Virtual Office 

Contents 

• Abstract 
• Introduction 
• Advantages/Disadvantages of Central/Distributed Work 
• Specific Tradeoffs 
• Technology Needed 
• Experience Report after Year One 
• Conclusion 
• Appendices 
• For More Information 
• Our Virtual Office Policy 
• Glossary 

Abstract 

Object Services and Consulting, Inc. (OBJS) is structured as a virtual office. At present it has eight full-time employees and a ¾-time office manager spread across six geographic regions of the U.S. and is executing two R&D software contracts and providing some external consulting. There is no central work location and employees report to work from their home offices. 

The first version of this paper was written when we were making the decision to adopt a virtual office. So most of the paper is written from the perspective of the company in its early months when we were collecting together thoughts, justifications, risks, costs, and other considerations that were considered in choosing this non-traditional path. The current version of this paper has been annotated with experience we have gained after a little over a year of experience. So far, we report success. 

The style of the paper is informal. If focuses at least as much on business processes and social issues as it does on technical requirements in a virtual office. For a review of some related technologies needed to Internet-enable the virtual office, see our Internet Tools Survey. One of our company's premises continues to be that Internet-based collaboration technology is now adequate and is fast improving to make this kind of work arrangement effective but there are many factors beyond technology to consider. Another paper on Visions covers other future directions made possible by Internet technology.

We believe the trend toward virtual enterprises will rapidly increase as people learn how to collaborate-at-a-distance crossing boundaries of time and space to solve information-based problems "better, faster, cheaper" using best-of-breed resources not necessarily available at a single location while gaining increased control over the quality of their lives. 

Introduction 

This paper describes considerations that led us to constitute our new small business Object Services and Consulting, Inc. as a virtual, or distributed, office. 

When we formed OBJS in September 1995, we assumed it would have a traditional centralized office setup. One of our guiding principals was to innovate technically but not reinvent the wheel where company administration was involved, instead just adopt traditional means of doing business. Certainly, it is traditional to have a central office and we mostly worked this way in the past with occasional telecommuting. We did not initially question this-we just assumed we would have a central office by default. 

But it takes time to build out and furnish office space, so we worked from our respective homes for the initial three months, while we organized the business and searched for office space, and, in that time, we found in many ways that this arrangement seemed to work as well or better than a traditional office. Furthermore, our OBJS General Counsel lives in Pittsburgh and one of our prospective first new hires lived out of state and hoped to continue to do so. Finally, our company focus is computer software R&D on distributed work, collaboration, and information access. So, as the decision to sign a lease was approaching, we questioned whether to do so or instead continue and scale the experiment to future employees. 

This was an important decision for us because it represented risk, a deviation from the norm, and might affect the way we work and interact, the kinds of employees we could hire, and ultimately the kinds of work we do. There are some but not many models of this kind of enterprise in our business so we needed to consciously assess this course, including providing for a graceful fallback position if it became apparent that the experiment was not working. 

This paper, then, provided us a vehicle for assessing whether to constitute OBJS as a virtual office, and, having made that decision, for recording our justifications for this experiment and how to monitor it as we moved forward. We are doing this for ourselves, for the benefit of our government sponsors, and because distributed work is directly relevant to our government contracts and our experience will provide scenarios and demo testbeds for remote collaboration relevant to our contracts. Our government sponsors initial reactions were: 

• Dave Gunning, Dec 19, 1995. "This sounds like an interesting experiment." 
• Barry Leiner, Dec 30, 1995. "While I am aware of a number of organizations that have satellite locations and people who telecommute, I am not aware of anyone that has gone "fully distributed" as you are suggesting. I agree that it is a worthwhile experiment. Let me know if there is anything I can do that would help, and keep me informed as you learn lessons (good and bad)." 

At the time of this writing (January 1997), OBJS has grown to eight full-time employees and a ¾-time office manager spread across six geographic regions of the U.S. The experiment is working very well but we are still learning as we put in place the technical and administrative infrastructure needed to continue this style of work. 

