Technology Trends

In this section, we survey some of the technological trends that affect the success of collaborative technologies and applications.

Trends in networking and communication

  • The development of the Internet and the World Wide Web: In recent years, the Internet and the World Wide Web have been expanding at a rapid, exponential rate, as shown in the figure below, which plots the number of Web sites (servers) as a function of time at 6 month intervals. Obviously, the number of hosts on the Internet is several times this, and is also increasing rapidly with time.

    Courtesy:
    Matthew Gray, MIT

    This has led to the development of many new and innovative network applications and has also caused immense demand for high-speed and ubiquitous Internet access, in both corporate and residential environments. Three important consequences of this trend are:

    We explore each of these in turn, focusing on how they affect collaborative technologies.

    Intranets

    In the office environment, many corporate and educational institutions are starting to move from the conventional 10 Mbps shared Ethernet and (sometimes) FDDI technologies towards switch-based, high-speed Local Area Networks(LANs) ranging from 100 Mbps switched Ethernets to higher-speed Gigabit Ethernets. Asynchronous Transfer Mode (ATM) does not seem to be living up to its earlier promise (made in many circles) of being the panacea to all bandwidth-related ills for integrated data and voice/video services in that it isn't the technology of choice for LANs; however, it is likely to play an important role in the backbones of large scale corporate networks as well as the Internet. One consequence of this is the emergence of medium- to large-scale corporate intranets.

    What is an intranet? While there is no standard an accepted definition of this term, it is essentially a collection of computers and networks within a (possibly global) organisation, connecting the organisation's members to a range of computer services, resources, and information. Despite having heterogeneous machines on many physically distinct and distant networks, the software and tools on the network provide the appearance of a single, large, homogeneous network, and enable efficient communication and collaboration between employees. One example specification is Netscape's vision of the Full Service Intranet.

    Clearly, the development of intranets, which has also been triggered by increasing awareness in organizations of the value of collaboration and horizontal integration, is an important catalyst for the spread of collaborative applications and technologies. In addition to intranets, extranetsinter-organizational business strategy and partnerships. Extranets extend a subset of an organization's intranet to subsets of the intranets of other partner organizations, to facilitate convenient joint work or producer-consumer relationships.

    Residential Access

    In addition to increasing speeds of conventional modems over telephone lines (which have inherent limitations in how fast they can go), technologies for high-speed residential Internet access, such as 128 Kbps Integrated Services Digital Network (ISDN), Asymmetric Digital Subscriber Loop (ADSL), cable modems, and Direct Broadcast Satellite (DBS), are being developed and are in various stages of deployment around the world.

    In an earlier project, we have found that most residential network access technologies are inherently asymmetric in nature. Depending on the exact details of the system, the ratio of bandwidth in the "forward" (towards the client) to "reverse" (away from the client) could be anywhere between 10 to 1000. The reasons for the development of asymmetric access technologies, such as ADSL and cable modems have to do with the fact that it is a lot easier and cheaper to provide higher "downstream" bandwidths than in the "upstream" direction. Upstream connectivity is provided either by means of conventional dialup phone lines or lower-bandwidth, higher-contention cable lines. The push toward asymmetric technologies for home use is not simply a consequence of technology -- it is also motivated by the nature of traffic in today's most popular network application -- Web access. Most Web sessions involve the transfer of far more data towards the client than away from it; the return channel is used mainly for Web requests, forms, and of course, TCP acknowledgments of received data.

    However, the continuing deployment of highly asymmetric access technologies implies that applications that involve more symmetric data transfer, such as symmetric video conferencing, would not achieve particularly good performance, especially in the upstream direction. Even the best compression algorithms known today need at least 128 Kbps for acceptable conferencing performance to provide the participants with the required sense of presence. This affects the level of participation telecommuters can achieve and make their presence felt in meetings that they participate in from home.

    Wireless Technologies

    In the recent past, there has been a proliferation of wireless network access technologies, both in research labs and in the commercial marketplace. Available technologies include both local-area and wide-area ones. Local-area technologies include in-room ones based on infra-red and in-building ones based on spread-spectrum technologies, and have bandwidths between 1-4 Mbits/s. Wide-area wireless networks, such as Metricom's Ricochet have higher range than LAN wireless networks, but substantially lower bandwidths (30-100 Kbits/s). They are therefore not suitable for applications like video conferencing, although two-way audio and shared whiteboards work quite well. Our interest in wireless is primarily at the LAN end of the spectrum, and particularly on how it can be used in conjunction with a network computer to provide ubiquitous network connectivity and enhance collaborative activities in office environments (e.g., hospitals, factory floors, etc.). An example of the sort of product we are talking about here is U.C. Berkeley's Infopad research prototype, which is a wireless multimedia terminal coupled with backbone network and application support that provides real-time multimedia conferencing capabilities.

