time. (p. 122)
Again, Farson agrees, "Many supposed communication problems are actually balance-of-power problems… It is only when the balance of power is relatively equal that a truly candid communication can and should take place." (p. 55)

Schrage argues forcefully that the most successful meetings are those where people collaborate to create a document; i.e., the meeting is used actually to do work rather than just talk about doing work. (p. 111) That assertion may come as a rude awakening to those who busy themselves rushing from one meeting to the next wondering how they can be expected to accomplish so much "work". While one definition of "work" is "force times distance," surely, the movement of bodies from one meeting location to the next cannot be the sole intended output of meetings.(12) Carrying the discussion forward to the importance of visual "objects" in establishing "a new ecology of meetings," Schrage says:

Most significantly, perhaps, is that what's displayed can be physically printed out, copied, and distributed to all group members and any other relevant parties. I can't overemphasize how significant this is. It transforms the traditional meeting ecology - to discuss a topic, come to a conclusion, or make a decision - into an act of shared creation. The ability to create a tangible product completely revolutionizes what a meeting is. The meeting becomes a group process to create a group product. (p. 126)
Not all meetings are collaborative or need to be. But when there are problems to be solved, innovations to be created, decisions to be made, the ecology of shared space offers tremendous potential to those willing to exploit it. Using a computer screen as a shared space can be a powerful collaborative tool and environment. (pp. 133-134)
With regard to electronic information distribution, transmission, and processing systems, Schrage argues, "In fact, the design of these systems usually reinforces the ability to communicate at the expense of collaboration." To expand upon and reinforce the point, he quotes Carnegie-Mellon University professor Sara Kiesler, who has extensively studied the way electronic mail is used in large organizations: "When social definitions are weak or nonexistent, communication becomes unregulated. People are less bound by convention, less influenced by status, and unconcerned with making a good appearance. Their behavior becomes more extreme, impulsive, self-centered. They become, in a sense, freer people." In other words, Schrage suggests, "the system often isn't designed to encourage any other behavior but self-indulgent communication." (p. 145)(13)

Farson says, "Most organizations are, in fact, overcommunicating: meetings, conferences, memos, phone calls, and electronic mail overwhelm managers and employees alike." To reinforce the point, he cites a study by organizational psychologist Alex Bavelas that came to be known as the "line and circle experiment." In one group, all information was fed to a central person, i.e., top-down, line management. In the other, information was shared around a circle, like participatory management. In a simple task, the line group performed much better than the circle group, but when complexity was introduced, the circle group adjusted to change more quickly and performed better since they were allowed to communicate with the person next to them as well as the "leader". However, when all lines of communication were open and group members could talk not only to the people next to them but to all other members as well, the problem-solving ability of the group diminished markedly and it became virtually paralyzed.(14) Farson concludes, "…there always seems to be an optimal level of communication beyond which further or expanded communication becomes dysfunctional. Communication has its limits." (pp.53-54)

Opper offers similar arguments and adds that as E-mail and facsimile transmissions have proliferated, they have been used increasingly for self-protective communications. As Amram (1994) understates, "one limitation of e-mail is that messaging may get out of control and actually lead to wasted time." The same is true of a discussion database. If it is viewed as an end unto itself, the discourse in a collaborative database may only serve to aggravate the negative aspects of E-mail. Schrage reemphasizes the way to avoid such misguided excesses:

The Capture Lab experience at General Motors indicates that group documents can be put together in days, rather than months. Indeed, Coopers & Lybrand's David Braunshvig says that group writing offers "the biggest bang for the buck" in the value of collaborative tools. This isn't propaganda; it's empirical knowledge that the makers of collaborative tools have gained the hard way. The implication is clear: both executives and middle managers will spend a greater portion of their time collaborating on documents deemed to be important to their organizations. (p. 192, emphasis added)
As Farson puts it, "… rapid and accurate communication of information may be less important to management than are other organizational concerns." (p. 56) He discounts the usual explanation that managers are computer naïve and suggests instead that the information systems are simply not giving executives what they want and need. In other words, the systems are not adding any real value or contributing to any worthwhile work product. Addressing the merits and methodology of a document-centric approach, Schrage says:
The focus of collaborative technology shifts from assigning work to doing work. Shared understandings become possible more quickly. Collaborative environments will give people in group meetings the incentive to produce something - anything - that can be displayed. The rise of collaborative environments should speed the cycle of document creation and the spread of new ideas. If the collaborative tools are handled well, the quality of ideas and information will improve. (p. 193)
It's best not to begin a group writing session with a blank screen. Experience dictates that an outline or actual draft of the document offer[s] the best jumping-off point. Indeed, the practice seems most successful when it resembles group editing more than group writing. (p. 208)(15)
It is no stretch to suggest that "collaboration" is the collective act of documenting knowledge and that discourse, by whatever means, that does not result in the production of a document is not collaboration. In essence, that is what Schrage says. However, he does allow that "… altering the meeting ecology also has its downside. Fuzzy thinkers and people who have difficulty articulating thoughts are penalized." (p. 131)

Success Factors

Moving beyond the thesis that documentation should be the central focus, Schrage suggests that 13 ingredients largely determine how successful a collaboration may be: 1) competence; 2) a shared, understood goal; 3) mutual respect, tolerance, and trust; 4) creation and manipulation of shared spaces; 5) multiple forms of representation; 6) playing with the representations; 7) continuous but not continual communication; 8) formal and informal environments; 9) clear lines of responsibility but no restrictive boundaries; 10) decisions do not have to be made by consensus; 11) physical presence is not necessary; 12) selective use of outsiders for complementary insights and information; and 13) collaborations end.

Stedman (1995) posited nine factors for successful implementation of groupware: 1) Commitment from the top down. The system "must be seen by all as a strategic environment." 2) A clear understanding of how the business functions and mapping of the information technology to those processes. 3) Central management and control of the development process. 4) Implementation of applications one at a time. 5) Implementation of the most "commercially visible" applications first. 6) User cooperation, training, and involvement. 7) Use of systems that are appropriate for the information being processed. 8) Assistance from experienced consultants. 9) Following a progressive series of stages, the first of which is to have an information services strategy in place. Second is definition of the business process. Third is breaking down the workgroups into comprehensible components. Fourth is definition of the application. The fifth and final stage is the application design cycle.