Type of Business We Are In 

Not all businesses are amenable to becoming distributed workplaces - in particular, businesses that depend on physical movement (e.g., construction, fast foods) may not be though even some of these necessarily are distributed (e.g., transportation and regional or customer-site sales). But many information service businesses are (e.g., travel, real estate, discount brokerages, regional sales, transportation, and military operations). Some of these involve fixed satellite organizations and others involve mobile teams that must be able to communicate effectively. Management consultant Peter Drucker estimates that as many as 30-40 percent of today's workers are knowledge workers. 

In our case, we are developing research software for distributed information management applications. Software is typically developed in teams. A lot of effort goes into early evaluation of software we might make use of and into collaborating together in design and later in integration with others on a team where occasionally team members come from other organizations. Software distribution and support is increasingly via the Internet. Debugging is often remote. Reporting to our sponsors is via the Internet and occasional in-person reviews. So the most critical area where centralized work seems to be required is in insuring good communication within and between teams. Physical proximity does not guarantee this but may help enable it to an unknown extent. An abiding question is, when do we need physical proximity if we can provide virtual proximity? 

Finally, we are working on distributed collaboration application scenarios in our contract. If we are successful and we document and advertise our virtual office model, then this may become a template for others in industry, including others in the DoD contractor community. Most DoD contractors produce one-of-a-kind products and not open systems, which we focus on. Still, it is important to separate our internal use of collaboration technologies from any interest we may develop in delivering a certain kind of collaboration solution as products. 

Past and Related Experience 

Our current team has a track record of remote, collaborative work. Two team members worked off the main project site for two years at one point. One worked for two years with a remote software development team in another city and also in field service with teams across the country. Another past team member worked for six months from India while developing software using fax and email. Collaborators at other sites have developed software related to our designs, and we have either integrated it into our system or invited them to visit for periods of 1-3 months to work with us to develop parts of the system. We have long used the Internet to distribute software and to remotely debug customer software and email for support. All of our work in the standards community is collaborative with only occasional meetings and most work happening between meetings. Our architecture work for the NIIIP Consortium has involved developing the technology infrastructure for virtual enterprises. 

Terminology 

The Glossary section defines these terms relevant to virtual offices: 

• alternative work arrangement, boundaries, business processes, collaboration, corporate memory, encapsulation, gratuitous communication, home office, home workers, infrastructure, Intranet, middleware, mobile workers, mobile office, nomadicity, personal environment, remote employment, remote work, tele, telecommuting, telepresence, telework, telework centers, videoconferencing, virtual, virtual office, virtual company, virtual corporation, virtual enterprise, virtual organization, virtual team

Advantages/Disadvantages of Central/Distributed Work 

It is clear that there are tradeoffs between fully central and fully distributed organizations. Also, it is clear that central and distributed are extremes and some hybrid in between may work best and in fact describes most organizations. Finally, it clearly depends on the kind of work one is doing. But what is not so clear is what factors dominate the equation, especially if an organization is committed to breaking down barriers to make a chosen approach work. 

Some advantages we see for centralized work 

• Human Communication. People in a hallway are collocated so they can easily interact one-on-one or in groups, just by walking next door, down the hall, or going out to lunch together (gratuitous communication). Sometimes, chance hallway interactions lead to main directions. Still, proximity is no substitute for communicating directions and deciding how to partition work. People on another floor, hallway, or in the same office but in another project or division can all be too far away to communicate with. Similarly, people are only one sort of information source; journals and now the web are increasingly excellent links to others with similar interests no matter the distance. There is no substitute for excellent people, careful planning, and good communication, whether in a centralized or virtual office. 
• Facilities. Facilities are centralized and shared so costs can be lower. This includes office space (offices, meeting areas, computer network, local telephone access, access to secretarial support, machine support, library, and supplies). But facilities themselves cost money to configure and maintain and may be difficult to evolve or scale as the organization evolves. 
• Been Done. The obvious one. The centralized office paradigm is what people have been assimilated into from kindergarten on. Most management schemes assume a centralized environment. Of course, occasional telecommuting today is tried and true, and management can adapt traditional methods to new concerns. In fact, this is the subject of a master's thesis by Bernie Kelley and Bruce McGraw. Kelley and McGraw surveyed 28 managers about how they have adopted their styles to facilitate telecommuting within their centralized setting. 