    The trend towards increasing network bandwidths and ubiquitous network access augurs well for collaborative and conferencing applications, especially ones that involve real-time multimedia dissemination. Improving network hardware, coupled with improvements in routing and data transport technologies (e.g., multicast) as well as conferencing software tools, are opening up the possibilities of large-scale collaborative applications on the Internet. In the remainder of this section, we discuss these issues.

    IP Multicast and the MBone

    Multicast is an efficient way by which distributed applications can achieve point-to-multipoint or multipoint-to-multipoint communication and data transfer without unnecessary and redundant replication of packets in the network, using bandwidth and network resources efficiently. IP multicast is the collective name given to the routing protocols that support multicasting over IP on the Internet. The Multicast Backbone, or the MBone as it is called, is the multicast-enabled virtual network that is laid over the traditional (unicast) Internet for multicast applications. Collaborative and conferencing applications are some of the main beneficiaries of multicast, and indeed, such applications were the main reasons for the development and recent proliferation of this technology. The MBone has grown substantially in its relatively short lifetime from its inception in 1988. Today, there are over 3000 multicast-enabled subnets in the world, and this number is increasing with time. A topology map of the MBone as of August 1996 may be obtained from here. Some interesting information about the history and future of the MBone, IP multicast, and conferencing applications can be obtained from this interview with Steve Deering, one of the pioneers of the IP multicast infrastructure.
  • Trends in computer and multimedia hardware

    Several computer vendors are adding extensive capabilities to support multimedia. For exampe, Intel's new microprocessor codenamed P55C by Intel, will be available in volume at the end of 1997. It has added multimedia capabilities and special instructions that are likely to benefit video conferencing applications. These chips enhancing the PC's video and multimedia performance will be used as co-processors in a Pentium-based system.

    Although video-conferencing has been present for a long time, prices for high-frame rate systems have been high and have precluded general and widespread acceptance. With the advent of efficient hardware and software compression, PC-based desktop conferencing systems are poised to grow rapidly in the next few years [IEEE Spectrum Forecast, Online version].

    LiveBoard, from Liveworks Inc. is another example of collaborative technology we are addressing. LiveBoard integrates multimedia presentations, document conferencing and video teleconferencing with an intuitive shared whiteboard that links geographically dispersed teams from anywhere in the world to collaborate in real time.

    Trends in Software

    In parallel with these advances in networking technology and computer hardware, software for real-time conferencing has also made significant progress in the last few years. A lot of work in the research world has led to technology transfer and commercial ventures aiming to bring conferencing tools to the mass Internet. Examples of companies involved in this include Precept, VXtreme, etc.

    Conferencing tools that are freely available include video-conferencing tools, audio-conferencing tools, and shared workspace tools. A screen-dump of these tools in operation on a desktop during an active day on the MBone is shown here (picture courtesy: Steven McCanne).

    Economic Factors

    There is overwhelming anecdotal evidence of the economic benefit of collaborative technologies:

    Increased productivity:Collaboration improves productivity in a number of ways. First, it can efficiently bring together a broad range of experts much quicker then conventional methods. By having all knowledgeable and impacted groups represented, decisions can be made quicker and with greater assurances that they are correct. Collaborative technologies that enable employees to telecommute increase productivity and job satisfaction by allowing the employee to avoid unproductive commute time as well as the associated stress. Various studies suggest that employees tend to work more hours when telecommuting, presumably using the extra time the did not spend commuting for the betterment of the organization

    Direct cost savings: Collaborative technologies can also offer an organization significant, direct cost savings, specifically on travel and other related expenses. Organizations have already effectively replaced some traditional meetings with teleconferences. As the functionality of the collaborative toolset expands, more and more meetings can be moved from the conference room to the network at a great cost reduction for the organization.

    Indirect cost savings: There are also a number of ways in which an organization can accrue indirect cost savings through the use of collaborative technologies. For example, the ability to increase worker productivity either through telecommuting, facilitated sharing and dissemination of information, and mobile connectivity with the office can result in significant cost reductions. In addition, the ability to collaborate with someone in a different time zone in real-time can have immeasurable value, particularly in fast changing environments. The problem with these cost savings, of course, is that they are very difficult to measure, and as the old saying goes: If you can't measure it, you can't manage it.