McCready and Palermo (1994) list eight success factors for Lotus Notes (pp. 15-18): 1) A charismatic leader - Lip service from senior management is not enough. They must make an "emotional commitment" as well as a financial commitment. 2) Information systems support - There is a fine line to walk between requiring standards and suggesting policies. A happy medium is to allow users to develop individual applications but to require centralized approval for general distribution. 3) Impact daily life of users - The first application must be so compelling that users will see Notes as the preferred application environment. 4) Define benefits - The more precisely the benefits are defined, the greater the benefits will be. Numerical goals should be established, along with methods for achieving each objective. Benefits should be classified in terms of cost reduction, productivity improvement, cost avoidance, and revenue enhancement. Then a more detailed plan should be developed on how to achieve those benefits. 5) Cross functional applications - The benefits should be extended across functions in the enterprise. 6) Imagination - This factor is cannot be quantified. Like pornography, you know it when you see it, but like sex in life, business cannot continue to exist without it. 7) Passion - It is especially important for the first application. The proponents must be able to take it to the boss with conviction. 8) Development methodology - Users should be involved early and at all stages of development, with mechanisms in place to ensure good communications with the developers.

Unlike most other analysts, McCready and Palermo argue that an investment is Notes is relatively risk-free since no specialized hardware is required nor long development cycles or specially skilled personnel.(16) However, they do add that "success can kill." Unexpected demands may arise for enhancements and new applications, and it is critical to stick to a reasonable phase-in schedule. They recommend the allocation of $150 per user per year to provide for continual training, and they caution that training and support resources can be overwhelmed. Finally, they note the importance of guarding against the "Fantasia Syndrome" - the belief that placing Notes on everyone's desktop will cause "magic in productivity." To drive home the point, they emphasize that "standards for the sake of standards is not the business imperative."

Opper recommends that the process of implementing groupware should begin with input from business, systems and organizational effectiveness specialists and that it should be supported by an on-going team of business and technical experts. She suggests that a series of structured pilots builds a solid foundation for enterprise-wide groupware. Effective training and skilled support are necessary. Careful measurement of organizational performance before and after implementation will lay the groundwork for benefits assessment to support ongoing internal marketing until critical mass is reached. As with all benefits, there are also costs - to the organization, individuals, and groups. Thus, seeing the project through to the payoff begins with a commitment to taking on major change. The effective use of groupware requires clarity of purpose, alignment with a common goal and commitment to results. Education and ongoing support are essential.

In her groupware "readiness checklist," Opper includes 12 factors in three categories - environmental circumstances, business conditions, and organizational goals. The factors are: 1) Share of the organization that is networked. 2) Whether the group regularly conducts business at multiple locations. 3) Whether an upcoming change will cause a massive response by the organization. 4) Whether there are breakdowns in communication that cost money, effort, or lost opportunity. 5) Whether the organization's delivery cycles are longer than the industry average. 6) Whether the work depends upon regular, close contacts with outside organizations. 7) Whether the work is highly project oriented. 8) Whether the work is highly document driven. 9) One or several senior managers are aware of the benefits of groupware and inclined to test its use. 10) Staying close to the customer is a high priority. 11) Quality improvement is a high priority. 12) The organization has a strategic commitment to using advanced technology.

Opper's major points concerning the three categories into which the readiness factors fall can be summarized as follows:

Environmental Circumstances -- Companies with few networks installed can still implement groupware, but careful planning and slow-paced implementations are required. Organizations that aren't heavily networked should score strongly on other items on the checklist before they inaugurate groupware. Groupware supports dispersed work groups, so regularly conducting business from different locations benefits from a networked solution. When geographic dispersion is relatively new, people are still looking for solutions to the inconveniences of a multi-location environment. An upcoming significant change generally has a heavy impact. When an organization faces a major but predictable change, such as new regulations, a merger, or a dramatic shift in policy, heavy group activity is often required. When there is advanced notice of these changes, groupware can be beneficial.
Business Conditions -- Slow delivery cycles and the need for close coordination with suppliers or outside professionals both present groupware opportunities. Similarly, companies that are project-oriented or document driven will benefit.
Organizational Goals -- Groupware champions have a powerful influence on the success of groupware. A company or department whose leader believes in the value of groupware can sometimes justify piloting groupware on that factor alone. Emphasis on customer service involves both delivering what customers want - an issue of making the right product and service decisions - and doing it rapidly, perhaps even by hooking customers into the system. The leader's position regarding the use of technology is key to the way the organization receives groupware. Early pilots should have leaders who are willing users of technology. More than any other single factor, the actual, regular use of groupware by the head person will determine the success or failure of the groupware experiment. Initial applications for piloting groupware should be relatively simple. They should address specific business problems in areas where groupware can obviously make a major contribution. And they should be situations where the payoff is evident and relatively easy to measure.
Opper says that for groupware to produce the greatest pay back, it should be implemented enterprise-wide, but that it can be implemented in stages or even in a single department. She suggests that four questions should be analyzed in order to determine which level of commitment is appropriate for the organization: First, how supportive is the corporate and information systems environment? Second, how well is the organization operating? Third, how responsive is it to its customers? Fourth, how does it stack up against the competition? Regardless of the scope of implementation, however, she cautions that groupware should be approached with "considerable care and oversight." In summary, Opper suggests, "…wise product selection and thoughtful implementation are keys to success."

Dallas (1995) says that three groups must agree on the need for groupware - upper management, information systems managers, and the team that will actually be using it. However, the attitude of the team itself is by far the most important. If the team buys into the system, Dallas asserts that it will succeed. To ensure that it does, he suggests the groupware proposal should include three components: 1) a sound business plan showing how the investment will pay for itself, or at least yield benefits, over time; 2) analysis of the impact on the existing infrastructure in an implementation plan approved by the information systems managers; and 3) a few early adopters on the team who will ensure measurable success during the critical first pilot program.

Although they are not specifically focused on groupware or even information technology in general, several of Clemmer's (1995) "deadly dozen failure factors" are very relevant to this discussion: leadership lip service that does not reflect a real commitment on their part; fuzzy focus on improvement efforts that are "not connected to the burning issues that keep senior managers awake at night"; partial and piecemeal efforts that are too narrow and segmented; and misaligned systems and processes. Of the latter failure factor, Clemmer cites evidence clearly showing that only about 15 percent or less of the time is an error, complaint, or problem rooted in people problems. He says that over 85 percent of the time the root cause is found in organizational systems, processes, or structure.