Some disadvantages we see for centralized work 

• Cost of Office Space. If distributed work succeeds, we save around $40-50,000 a year in lease costs for a 10 person office, which we can use in other ways, including to improve our communications/collaboration support. This would not be a good enough reason to make the experiment if we did not see other bigger advantages. 
• Cost of Commuting. If an average commute to work is 30 minutes one way, then there is a built-in, every day requirement for wasting an hour a day. Also, there is a deterrent for returning to the office on weeknights or weekends or to take care of small tasks. Not to mention pollution. 
• Cost of Attracting Talent. Experts by definition have experience; they are often mature and have settled somewhere. Relocating them and their established families is prohibitive in expense and quality of life unless it can be amortized over many years of service. 
• Flexibility of Office Space as the work group grows and shrinks. Constantly upsizing and downsizing facilities as the amount of work changes adds additional overhead. 

Some advantages we see for distributed work 

• More time to work, easier access to work, employee controlled quality of life. Telecommuting is becoming more widely accepted. Department of Commerce is actively experimenting.. A home office makes it easy to get to work, meaning more work is done over a day since employees can readily access their computing environment. A mobile office extends this high availability to going where the person goes. In addition, the annoyance and cost of the commute is eliminated. Generally, the money saved by not commuting should cover the cost of extra utilities. 
• Self-Management. Employees in a Virtual Office are free to work without being intruded upon by management. Self-management also cuts down on non-productive work such as repetitive memos, reports, presentations, administrative meetings, etc. which can plague some corporations. Also, employees now have a better opportunity to fully develop their ideas as they envision them with less interference and micro-management. 
• Potential to hire remote workers who do not want to move to the home office region. It may be much easier to attract talent if physical relocation is not a major issue. Depending on international law and contract restrictions, it may also be possible to hire foreign workers for some tasks. 
• Part time workers. It might make sense to hire some experts part time, especially university collaborators who might retain their faculty position but team with us as industry partners trading time for equity or salary and the chance to affect industry directions. 
• Distributed collaboration might be our best chance of competing with much larger corporations in the open systems area. It uses electronics to break the space barrier, keeping us on the leading edge, giving us an interesting competitive advantage. Looser bindings to a single large corporation may provide us other innovative business model opportunities: for instance, we could provide distributed R&D to large companies who no longer maintain in-house R&D staffs along the lines of a distributed MCC. Or we could evolve a model of distributed ownership and micro/pyramid equity. 
• Because infrastructure technology available for virtual offices is now nearly comparable to enterprise infrastructure technology. With home ISDN bandwidth and laptops, reasonable communications can be in place for point-to-point and team communications that are beginning to be comparable to those supplied in large centralized organizations. In addition, at least as much communication occurs with those outside the organization as with those inside so there is less special benefit to proprietary internal communications since it is likely that standard versions will become generally available a few years later. 
• Home office income tax exemptions. While many people maintain a home computer, they often do not qualify for home office exemptions. A distributed, home-based work environment changes that. 
• Maintaining just one and not two offices. Many employees maintain a home office with equipment supplied by their employer, meaning the employer is maintaining two officer per employee and the employee is porting information back and forth between the two on a daily basis. A single home office reduces the cost and "centralizes the work." 
• Preexisting Equipment. Many new hires already own computers. This is a small bonus for the company because any appropriate equipment that the employee already owns might be available for free in the Virtual Office, perhaps at least reducing up front costs and improving cash flow for new businesses. Next to the actual land and building, the cost of workstations constitutes the greatest up-front cost for a centralized office. Appropriate policies are needed to separate employee and company owned equipment and software. 
• Practicing what we preach, taking our own medicine, eating our own dogfood. The virtual office encourages us to use technologies we develop and be a test bed and early adopter for collaborative technologies being developed by others. If we learn how to work in a distributed environment and what the short-comings are, we will be better equipped to develop technologies that will help our customers do the same. Learning to work this way is what some of our customers, soldiers in the field, experience. In addition, it is likely to be the wave of the collaborative future, if we believe and champion the virtual enterprise story that small distributed best-of-class teams will need to be able to work together to develop best-of-class technologies. 
• Prototype how DARPA could rapidly assemble distributed high-performance teams. The logistics systems for Desert Storm was successful but required relocating a team for several weeks to build it. What if we can demonstrate a similar capability but leave the people in place. We should consider a scenario in which we demonstrate a capability including the organizational aspects and deliver technology to make it work for a new team. Such an arrangement might lead to distributed teams of armchair combat programmers including remote expert observers that follow teams and provide them advice on an as needed basis. 