    On the other hand, there are significant costs associated with the adoption of collaborative technologies:

    Hardware and software: An obvious and important cost of implementing collaborative technology is the actual costs of the hardware and software, plus the necessary infrastructure investments. The following graph illustrates how quickly these costs can rise. While an email "seat" costs an organization around $50, the cost of a "seat" for collaborative tools can be 6 to 20 times as high.

    Switching costs: Aside from the acquisition costs of the required hardware and software, there may be significant switching costs due to the learning curve effects as well as the displacement of existing tools used to achieve some of the desired outcomes. In addition, there are significant path-dependent effects, particularly since the industry is new and standards are still in there infancy. As a result, a company operating in a Windows NT environment may be limited as to the collaborative tools it can implement without incurring additional costs.

    Network externality & Intranet applications: Network externality is also a significant factor in the deployment of extranet based collaborative tools; however, these network effects are substantially reduced for intranet implementations. For intranet applications, the organization's value proposition is derived primarily from the internal use of the application, which in turn is determined by the organization. When the technology is deployed, the firm instantly gains as much value from the technology as it chooses to invest. It has control of where to deploy the tools, at how many locations, and for what purposes. While this scenario might change for a single profit center within the firm, from the perspective of the organization as a whole the network is self-contained. This fact implies that such decision should be made at the corporate level when possible since a primary benefit of the technology can be cross-division collaboration. It also implies that, at least for intrabusiness applications, many diverse collaborative technologies can survive since wide-spread adoption is not required for the application to provide value

    Network externality & Extranet applications: When the discussion moves to extranet applications, the equation completely changes. Network externality effects become increasingly important, and successful development firms will need to be widely adopted or comply with the prevailing standards. The desire for organizations to collaborate with each other will eventually push the market beyond internal solutions. At that point, the value of the application will increase significantly with each new adopter of the technology. Since firms do not have much (if any) control over the computing environment of their partners, interoperability and wide spread adoption will be critical to the success of the tools. These applications will require a different pricing model then the one employed to enter the intranet markets since the value is now tied to the number of adopters.

    Because of these factors, we believe that collaborative technologies will first become widespread in intranet realms (i.e. for use primarily within a single firm) and some highly coordinated cross-firm efforts. These applications will prove to be the testing ground for the technology and new work paradigms. As these paradigms are refined and the technology becomes more universally accepted, dominant players will emerge that can provide value beyond just the organizations network either because of their market penetration or the emergence of industry standards. At that point, network infrastructure issues and interoperability will become the key determinants of the success of the industry.

    Until then, we believe that the technologies will develop to meet the needs of the internal, or intranet, organizational markets. We believe that the business case for organizations to use collaborative technologies internally alone is sufficient to justify the expenditures they will incur, and many organizations are already validating this belief. Our case studies discuss the economic benefits accrued to the organization and shed some light on how to capture the cost-benefit information required to make the decision.

    Human Factors

    Human factors will likely play a very significant role in the success or failure of collaborative technologies. We looked at different human factors and their impact on both the development and implementation of collaborative technologies.

    Human resistance to change: Humans are inherently averse to change. This resistance to change is not unique to collaborative technologies. It is true of any new technology or change in business process. The upside and downside in these technologies is they have such a great potential impact on the way people work and communicate, that it magnifies the degree of change and can engender strong opinions either for or against the technology. Effective planning for change drastically improves the probability of success.

    At General Foods, which implemented a more basic collaborative tool - scheduling, fear kept the project from taking off for five years before it was finally overcome. The most common concerns were loss of control of personal calendars and a fear that they would be inundated by meetings because they were now easier to schedule. Management eventually overcame these fears with a two-pronged strategy. First, GF rolled the product out in a phased manner and only gave it to specific functional groups to start. These were groups that had considerable contact with everyone else in the division, so the word spread that these fears were unfounded. The second prong of the attack was to publicize the success of the project. GF published articles for the company TQM newsletter about how much time the scheduler saved in the initial pilot tests and how easy it was to use. The word got out, and the fears evaporated.

    If this much fear can be generated by a scheduling tool, imagine the range of reactions, for example, to a system requiring individuals to collaborate over the internet with a company that was once considered a competitor. Even intra-firm collaboration can be perceived as a prelude to being replaced, laid-off, or relocated to a new division. Middle managers tend to be particularly fearful of collaborative technologies, viewing it as a threat (perhaps real, perhaps perceived) to their power and organizational domain.

    We will discuss in further detail how an organization can address these concerns through training, communication, and reshaping of the corporate culture in the section on Industrial Organization. Suffice to say for now that an adopting organization will fail miserably at implementing these technologies if these human factors are not considered. Collaborative technology companies, for their part, must be prepared to help organizations recognize and plan for these issues. If the industry does not help prepare organizations, they run the risk of seeing the implementations fail and having the technology blamed for the failure. An early set of such horror stories could permanently stifle the young industry.