Farson suggests that executives are more like to rely on the advice of their colleagues than any comprehensive display of data. He says they need interpretations, opinions, and information that has been "massaged" and that's why they rely on meetings, memos, and telephone conversations. While a collaborative database might be used as a substitute, in part or in whole, for those forms of communication, Farson forcefully concludes:

Change and Change Management

In his overview of groupware, Coleman (1995) suggested that change management is critical with groupware and that planning for change drastically improves the chance of success. He said organizations tend to resist change in proportion to their size. The larger the organization the greater the resistance, and the bigger the change, the greater the resistance. He cited seven major forces that provide the initial propulsion toward groupware: 1) a network infrastructure capable of supporting groupware is now available; 2) improved price/performance of both groupware hardware and software has made it more available to a larger population; 3) recession and downsizing force increased white-collar productivity; 4) well known vendors such as Microsoft, WordPerfect, Lotus, IBM and Digital Equipment Corporation (DEC) are promoting groupware products thereby increasing awareness in the marketplace; 5) increased competition imposes change on organizations, making them flatter and more flexible, often requiring groupware for this transformation; 6) increased complexity in today's products and business procedures is driving the use of ad hoc teams supported by groupware; and 7) articles in the trade and business press have increased awareness of groupware and aroused the curiosity of business leaders.

Coleman also said there are equal and opposing forces that inhibit the growth of groupware including: 1) a low level of education in the business community about groupware; 2) confusion in the marketplace as to the nature of groupware; 3) decreasing budgets and many firms perceive they cannot afford the investment in groupware; 4) the distribution channels for groupware are new and not fully developed; 5) MIS shops worry that they will become dependent on a groupware vendor; 6) organizations are resistant to change; and 7) there are few standards in the groupware market to foster rapid growth.(17)

Coleman pointed out that when 500 groupware users were surveyed at the GroupWare'93 conference about their success and/or failure with groupware, those who were not successful noted the greatest problems were not technological. Instead, problems stemmed from the lack of support from top management or lack of a well defined business problem.(18) (Users were surveyed again at the 1994 groupware conference with essentially the same results.) To overcome the problems and ensure success, Coleman suggests 20 common sense rules for deploying groupware:

1. Find a groupware champion! The higher up in hierarchy, the better. Get managements' hands on the keyboard. By getting top management involved, they see the benefits, and you will get a lot more support!
2. Groupware changes the corporate culture. Plan for it!
3. Pick a pilot project rather than trying to roll groupware out to the whole organization.
4. Pick a bounded project with a group that is supportive of both technology and innovation.
5. Pick a project with visibility and financial impact.
6. Realize that training, maintenance, and support will be the majority of the cost, rather than the initial cost of the software.
7. Measure productivity factors before and after the project has started. This is a good way to cost-justify groupware!
8. Pick groupware software based on a specific business problem that needs to be solved and has not been solved successfully using traditional methods. Corollary: Don't pick the groupware first and then try to find a problem.
9. Make sure you have adequate planning, support, training, and maintenance for your project.
10. No single groupware product can do it all. Don't expect it to!
11. Don't expect software vendors to offer you all the services you need for groupware. You may need to use internal people or consultants to ensure your project's success.
12. Groupware is not a quick fix! As part of a re-engineering effort, it may take 2-4 years to see the results.
13. Listen to the people involved in the pilot project. They are experts on what needs to be done and can often suggest ways to better the process.
14. Don't be afraid to make changes! A pilot project is an experiment. Learn as you go.
15. Make sure the software you pick fits with existing systems. Try to amortize your LAN investment by connecting to your mainframe or other legacy systems.
16. You can't change people overnight. Be prepared for resistance!
17. People take time to change. Organizations take even longer!
18. It takes courage to change a corporate culture! Applaud those who are willing to change.
19. Be careful about paving the cow path. There is no point in automating a very inefficient process. There are no big productivity wins here!
20. Groupware can be very political. Make sure it is a big win!
Burns (1995) suggests groupware is important to the information technology strategies of organizations for two primary reasons -- downsizing and competition. Downsizing meaning that the "survivors" must do more with less, and groupware helps them work more efficiently with fewer resources. Meanwhile, competition is driving organizations to improve their customer service, to reduce their costs, and to more rapidly respond to changing market conditions. Groupware helps organizations meet the competition through greater coordination of activities, by minimizing obstacles of time and geography, and by expediting the decision-making process.

Burns points out that groupware use is now growing rapidly due to a number of factors. The infrastructure necessary for groupware to be widely implemented has reached a critical mass in many organizations. The typical desktop computer now has sufficient power to collaborative computing features. Perhaps most importantly, organizations are turning their attention from personal computing toward workgroup, or team-based, computing to improve enterprise productivity.

Finally, Burns says several key trends are driving the use of groupware: First, products like Lotus Notes are steadily and methodically becoming fixtures in the daily lives of knowledge workers, as companies move from pilot phase to deployment. Second, companies struggle with decisions about which products and technologies will solve the business problems that they have clearly identified. Third, these companies are gambling on the outcome, deploying groupware products without clearly formulated, objective criteria for measuring the benefits. These factors lead Burns to the conclusion that there is pent-up demand for groupware that is just hitting the "knee in the curve" as groupware moves from the innovator stage to the early adopter stage.

As evidenced by IBM's multi-billion dollar takeover of Lotus, Burns is not alone in her belief. No doubt, the first trend she cites is well taken. If only by trial and error, but certainly in large measure because of the vast technological advances of recent years, people in organizations are beginning to realize and take advantages of the improving tools that are now available to help them do their work. However, the second and third "trends" that Burns cites could just as easily lead to the conclusion that groupware will fail. If companies truly understand what their business problems are, figuring out how to solve them should not be such a struggle. Likewise, if they are "gambling" on products without understanding the benefits, the risk of absolute or relative failure is high as, over time, it becomes apparent that they have chosen tools that are less than optimal for the tasks at hand.

Notes v. the Internet

It might be said that the Internet is the ultimate groupware tool -- potentially linking together the whole wide world for collaboration and information sharing. While any long-term predictions are risky in the rapidly changing arena of information technology, if anything is certain it is that the information resources, applications, and use of "everyone's network" will continue to expand. Thus, any discussion of groupware would be incomplete without reference to the Net as compared to the currently recognized leader in groupware, Lotus Notes.

As Schrage (1995) pointed out, IBM's takeover of Lotus served notice to Big Blue's customers and competitors alike of its intent to make the world's largest computer company a force with which to be reckoned in the fast-growing groupware marketplace. Schrage said, "there's no question that this … is really about who gets to define and control the future of corporate computing. IBM desperately -- and desperately is the operative word here -- needs to be seen once again as a company that can set technical standards in the information marketplace."