Some disadvantages we see for distributed work 

• Communications and Camaraderie. Communications between employees is more difficult than walking down the hall. This is a central concern since interactions and negotiation are important not just in initial brainstorming but also at later stages of software development. Building camaraderie from a distance is more difficult. It is not the same as being with each other frequently -- the socializing aspect is missing. If we could see and hear each other over the Internet, a large part of this need for human interaction might be reduced.
• Physical Meetings. Finding occasional meeting space on short notice or for short periods is not easy. 
• There are only a few situations where we currently depend on physical proximity - officers co-sign checks and local people get together for interviewing new job candidates or for occastional discussions.
• Small internal business meetings of 2-4 can take place at homes and restaurants. Scaling this to larger meetings of 10 or 20 requires preplanning for face-to-face interactions. But such meetings are infrequent, are often information dissemination meetings, and may not be required. Reviews with customers or sponsors can take place at remote sites but there is occasional need to meet locally without much preplanning. Hiring summer students means housing them and working with them closely. We can rent space for occasional meetings and reviews, and seasonal junior workers may use their living space without creating insurance problems. 
• Also, if we hire employees across the country or across the globe, we may need to meet occasionally. Actual meetings require air travel and accommodations and per diems, must be preplanned to be cost effective, and cannot happen with high frequency. Technology alternatives are phone, teleconferencing for groups of three or more, and Internet-based video conferencing. Today's video conferencing solutions are either expensive and good quality only in a LAN but not WAN or cheap and unacceptably low in quality; they are limited to one-on-one situations with very little motion and scale poorly to team meetings. In 1997 we expect this to begin to change. 
• "Seems unprofessional." Many customers will not feel comfortable meeting in home offices for many reasons (e.g., invading privacy, atmosphere not professional, distractions, locating homes in neighborhoods). There is also a general impression that employees are not really working when they are at home, that the business is not real. This may not be as much an issue for us as for some other kinds of businesses since most visitors are close collaborators who already have an opinion of our work. Job candidates meeting in home offices have reported the interview experience as unusual but not uncomfortable. DCAA auditors are accepting of a professionally run home-office. 
• Opportunity for abuse. Are employees really working 100% when distracted by TV, family, perfect weather, etc.? Are timesheets accurate? Although abuses can happen in central workplaces as well, there might be an initial thought that abuse is less common with management's close presence. Our experience is that trust is more tangible in a Virtual Office than in a centralized setting. Ultimately, the company needs to hire people it can trust. Reports on the web indicate that trust in distributed offices appears to improve employee job satisfaction and productivity. Still, mechanisms should be put in place to help combat defrauding the company. One of these, required by our government customer, is timesheets which can be filled out by the employee and logged weekly or monthly. Another safeguard is two signatures on checks. Policies and Procedures are needed to insure that employees are available in their work area during business hours or leave a message telling when they will return or leave a forwarding address; the use of an electronic whiteboard may also fulfill this function. Monitoring machine activity is another possibility, but this would be unilaterally unpopular and might require monitoring computers which may also contain employee personal correspondence-this we deem unacceptable. Employees need to know what we expect them to do, what the guidelines are, and need periodic feedback. Ultimately, the best measure is to assess the quality and timeliness of work to insure fair value is returned for competitive pay. 
• Employees Work Environment. Not everyone is cut out to work in a virtual office. A person must have self-discipline and be self-motivated. Working from home day-after-day is not the same as occasional telecommuting There is also the aspect of one's spouse having you around all day when they are not used to it. It is very important to separate ones work space from ones living space, e.g., having ones office upstairs or in the back of the house so distractions are minimized. Some employees may not have a work space that is conducive to remote work or may not wish to work from home. Of course, these people may choose not to work for the company. Or the company can rent individual office space though that may not be cost-effective. 
• Training. In a centralized office, training is usually accomplished by bringing a group of employees to a meeting place. This is a problem in the Virtual Office where such areas do not exist. One solution is the use of seminars instead of on-campus training, but this has the disadvantage of being expensive and inconvenient, requiring that employees to travel from their workplace. Videoconferencing provides a reasonable alternative. In our case, much "training' involves attending conferences, which requires individuals to travel anyway. 
• Updating Technology. With each employee in the Virtual Office using his/her own machine, updating both hardware and software across the company might appear to become harder. It can no longer be a central or work group operation but instead every employee must bear some responsibility for environment set up and maintenance. Actually, large centralized organizations share this same support problem, especially when the ration of support staff to technical staff gets high, and this is driving technology toward zero-effort maintenance, incremental hands-off upgrades, and Java, but these capabilities are by no means standardized or universal at this point. In our case, we have a well-trained workforce and this has not created a problem. Another factor is that employees living in different regions may not have access to new communications infrastructures like ISDN, especially in rural areas, and employees in other countries may be subject to later release dates for software. We have not found this to be a problem. A downside to distributed work it that employees currently must maintain their own computer and its software. Also, we report some problems with workgroup software interoperability, especially emailers that process HTML tags incorrectly.