    Human-computer interface: There are several very critical interface issues key to the successful integration of collaborative technologies into organizational work processes. First, the interface must be easy to use and intuitive. Given the high costs of training, successful tools will be developed in a way that minimizes individual learning curves. In addition, the tools should be developed to minimize the organizations learning curve. In other words, the tools must not only be easy to use, they should also fit seamlessly into the organizational processes. These goals can be accomplished through careful attention to the interface design details, standardization of basic commands across vendors, and by leveraging off the basic look and feel of existing GUI designs.

    In addition, collaborative technologies will need to find a way to address the more challenging problems of simulating physical presence in a collaborative setting. The human interface will need to capture important non-verbal gestures in order to more fully capture all communication in a virtual meeting space. Body movement and facial expressions convey important information that, if successfully captured in virtual space, can greatly enhance the value of these meetings. A related issue is the subject of floor control. While neither of these issues is likely to stunt the growth of the collaborative technology industry, companies that come closest to emulating these important, real-world meeting dynamics will be able to push video conferencing applications beyond existing bounds.

    Socialization implications: There are also many social habits that, while impacted by the human interface, are not likely to be overcome by it. Humans have a basic need for presence, a need that is greater for some interactions and less for others. Collaborative technologies will attempt to make up for the lack of true physical presence through the use of sophisticate GUIs and other objects. However, we believe that certain social customs, like the handshake symbolizing agreement to a deal, cannot and should not be replaced by these tools. Clearly people will continue to meet in person for significant, lengthy, or intimate discussions. Collaborative technology companies should recognize this distinction and focus their energy on features that will best support interactions best suited for remote interaction.

    Industrial Organization and Management

    Industrial organization issues will play a major role in the development of collaborative technologies. The collaborative technology industry is a young, horizontally integrated industry. It does not have a set of standards because of the immaturity of its products and the overall market. These characteristics create a complex set of organizational challenges for companies competing in this market. In addition, their customers are faced with new organizational challenges and opportunities created by the technology . This section discusses these challenges and pitfalls.

    There are several organizational decisions which firms must address:

    Complementary products: Because of the horizontal structure of the computer industry, there are numerous opportunities for niche players to develop successful products. These companies benefit greatly from the complementary nature of the technology industry. The collaborative technology industry as a whole was developed in part by these niche companies feeding off the explosive growth of computing power in the United States. The industry still depends heavily on complementary products, none more important than the network, in order to help promote the growth of the industry. While the presence of large companies like Xerox, Microsoft, and Netscape has given more credibility to the industry, it is still very vulnerable to the cycles of its complementary products. Different tools within the industry are also collaborative. As a result, firms may want to work together on design or implementation efforts to leverage these complementary skills for the benefit of both companies.

    Standard setting: There are very few standards in the collaborative technology market. The Workflow Management Coalition (WfMC) has developed a number of specification for the transfer of data between process oriented workflow programs, and some initial standards discussions for group calendaring and scheduling software for the inter/intranet is still being debated by the IETF. However, standards for more robust tools like shared whiteboards are still several year from materializing.

    Perhaps because of the youth of the industry, collaborative technology companies have not yet rallied around a single standard. This development is detrimental to the industry in many ways. First of all, companies are more reluctant to purchase a collaborative technology solution for fear of lock in to that provider. This concern is amplified by the fact that the technology may be incompatible with the technology being used by a potential partner, making it either useless or expensive to integrate with the alternative application.

    Standardization does not, of course, mean commoditization. The standard that emerges will likely be one of the least common denominator, i.e. the set of tools or functions that all the warring parties could agree to support. Differentiating functions will be implemented on top of the standard, and the battle will move from pushing for a standard to delivering more robust functionality.

    New relationships with partners: The collaborative technology industry needs to develop new relationships in order to push its way further along the adoption curve. Traditional relationships with hardware vendors will continue to be important, but the new technology requires new competencies to be successfully implemented. Many of these developer simply do not have those skills.

    For example, companies may want to form alliances with consulting firms. The organizational changes required of companies implementing collaborative technologies is tremendous, and it is critical to the success of the industry that these early adopters have a successful experience. Developing relationships with consulting firms, particularly firms with core competencies in change management, will allow collaborative technology companies to play a proactive role in supporting the growth of their market.