Schrage points out that Notes is cleverly designed and has a wide following in organizations ranging from General Motors to McKinsey & Co. to Price Waterhouse to the Central Intelligence Agency. Notwithstanding competing groupware packages, such as Novell's GroupWise and Collabra Share, he asserts there should be no argument that Notes has had an extraordinary impact in causing organizations to rethink the role of information technology in management. While some organizations have merely used Notes for E-mail, many others have transformed the way they share information. Concerning the transformation, Schrage says:

… some pundits declare that what's important … is that corporate computing is shifting away from being a collection of computational devices to becoming more of a medium of communication. That's accurate without being quite true. What's really going on is the gradual realization that computers and their networks are more valuable to corporations when they are used to create and manage interpersonal relationships, not just to create and manage information. Groupware products such as Lotus Notes shift the technical design sensibility away from using software to structure data to using it to structure relationships -- not unlike the way we use our voice mail systems to screen our calls as well as exchange messages. The goal becomes the creation of value through human interaction, not just better information. Groupware thus becomes a kind of operating system that helps manage and coordinate critical relationships and the flow of information in a corporation. That's why an IBM feels compelled to seize on a product that's perceived as a winner in this emerging marketplace.
However, Notes and other proprietary groupware systems are facing tumultuous and volatile changes in network communications, particularly the phenomenal growth of the decidedly un-proprietary, wide-open World Wide Web.(19) Now it is possible for every large and not so large company and organization to have its own home page on Internet, and many individuals are even establishing their own personal pages. Work groups can easily set up their own sites on the Web. It doesn't yet have all the functionality nor is it as secure as IBM's Lotus Notes but multitudes of entrepreneurs and established software companies, including Lotus and IBM, are hard at work on ways to make the Web more useful to organizations.

Schrage points out that innovative companies can get most of the value of Notes at far lower cost by using the Web as their networking medium. In terms of cost-effectiveness, he says technological infrastructures such as the Web are rivals as much as partners with Lotus Notes, and a company like Netscape is as much a rival to Lotus as is Microsoft.(20) Indeed, Netscape has established itself as a direct competitor by acquiring the primary discussion database alternative to Notes -- Collabra Share. In at least one recent head-to-head comparison (PC Magazine), Collabra Share was judged the superior product. Moreover, it will be widely available since Netscape intends to blend Collabra Share's discussion features into its Navigator Web browser.

Notwithstanding competitors who are boldly bidding "adios to Notes" on the Net even as they tout their alliances with IBM (Open Text), Schrage suggests that IBM may have some success in establishing Notes as a groupware standard. However, he says the most important point is that, "… as we see our technologies more as a medium to design how we should work together rather than merely design how we share our information, we are going to see all kinds of new groupware models and approaches enter the marketplace." Although that may not bode well for Notes, Schrage says it's terrific for companies that can use technology to create value internally and with their customers. Whether in fact the benefit to organizations can be optimized, however, depends upon at least two factors -- whether they choose the proper and best tools for the tasks at hand and whether their people are ready, willing, and able to use them effectively.

On that score, Mezger (1996) reported the results of a survey conducted for Lotus. Of 1,000 randomly selected managers in a variety of large and mid-sized companies, 90 percent said technology and teamwork will be critical to success in the global economy. However, 70 percent said their teams are not fully utilizing technology, and only 47 percent indicated the workplace encourages teamwork. Some questioned the notion that technology works best when it fosters teamwork and others suggested that it actually slows decision-making, as compared to face-to-face communications. Mezger quotes one researcher as saying, "… the technology side has often been oversold, has frequently disappointed, has generated its own kinds of confusion, and managers have a skepticism about it."

Human Factors

Drawing theoretical inspiration from educational philosophy and learning theory which stresses that learning is a social and cultural process and that knowledge is constructed in interaction with others, Gay (1994) examined how students at physically separate locations solved a design problem by using various communication channels, including audio, video, computer text, online resources, and face-to-face discussion. Her analysis showed that video-audio was the preferred medium of intergroup communication, corroborating the findings of others (e.g., Chalfonte, Fish, & Kraut, 1991; Daft, Lengel, & Trevino, 1987) suggesting that richer and more interactive media provide better support for "nuanced" communication and thus collaboration.

Comparing the use of text versus voice in collaborative writing, Chalfonte et al. found that the expressiveness of oral language facilitates the communication of more complex ideas more quickly and places fewer cognitive demands on the interlocutors. While speech is not inherently more expressive and quicker than text, speakers carry a greater awareness of a partner than writers do and they modify their speech in response to various cues. As a result, more attention is given to the social dimension of communication than to the content of the message, and Gay suggests that may promote smoother collaboration.

Thach (1994) highlighted other disadvantages of conducting interpersonal communications through machines. In particular, employees with low computer literacy have difficulty becoming accustomed to the systems, and extroverts are often frustrated when typed text become the medium for communication. Thach raises some key questions:

The issue of automate or informate has also not been resolved. Will the technologies be used just to increase speed of business practices and reduce staff, or will people be retrained to use the data and improve processes? Some experts warn that how well the information technology is used is a function of organizational learning. Have organizations learned that IT is not only for automation, but for informing "smart" employees who synthesize the data for improved organizational efficiency?
Another major issue that concerns some experts is the psychological side effects of interacting through a machine all day. In the organization of the future, it may not be necessary to meet face-to-face every day, to shake hands, to touch people. Instead all of this can be done via videophone, videoconferencing, telepresence, networked portable computer, or virtual reality. What are the psychological consequences of this? It does seem that organizations are poised to relearn some of the harsh lessons from socio-technical systems theory. The imposition of new technology, which ignores the impacts of such changes on the social fabric of the organization, may come at great cost and prove to be less effective than anticipated.
Thach cites earlier work with employees who experienced the impact of having their manual tasks replaced by a computer, in which Shoshana Zuboff found that workers complained they were no longer in touch with their work. They felt as though they were "floating in space," with their work "lost behind the screen." Other people who have been working with computers longer reported feeling more productive, but less personally fulfilled. Thach asks, "Will organizations have to plan and hold social functions after and during work hours to enable their employees to see and touch 'real people'?"

Thach also questioned implementation of the technology itself, suggesting that the continual need to replace equipment, combined with the lack of predictable standards, could prove costly.(21) In a similar vein, Radosevich (1995) noted the loss of network connections and heavy support requirements as commonly occurring problems with discussion databases. In fact, she said discussion databases require a "huge" amount of support.

Beyond Computing magazine (1995) reports that 75 percent of information technology managers doubt the ability of their organization to manage critical data. Not only is the task of managing client/server databases complicated and risky, but they generate vast amounts of information. Keeping them synchronized can be a "data administrator's nightmare" and other concerns include compatibility, security, backup, and recovery. Not many good tools are available for managing distributed databases. Some information managers believe "inherent problems in the structure … make distributed databases more trouble than they're worth" and that most organizations should instead provide distributed access to centralized databases. Beyond Computing suggests that groupware may bring the issue to a head. Replication is one way to balance the advantages and disadvantages, but replication is not easy to manage. The lag from the master to the slave may be a problem, and the replicator competes with users for the same data and network resources.