Specific Tradeoffs 

We considered issues related to the following areas that could be affected by distributed work. A main metric is reducing abstract distance, which includes not just the physical distance between sites, but also visual communication, low cost, and effective office processes and protocols. 

Legal 

• Overall. Nothing about the form of incorporation or the company operation is affected. Nothing about approvals of contracts, licenses, patents, etc. is affected except in so far as the corporation obeys the laws of the states and jurisdictions it operates in. When we first consulted our lawyer for potential pitfalls, he saw none and instead asked how to configure his laptop to best connect to us. 
• Approvals from Government sponsor. The Government sponsor needs the assurance that the distributed office approach does not (a) cost more, (b) risk our success, (c) expose our results to those without approval. 
• Jurisdiction. The company is required to follow the state and local laws of the area in which an employee works, independent of the state in which the company is incorporated. 
• Legal postings. The company is obligated to display posters informing employees of their legal rights. In a Virtual Office, these can be scanned and made available via informational mailings or corporate shared electronic bulletin boards. 
• Employment papers. Some things need to be witnessed, others need to have documents checked (I9). Notaries may be required in the virtual office. 
• Zoning Laws. Many cities restrict business use of homes. Common ordinances preclude excessive noise, signs, excessive traffic, employees other than those living in the residence. 
• Taxes. Each employee is a source of revenue so we will have to file tax returns in each state that we have people working. We will also have to make sure that we know the taxing requirements of each employee's locale. The decision was made to go with a Big Six accounting firm since they have access to CPAs in the various localities and world-wide if we need the help. If we have an employee doing the bulk of his/her work out of Texas, we have to register as a corporation in that locality, withhold appropriate taxes from the employee's income, pay any unemployment taxes peculiar to that locale and the owners would have to file income tax with that locale for the profit that was generated there. It does not help to pay the employee through an account in a Texas bank. The criteria is where is the bulk of the work actually performed. 
• Paying Rent for Home-Office Space. Our CPA advises us that we have no requirement to rent office space for an employee or to rent their home-office. The working conditions are part of the employment agreement which is up to us and can be worked out on an individual basis though we should not follow any discriminatory patterns in allowing some (non-owner) employees to have certain benefits or privileges and others not. 
• Non-U.S. Citizens. If an employee is not a U.S. citizen and works outside the U.S., then what are the considerations? (This is true for any multinational company.) Foreign employees present an even more complex problem: if OBJS employs someone from Jamaica, are there no taxes or FICA? Is OBJS legally or ethically bound to provide health insurance and all the other protections U.S. workers get? These are bridges we will have to cross if we come to them. 
• Export of Technical Data and Software. The U.S. Government restricts the export of encryption technology and some other kinds of software. Any export or import of technology must consider U.S. and other country's customs. This would be an issue in a virtual office that is spread across national borders. 