    For companies implementing collaborative strategies, they must be prepared for:

    Organizational resistance to change: As much as individuals resist change, organizations resist them even more. Few things are as unnerving as change, particularly in the business environment. Most change is assumed for the worse, and fears of job security, changing reporting responsibilities, and loss of control are particularly high with changes involving collaboration. The fear and resistance should be met by open, honest communication that sheds light on the true nature and intent of the proposed changes

    Organizational restructuring: With the fear of change relieved, companies must then face an even greater challenge: managing the entire spectrum of organizational changes required to create an environment supportive of collaboration. More then just the implementation of new technologies and procedures, collaborative technologies require a rethinking of the entire organizational structure. It is important to note here that while these changes sound ominous, they are in fact quite healthy for the organization (if perhaps potentially painful). Done correctly, a company emerges from these changes more focused and more competitive in its market.

    A study that examined the results of an implementation of collaborative technologies in the office at a large corporation found that key organizational elements influencing the effectiveness of the new collaborative tools were the structural properties of the organization such as policies, norms, and reward systems. The findings suggest that where the premises underlying the collaborative technology (shared effort, cooperation, collaboration) are counter-cultural to an organization's structural properties (competitive and individualistic culture, rigid hierarchy, etc.), the technology will be unlikely to facilitate collective use and value. That is, where there are few incentives or norms for cooperating or sharing expertise, collaborative technology alone cannot engender these.

    Organizations looking to implement collaborative technologies must be prepared to reevaluate themselves from top to bottom. They must pay particular attention to the culture, reward systems (both implicit and explicit), and reporting structure of the organization and take the necessary steps to realign the organization with its new objectives. The organization must follow up these changes with comprehensive communication and training plans.

    New relationships with business partners: It is clear from the preceding discussion that to fully embrace collaboration is not a simple endeavor. For this reason, companies may find themselves in need of outside help to manage through these changes. Particular expertise in organizational design and change management is critical to the successful implementation of collaborative technologies.

    Of course, implementation of collaborative technologies also brings to the organization a multitude of new partnering opportunities with suppliers, customers, developers of complementary products, and even competitors. The course of action an organization should take will differ drastically depending on its unique competitive environment and the collaborative "readiness" of its would be partner. These collaborative efforts could result in joint developments of new products, improved efficiencies in the supply-chain, and a host of other opportunities.

    Legal and Regulatory Factors

    There are two primary ways that legal and regulatory factors are likely to influence the adoption of collaborative technologies in organizations.

    Regulatory actions: The first way is through possible regulatory acts related to the network. These acts are most likely to take the form of regulatory measures intended to support the development of a network that is better suited for symmetric transfer of information, to achieve universal service of this w network, or to regulate the use of the existing network

    For example, the emergence of Internet telephony as a potentially viable alternative to PANS for at least some uses has raised the ire of some telecommunications companies and prompted an FCC inquiry into the subject. If the FCC rules that voice over the Internet may be subject to regulation, it could have serious consequences on the development of collaborative technologies by reducing at least one of its cost advantages over traditional collaborative communication techniques. It would be a small, perhaps inevitable step to regulate the transmission of video over the network as well. The impact on desktop video conferencing as well audio conferencing, along with the collaborative technologies built around them, could be devastating.

    We believe the U.S. government should resist the urge to regulate this developing market. Public policies to promote universal service for the new network infrastructure should not be based on a monopoly model of a network. The environment now is completely different then the one that existed at the turn of the century, when the telephone network was first being developed, and we believe that there are sufficient market incentives in place to ensure the continued development of the infrastructure within the context of a competitive environment.

    The convergence of communications technologies provides a variety of infrastructures over which network services can be delivered. The wide availability of alternative telephone, cable television, and wireless networks will broaden the reach of the network. Regulatory conditions governing interconnection and charges for access can increasingly bias the use of particular supply arrangements. Government efforts should instead strive for neutral policies that avoid pricing distortions among messaging systems. The mere threat of government intervention is likely enough to prevent monopolistic behavior in the market.

    Legislation: The second way is through legislation. Environmental concerns and civil liberty issues, particularly for the physically disabled, are both significant issues that the government is actively involved with. Two recent acts, the Clean Air Act and the Americans with Disabilities Act, reflect the sentiment in government. Unlike regulation of the network, however, these acts play into the hands of collaborative technology companies. A benefit of collaborative technologies is that they provide fairly extensive social benefits in the forms of reduced air pollution, greater accessibility to information, and an alternative form of communication and socializing.

    Since the trend in these two areas continues to be strong, continued legislation in these areas is likely and may help swing the tide in favor of greater use of collaborative technologies by incentivizing organizations to adopt them.

    Case Studies

    Overview of Trends

    Contents