On the positive side, Thach noted that discussion databases can replace some other forms of communication, such as fax or E-mail or even live meetings, and that they provide a means to document and coordinate information and discussions. Amram (1994) points out that the groupware and electronic mail markets are both growing fast, while offering different approaches:

While a groupware approach offers the benefit of an integrated database and a standard that runs across multiple platforms, others may promote e-mail as being most important for a computing foundation. One limitation of e-mail is that messaging may get out of control and actually lead to wasted time. Groupware, on the other hand, facilitates discussion databases, so each employee can check into the appropriate database without having to open unnecessary messages. Groupware also creates an archival record that will not disappear, and it can also accept several media formats, such as graphics and sound. Some hybrid products extend standard e-mail with groupware features to offer the best of both worlds.
Indeed, it is likely that the distinctions between E-mail and discussion databases will narrow, if not dissolve as the features of each are integrated into the other. Most major E-mail systems run on multiple platforms. Many have "bulletin board" features that are similar in function to discussion database threads, and E-mail messages are routinely preserved and may or may not be considered official records from a legal standpoint, depending upon the content. Still the fundamental issues remain with respect to whether E-mail or even more flexible discussion databases are the proper and best tool to provide the information management services required by an organization -- and whether the people comprising the organization are ready, willing, and able to use them. Even if a decision is made to reduce the time spent in meetings, advances in telecommunication mean that desktop video and televideo conferencing will increasingly compete with and perhaps render sterile discussion databases still-born in the embryonic stage.

Conclusion

The need and opportunity to realize the advantages and the ability to accept or overcome the disadvantages are prerequisites for successful implementation of a collaborative database. The need and the advantages may be accentuated in a large, geographically dispersed organization that works with information that is widely distributed, maintained in the minds of its employees, and routinely shared in meetings, conversations, hand-written notes, memoranda, and other traditional forms of communication. However, if the people comprising the organization are unwilling or unable to use collaborative software in lieu of or at least to supplement more traditional means of communication, it cannot succeed. On the other hand, if the organization is faced with external pressures or constraints, it may have little or no choice but to begin to take advantage of on-line collaboration. Such pressures and constraints may include reduced travel funding, personnel "streamlining," increasing workload, and/or constricted time frames for decision-making and action.

For commercial enterprises, external pressures may also include life-threatening competition in the marketplace if they fail to provide their employees the very best tools to do their jobs. Nonprofit organizations are also subject to competition for resources, albeit perhaps in a different sort of marketplace. Governmental organizations may be insulated somewhat from competitive concerns and thus lack the same urgency and imperative for efficiency and effectiveness. However, as technology evolves and improves, even governmental agencies that fail to capitalize on its potential risk criticism and sanction -- from auditors, inspectors general, the news media, interest groups, Congress, or even in some rare instances from enlightened superiors. Technology enablers raise the standard, and what was formerly "good enough for government work" may no longer be so.

To whatever extent they are implemented and are to succeed, groupware applications must focus on real and meaningful outputs. As far as collaborative, "discussion" databases are concerned, the issues are whether they are: 1) needed and justified to replace or implement informal communications by E-mail, 2) accepted and will be used to save time and money that would otherwise be invested in personal gatherings, and 3) the best means to support the creation, review, and approval of work products, particularly outputs that warrant documentation.

At a functional level, in Norman's terms, the issue is whether a discussion database is the best or most important vehicle for an organization's knowledge to be maintained "in the world" so that it does not need to be maintained in the crania of the people comprising the organization.(22) Is the information best and most appropriately maintained as discrete data elements, i.e., in fields in a database? Or should it be encapsulated in more complete chunks that organize a set of elements into a logical, comprehensible whole, i.e., documents? Also, to the extent that on-line collaboration is justified, is it most efficiently and effectively carried out in a free-flowing, multi-threaded database or, as Schrage argues, should it be focused directly on a document? If Schrage is right, are collaborative notes best provided and maintained within the document itself? Or is there a need to maintain them as distinct objects that can be manipulated and analyzed within a relational database? Is it the relationships among the comments themselves that is important? Or are they only meaningful insofar as they can be brought directly to bear on the output?(23)

Covey's (1989) third habit -- putting first things first -- is of particular relevance. If collaboration begins and ends with documentation, the question must be asked as to whether a discussion "database" is the best place to begin to collaborate. The shortest distance between two points is a straight line, and the most direct access to a document is within the wordprocessor in which it was created.(24) As Fisher and Ury (1981) point out, "You do not need to get anyone's consent to start using the one-text procedure." (p. 122) However, before creating a document, you are likely to consider whether it is worth the time and effort, and that in itself is a value-additive filtering process. The false presumption underlying a discussion database is that all thoughts are worth recording. Not only is that obviously not true but the actual effect is perverse -- aggravation of the problem of information overload -- and counterproductive to the purpose of true collaboration.

That is not to say there is no place or role for a discussion database anywhere or anytime. It is merely to suggest that such a system may be a luxury that many organizations cannot afford unless and until their more basic and direct information management needs are met -- those needs that relate to the organization's most important information, that which warrants documentation. Analysts are virtually unanimous in recommending careful selection of the proper tool to fit the most critical tasks at hand, a cautionary note that those who contemplate installation of a discussion database would do well to heed.

Bibliography

Ambur, O. (1995, August 3). Functional, Technical, and Resource Requirements for the Servicewide Document Management System: Findings of the Requirements Analysis Team, Together with Recommendations and Alternatives. Division of Information Resources Management, U.S. Fish and Wildlife Service. This report grew out of the recommendation of the Surnaming Task Force for implementation of an agencywide electronic document management system. The Requirements Analysis Team was asked by the Director of the agency to conduct a more detailed analysis, identify alternatives, and submit recommendations. Ambur led the 20-person team and wrote the report. The team was comprised of those designated by the leadership of the agency as best qualified to address the document management needs of the organization. The report was widely circulated for review within the agency, and as recommended in the report, efforts are underway to conduct an expanded pilot prior to agencywide implementation of the project.