Insurance 

• Each state. The company needs to file and get workmen's compensation and liability coverage in each state or country where it has a permanent employee. 
• Liability. Meetings now sometimes happen in individuals homes or at remote sites. Full liability insurance (at least in Texas) for the Virtual Office directly held by the company seems not to exist since the state rates office buildings and designs its insurance around that rating. OBJS has obtained computer/peripheral insurance and is protected from being sued (as a company, not individually); however, to cover furniture, books, etc. in the home office employees will have to get riders on their homeowner's insurance, which OBJS should probably pay for. 
• Pre-existing Policies. Several of OBJS's insurance policies were granted while we were still planning to be a centralized company, but the change in direction seems not to have much effect on preexisting policies, although there is the possibility of problems with the individual and group disability coverage. 

Accounting 

• Accounting. Payroll becomes more complicated as we have to keep track of various state unemployment insurance's and state and/or local income taxes. We recently outsourced bookkeeping and added direct deposit to provide (even more) timely payroll. 

Recruiting/Retention 

• Recruiting. As mentioned above, recruiting for a virtual company may be much easier since it does not necessarily involve a move to a central site. This has definitely proved to be true. 
• Employee retention. The virtual office may make it much harder to retain employees. They may see this as an interim move, feel no company loyalty, not feel connected in human terms, see this as playing at working (not real). Continuous productivity and reward, in addition to semi-continuous communication via e-mail, phone, and videoconferencing seem antidotes to this problem. 
• Interim employees. Central businesses hire subcontractors, contract labor, consultants, part-time, students, etc. to fulfill needs for deep expertise, peak loads, and infusion of new ideas. There may be a finer line in a virtual office, where this expertise is distributed. 
• Providing Office Space. Should we provide central office facilities for employees who want that option? That is, do we require the office to remain distributed or migrate to a semi-central, semi-distributed office over time just as central business is experimenting with allowing telecommuting? The best idea seems to be to initially only hire employees who wish to work as a Virtual Office since OBJS is pursuing a fully distributed environment at this point. Later, if the company semi-centralizes, we would consider moving employees if financially able and wanted to move. 
• Moving to the central site. Should the company pay if employees want to move to a central site? Probably not, if that is an employee requested move. This can be decided on on a case by case basis. 

Personnel Policies 

• Virtual Office Policy. We adopted a virtual office policy which we communicate to employees when they join the company letting them know some of the effects of the virtual office. We also give them a copy of this paper to help them know what to expect. 
• Flex Time. Employees are expected to be available for scheduled meetings, to be generally available during working hours, and to indicate when they will return to their work area when absent. Employees can trade off other hours from work day to evening or weekend. Over time, availability is confirmed via weekly progress reports, contributions and via interactions by phone, email, and video conferencing. 
• Timekeeping Accountability. Employees must follow normal timekeeping rules. They need to keep their timesheet current on a daily basis and be prepared if an auditor visits their home office, although an auditor would probably be more likely to contact an employee electronically. 
• Auditability. We may need to have employees sign a release stating they understand that Government auditors may visit and will comply. Similarly, a release to allow work to be done in their home when other employees or customers visit. 
• Floor check. We have planned that the virtual office would be implemented in such a way that if an auditor came to one site, the auditor contact employees at any other site to be shown copies of current records like timesheets. 
• Who pays the rent? The company is not required to pay for the home office if a commercial office is an option or if it is a condition of employment. 
• Who pays for equipment and supplies? Clearly, the company does else the employee is bordering on being an independent contractor by IRS and legal guidelines. 
• Dress Code. This affects when employees meet each other or customers in person but also via home video conferencing. Attention to home office scenery, lighting, and other look and feel standards (e.g., avoid plants directly behind your head) in video conferencing to provide the image of a virtual office of professionals and not a pajama party. Still, this could be emulated, however, as in a Dilbert cartoon showing pajama-clad, unshaven Dilbert videoconferencing with his boss using a hand puppet. 
• Basis for Raises. Remains the same as in a centralized office, namely contribution to the company. Care must be taken in this area to fairly assess accomplishment often. This is not so different in the central office, just that its effects could be magnified via remote working situations. Several safeguards help alleviate this problem: weekly progress reports, constant communication, planning results that require collaboration and monitoring and reviewing progress often. 
• Employee Agreements. What safeguards are in place for insuring that employees do not freelance within areas that are competing with the company. Same as any organization. Just be sure employee knows that the IP they use and develop is company owned and have them agree via an explicit Employee Intellectual Property Agreement, making who owns what a matter of record. This might come up more and more if the virtual office becomes a reality and individuals time share their expertise with many companies. Then we see the Society of Experts model, with a Trader Service that matchmakes needs and providers. We are not there yet. But the half step of a virtual office makes it more important to draw clear legal ownership lines. 