Ambur, O., Bauer, T., Bender, K., Kilcullen, K., and Maruca, M. (1994, November 30). Report of the Correspondence/Surnaming Task Force, U.S. Fish and Wildlife Service. The objectives assigned to the Task Force by the Director of the agency were to: 1) make the surnaming process more efficient and less lengthy, 2) improve the quality of documents, and 3) ensure that key people stay informed. Among the recommendations of the Task Force were: greater delegation of signature authority, minimization of the serial routing of documents in favor of simultaneous reviews, empowerment of 2-person teams with authority for each document, and implementation of an agencywide electronic document management system. In other words, in urging greater delegation of authority, the Task Force in effect recommended less "collaboration" as it is traditionally recognized in terms of an iterative process of comment and response. The 2-person team - the person with the lead and the person with approval authority - would be empowered to finalize the document whenever they choose. Collaboration would still occur but it would take place "simultaneously" rather than serially, within the context of a draft document, and the empowered individuals would be free to solicit and/or ignore the comments of others as they see fit. Of course, they would also be held accountable for the quality of the document.

Amram, Y. (1994, July 1). Is It Going to Be Groupware or E-Mail? Computerworld (p. 37). Available at: http://www.elibrary.com/cgi-bin/hhweb/hhfetch?35282257x0y566:Q001:D003.

Baecker, R.M. (1993). Readings in Groupware and Computer-Supported Cooperative Work: Assisting Human-Human Collaboration. San Francisco: Morgan Kaufmann Publishers.

Burns, N. (1995) Groupware: Myths and Realities. Creative Networks, Inc. Available at: http://www.collabra.com/articles/myth.htm. This on-line article excerpted from Ms. Burns' work summarizes the results of a research program which studied how North American organizations use groupware and why they are using it.

Chalfonte, B., Fish, R.., and Kraut, R. (1991). Expressive richness: A comparison of speech and text as media for revision. Proceedings of CHI 1991, 21-32. New York: ACM.

Clemmer, J. (1995). Pathways to Performance: A Guide to Transforming Yourself, Your Team, and Your Organization. Rocklin, California: Prima Publishing.

Coleman, D. (1995). An Overview of Groupware. Groupware: Technology and Applications. Prentice Hall. Chapter 1. Available at: http://www.collabra.com/articles/overview.htm#20rules.

Communication, Collaboration, Coordination: The "Three Cs" of Workgroup Computing. Boston: Yankee Group. Available at: http://www.collabra.com/articles/yankee.htm.

Covey, S. (1989). The Seven Habits of Highly Effective People. New York: Fireside. Covey's seven habits are: 1) be proactive - personal vision; 2) begin with the end in mind - personal leadership; 3) put first things first - personal management; 4) think win/win - interpersonal leadership; 5) seek first to understand, then to be understood - empathetic communication; 6) synergize - creative cooperation; and 7) sharpen the saw - balanced self-renewal.

Daft, R., Lengel, R., & Trevino, L. (1987). Message equivocality, media selection, and manager performance: Implications for information systems. MIS Quarterly, 11, 355-366.

Dallas, A. (1995). Making Your Team More Productive. Excerpted from Special Edition, Using Collabra Share. Que Corporation. Available at: http://www.collabra.com/articles/product.htm.

Davis, S. and Davidson, B. (1991). 2020 Vision. New York: Fireside.

de Sola Pool, I. with Noam, E.N. (Ed.). (1990). Technology without Boundaries: On Telecommunications in a Global Age (pp. 87-88). Cambridge, Massachusetts: Harvard University Press. de Sola Pool's book was published posthumously, six years after his death in 1984.

Di Pentima, R. (1996, February 26). Statement before the Subcommittee on Government Management, Information and Technology, Committee on Government Reform and Oversight, U.S. House of Representatives. Renato A. Di Pentima is Vice President and Chief Information Officer for SRA International, Inc. He joined a distinguished panel of experts testifying before the Subcommittee concerning the use of best practices of information technology in government. In addition to highlighting the importance of measuring "outcomes" instead of "outputs" (or inputs), he pointed out that the most successful enterprises in the private sector make corrections quickly "with a primary focus on future outcomes - not sunk costs." Too often organizations continue to try to maintain systems merely because they have so much invested in them, even when it may be better to dump them and implement an entirely new system.

Emigh, J. (1995, November 27). Lotus Intros SmartSuite 96 For Windows 95. Newsbytes News Network. Available at: http://www.elibrary.com/cgi-bin/hhweb/hhfetch?35282257x0y566:Q003:D006

Engberg, M. (1995, July 1). Collaborative Efforts Key to Successful Groupware, Workflow Implementations. Imaging World, 4, 7, 19.

Farson, R. (1996). Management of the Absurd: Paradoxes in Leadership. New York: Simon and Schuster.

Fisher, R. and Ury, W. (1981) Getting to Yes: Negotiating Agreement Without Giving In. Boston: Houghton Mifflin Company. p. 122.

Gay, M. (1994, April 1). Collaborative design in a networked multimedia environment: Emerging communication patterns. Journal of Research on Computing in Education, 26, 418. Available at: http://www.elibrary.com/cgi-bin/hhweb/hhfetch?35281384x0y563:Q001:D023.

Grohowski, R., McGoff, C., Vogel, D., Martz, B., and Nunamaker, J. (1990). Implementing electronic meeting systems at IBM: Lessons learned and success factors. MIS Quarterly, 14, 369-383.

Groupware could win big in Japanese corporate culture. (1994, December 1) Computerworld, 45. Available at: http://www.elibrary.com/cgi-bin/hhweb/hhfetch?32432879x0y940:Q001:D005.

Groupware's Culture Problem. (1994, May 1). Information Week, 52. Available at: http://www.elibrary.com/cgi-bin/hhweb/hhfetch?32432879x0y940:Q001:D009.

Hoenig, C. (1996, February 26). Information Technology: Best Practices Can Improve Performance and Produce Results. Testimony before the Subcommittee on Government Management, Information and Technology, Committee on Government Reform and Oversight, U.S. House of Representatives. Mr. Hoenig was one of a panel of experts who testified on best information management practices that might be implemented by Federal agencies.

Khoshafian, S. and Buckiewicz, M. (1995) Introduction to Groupware, Workflow, and Workgroup Computing. New York: John Wiley & Sons, Inc.

Kinghorn, C. (1996, February 26). Statement before the Subcommittee on Government Management, Information and Technology, Committee on Government Reform and Oversight, U.S. House of Representatives. Mr. Kinhorn was among a distinguished panel of experts testifying on information management best practices in government. In addition to setting forth his "four C's," he suggested that incentives should be provided for using commercial off-the-shelf (COTS) software. He asserted, "There is an inherent bias in government that government is so different that no off-the-shelf system could possibly work in that operating environment… Agencies, like successful private companies, need to focus on their core mission and let people whose business is technology run those investments."

Let's Talk. (1995, November 21). PC Magazine, 205-234. In addition to this article, which suggests that Collabra Share "continues to outpace its competitors in the groupware market," this edition of PC Magazine contains an article suggesting that E-mail systems may be swallowed up by wordprocessors. The article, entitled "WordProcessors: Front and Center," appears on pages 103-154.

Malone, T., Director, and Halperin, R., Executive Director. (1996, April 25). Research Overview. Center for Coordination Science. Massachusetts Institute of Technology. Available at: http://ccs.mit.edu/research.html.

McCready, S. and Palermo, A. (1994). Lotus Notes: Agent of Change, The Financial Impact of Lotus Notes on Business. Framingham, Massachusetts: International Data Corporation. Anne Palermo recently accepted a position as International Vice President for Marketing for PC DOCS, Inc., one of the market leaders in document management software, whose software is called DOCS Open. Subsequently, Scott McCready conducted a similar analysis of DOCS Open as the two of them previously conducted for Lotus. The analysis shows high rates of return on the software at various businesses and governmental agencies, with payback in the range of 1 to 1.5 years.

Mezger, R. (1996, February 25). Managers Wonder How to Bridge the Technology Gap: Survey Finds Them Unsure of How to Sift Through What's Available and Know Whether It Fits Their Future. The Washington Post, p. H15.

National Archives and Records Administration. (1995, August 28). Electronic Mail Systems, Final Rule. Federal Register, 60, 44636.

Norman, D. (1988). The Psychology of Everyday Things. New York: Basic Books, Inc. p. 54.

Open Text Corporation. (1996, May). Internet home page. Available at: http://www.opentext.com. Even as Open Texts touts its OEM status with IBM, it bids "adios to Notes" in offering its Livelink Internet alternative. According to the company's sales pitch, Livelink is "… the first and most comprehensive suite of intranet applications that allow people in businesses to: collaborate on projects from multiple, global locations; search, manage, and maintain vast stores of critical business documents anywhere in the world; access and share information across multiple computing environments, including local and wide-area networks, the Internet, and intranets."

Opper, S. (1993, November). Choosing and Implementing a Groupware System. Available at: http://www.collabra.com/articles/choosing.htm. This article is a short form of a book published by Van Nostrand Reinhold called TECHNOLOGY FOR TEAMS: Enhancing Productivity in Networked Organizations, ISBN 0-442-23928-9. It also appeared in Datapro's "Workgroup Computing Series: Strategies & LAN Services" (1995). Delran, New Jersey: The McGraw-Hill Companies, Inc.

Pappano, L. (1996, April 28) Publisher Plans to Book 'Library of Language' on the Internet. The Washington Post, p. A3.

Patel, J., Watson, J., and Fenner, J. (1996, May) Comparing Two Groupware Heavyweights: IBM's Lotus Notes 4.0 vs. Microsoft's Exchange 4.0. Imaging Magazine, 56-60.

Radosevich, L. (1995, July 1). Oh, What a Relief it is! Computerworld, 78. Available at: http://www.elibrary.com/cgi-bin/hhweb/hhfetch?35282257x0y566:Q001:D001.

Schrage, M. (1990). Shared Minds: The New Technologies of Collaboration. New York: Random House.

Schrage, M. (1995, June 8). What's Good for the Groupware . . . The Los Angeles Times, p. D-1. Available at: http://www.elibrary.com/cgi-bin/hhweb/hhfetch?

32432879x0y940:Q001:D001. Michael Schrage is a writer, consultant and research associate at the Massachusetts Institute of Technology. He writes this column independently for The Los Angeles Times and can be reached at schrage@latimes.com

Stedman, M. (1995, May). A Practical Guide to Lotus Notes Implementation. Information Management and Technology, 28, 3, 112-116.

Thach, R. (1994, June 1). Organizational change and information technology: Managing on the edge of cyberspace. Organizational Dynamics, 23, 30. Available at: http://www.elibrary.com/cgi-bin/hhweb/hhfetch?35279433x0y275:Q001:D022.

Weaver, J. (1995, December 4). Personal communication with the author via E-mail. This exchange concerned the aura surrounding Lotus Notes. Ms. Weaver pointed out the difficulty of getting Federal managers to focus on the need for orderly management of important documents when they are distracted by more glamorous technology and applications. The author's response was: "You make a good point. Everyone likes to have fun and that will continue to be a strong motivator. However, … it [shouldn't] … be the prime consideration as far as the use of tax revenue is concerned. In order to form a more perfect union, we need a little discipline to go along with the fun."

Yourdan, E. (1993). Decline and Fall of the American Programmer. Englewood Cliffs, New Jersey: PTR Prentice Hall, Inc.

End Notes

1. Malone and Halperin (1996) at MIT note that the rapid spread of networks has created an unprecedented opportunity to use computers to help people work together, suggesting the shift to "interpersonal computing" will be as important as the earlier shifts to time-sharing and personal computing. However, they eschew use of the word "collaboration" in favor of "coordination," which they assert is a "core concept" that will help companies use information technology to organize themselves more effectively.

2. Khoshafian and Buckiewicz include E-mail in the "document and forms-based" category, together with workflow automation and document management software. In the high-volume, transactions-based category, they include high-performance database management, information retrieval, and document imaging systems. "Organizational communications" groupware, they suggest, includes group calendaring, video conferencing, electronic meetings, and group authoring (but not E-mail).

3. In technical language, this category might include all "objects" that are worthy of more than transitory encapsulation and thus might be termed "object management". However, in practice and reality, the office workers who are the customers for such software still deal with "documents" rather than "objects". Thus, in the spirit of service to the customer, "document management" seems to be the best way to characterize the functionality.

4. According to Enberg (1995) analysts estimate that roughly 12 percent of all office work can be formalized, but that figure seems as much a matter of social practice and custom as of any intrinsic business requirements. In the land of the free and the home of the brave, it may be on the high side. However, in the brave new world that may be contemplated through the use of artificial intelligence and employees empowered by optimized information systems, it is surely low. The real issue is how ready, willing, and able people are to take advantage of the potential for greater organizational efficiency and effectiveness.

5. In responding to public comments while publishing its final rule on record-keeping requirements for E-mail, the National Archives and Records Administration (1995) said, "… the final standards now being issued will put e-mail in proper context with all other records … If agencies fail to create and maintain on another format full documentation of their policies and activities … e-mail could assume an inflated importance. Agencies have the opportunity to put e-mail in its proper context by issuing … record-keeping requirements that clearly state what records are to be created and maintained and on what medium."

6. If databases are used to supplement or supplant oral communications, the highest and best applications would be those that would otherwise involve formal meetings, particularly decision-making meetings. Using elaborate, highly structured, maintenance-intensive database software for transitory, informal communications would certainly be a lower priority. Thus, a key question is whether the decision-makers are ready, willing, and able to use a collaborative database in lieu of oral communications to support decision-making.

7. Dallas points out several problems with relying on mail for communication within a group: Electronic mail cannot handle the overwhelming information flow. Mail supports one-to-one or one-to-many, not many-to-many. Groupware supports many-to-many communication, while electronic mail is only designed for a one-to-one or one-to-many model. The mail discussions that occurred before the arrival of new employees might help them up the learning curve, but they have no access to them. Discussions aren't threaded; replies look like new topics. If mail has categories, it doesn't show which categories contain new information. Group communications often require more than text to make a point and extracting a file and running a separate application is too cumbersome.

8. Norman (1988) pointed out that knowledge may be in one of two places - in the heads of people or in the world. People cannot possibly keep in their heads all of the information needed to live, much less to make a living, in the modern world. Knowledge in the world can take many forms, but documents display wisdom accumulated in the heads of people reflected back into the world in a form that can readily be used by other literate people. The term that Davis and Davidson (p. 104) use to extend Norman's principle to automation systems is "embedded intelligence". With advances in microelectronics, information services can now be built into products to meet the particular needs of the customer. With advances in collaborative databases, information unique to each user will be built into such systems. However, the concept is anything but new; well drafted documents have always been responsive to the particular needs and interests of the recipients.

9. Documents facilitate formal collaborations, from contract to treaty negotiations to the passage of laws. In their book on negotiations, Fisher and Ury (1981) discuss the "one-text procedure" as a useful and often essential means of reaching agreement. It is of more than anecdotal interest that politicians and diplomats are known and satirized for talk, but that they accomplish nothing of substance until something is rendered in writing.

10. In testimony before Congress, Kinghorn (1996) posited "four C's" as being part of a rigorous skill set demanded for the successful use of information technology - critical thinking, communications, collaboration, and computer literacy.

11. Pun intended; "chase our tails" may be substituted for a more literal meaning.

12. In Congressional testimony, Di Pentima (1996) stated, "Outcome measures are far more appropriate than output measures in assessing the real value of an investment opportunity. Output is level of effort; outcome is the objective we are after." While there can be no doubt of the validity of this statement, measures of output (e.g. documents) are certainly preferable to measures of input (e.g. meetings, or number and length of collaborative contacts). Sadly, many organizations have yet to demonstrate an ability to move up the "capability maturity model" even to the point of being able to measure output reliably, much less outcomes. (For a discussion of the Software Engineering Institute's process maturity model, see Yourdan, pp. 74-91.)

13. In personal communication with the author via E-mail, Weaver (1995) addressed the difficulty of getting people in organizations to focus on that which is needed for the good of the organization versus that which tickles their personal fancy. She said, "You're asking them to spend money on something that will organize them and allow tighter management controls vs. something they perceive to be fun! That's like giving them the option of buying a new kind of tire to replace some that still seem to have some mileage left or buying a new suit of dashing clothes to go to a party where celebrities will be present."

14. While Farson doesn't say so, this is consistent with the next-in-line principle of Total Quality Management, wherein the "customer" whose needs are to be met is the person who is next in line in the production process.

15. In an extreme example of group editing in document-based cyber-collaboration, Pappano (1996) reports in The Washington Post that Mirriam-Webster, Inc., the nation's oldest and largest dictionary publisher plans to place a draft of Webster's Fourth New International Dictionary on Internet as a work in progress, for review and comment by the world. Because the process of updating and publishing voluminous (more than 2,700 pages and 470,000 entries) hard-copy is so cumbersome, this will be the first major revision in 35 years and "all literate people" are being invited to become contributors.

16. In testimony before Congress, Hoenig (1996) reported that "… one-third of all systems development projects are canceled before they are ever completed… only 16 percent of all IT projects were considered successful - that is, judged to have accomplished what was expected within the budget anticipated at the outset… only about 42 percent of the largest companies are meeting their initial objectives… over 50 percent of IT projects exceed their original cost estimates by almost 200 percent."

17. Standards are being established for the category of groupware known as document management software. Whereas IBM is endeavoring to market a proprietary product, Lotus Notes, as the de facto standard for all groupware applications, the Document Management Alliance (DMA) is establishing open-systems standards for interoperability of document management systems (DMS). DMA is organized under the auspices of the Association for Information and Image Management (AIIM). All of the market leading DMS vendors are participating, along with more than 90 companies in total, including IBM, Microsoft, Xerox, and Novell. Information on DMA is available on-line at: http://www.aiim.org/dma/

18. Among those surveyed at the GroupWare '93 Conference, 24 percent felt they were currently successful and another 29 percent felt they were somewhat successful, but 47 percent felt they were not yet successful with groupware.

19. Although Notes utilities are capable of exchanging data with SQL databases, the Notes database itself is proprietary. In other words, to use a Notes application an organization must purchase Lotus Notes from IBM or one of its value-added resellers. Thus, it is a misnomer to suggest that Notes might become a groupware standard in the sense that "standard" implies an open set of specifications within which challengers can compete and excel.

20. In its review of the "two groupware heavyweights," Imaging Magazine (1996, May) gave the nod to Notes over Microsoft Exchange. However, the magazine cited the following limitations for Notes: The forms builder, Lotus Script, is not easy to use; Notes lacks direct access to relational databases, and it entails high administrative costs.

21. While it is true as Thach suggests that replacing computer hardware can be costly, Ambur et al. (p. 87, endnote 58) found the incremental cost of hardware to provide document management services for files already stored electronically in a disorganized, inaccessible manner to be as low as $1.25 per user per year. The cost of failing to provide such services is obviously many times greater, perhaps thousands of times greater!

22. Another alternative for maintaining organizational knowledge in the world is to build it into the tools that are used to do the work, thereby eliminating the need to maintain it in crania, documents, or databases. Such knowledge might be said to be "hard wired" into the device, or in terms of animal life forms, the concept is functionally equivalent to instinct. In any event, no conscious thought must be devoted nor are humans required to generate any documents or database records in order to produce the desired outcome. For a discussion of embedded intelligence and agent-based interfaces, see Negroponte, pp. 92-93, 102 and 155.

23. At the data level, the issue is whether the organization's most important information is discrete or continuous, i.e., digital or analog.

24. IBM/Lotus are hedging their bets by building collaborative features into their wordprocessor. By virtue of those features, PC Magazine (1996, November 26) rated Word Pro as "most improved" but still not on par with WordPerfect or Microsoft Word. The magazine also suggested that wordprocessors may even swallow up E-mail applications and become the "hub around which your computing day revolves, no matter what kind of work you do." (pp. 103-104)