Information Availability 

• Online Information Access. Internet access, the World Wide Web, email, phone, and fax provide much of the access we need to the computing world outside the company. Inside the company, it is easier to share information in an environment where electronic information is easily copied. Our initial workable approach to the problem was to send email and attach documents others want to see. In this approach, each individual maintains a personal organized information space and explicitly pushes information to others who may have requested it. A complementary approach is to also establish and maintain shared information spaces, for instance, a shared file system, newsgroups, and both internal and external web servers. Then individuals can push or pull information from the shared repository. An additional requirement is to keep the information space secure (see below). 
• Hardcopy Access. Books, periodicals, and documents are harder to share in the distributed workplace than in the central workplace. Even in the central workplace, many employees redundantly order desk copies of periodicals and books. The issue then is, how much cost is associated with additional redundant "library" copies or in waiting to gain access via sneaker net or mail. Since information is increasingly available on-line and since we cannot read most of what we receive, the delta is likely to be only a few hundred dollars a year. A solution is to set a limit on purchases via a company policy. A further issue is, who owns the periodicals and books? A company policy makes this explicit. A central repository in a city is reasonable. Heavily used items should be replicated as desk copies. At the same time, this is becoming less of a problem over time as publications increasingly go on-line. This is especially a consideration in our particular profession (computing), since many of the journals and documents that we most rely on are moving fairly rapidly in that direction, and some of the most current documents we need to access are available online before they are available elsewhere. Also, having the coming ability to pay electronically per page downloaded will provide the equivalent of physically Xeroxing the pages on-site.
• Office records. We need to maintain a central and secure repository for at least some key office records. 
• Software Access. The issue we foresaw was how to share software when licenses do or don't comprehend remote work? So far, this has not proved to be an issue, because we are mostly using PCs, PC software is generally pretty inexpensive, and often multi-user licensing isn't available, or is only available in much larger quantities than we need. Also, multi-user licenses are generally per concurrent seat, which best suits infrequently used or quickly-used packages. Most of the software we've purchased is neither. Also, we'd need to maintain a server to host multi-user licenses, which would be an additional expense. As we are introducing Sun workstations into our mix then this situation is changing. Software is more expensive for Unix, and the benefits of multi-user licensing more pronounced. In this experiment, it will probably be more cost effective to maintain a server. Our kind of distributed environment will not be a significant obstacle to maintaining a Sun server (and we have the skills in house to do so). In general, most Unix licensing appears to permit local copies of the shared software; only the license server needs to be centrally located. ISDN is probably too slow to be constantly downloading software from a server, or to be running X-windows over, but it is certainly not to slow to use for checking out licenses, or invoking non-graphical programs on a server (e.g. compilation). So, its possible this won't really cost any more at all, or there might be some additional expense if we really need to use a package with a multi-user licensing scheme that won't work in our environment, but our expectation is that that is not likely to be the case. In fact, we expect that most of the research software we'll be using under Unix, or for software development, will be free. 

Information Security 

• Technical Data and Software. Our contract requires that the Government sign off on technical papers and software before releases. It cannot therefore be always visible outside the organization. Initially, we have made it visible to Government sponsors via printed copies (and we continue to do so since they are required for administrative purposes). But we now have the capability to provide customer secure, password-controlled views into our information space to allow Government customers to review documents on-line and remotely (via the web) before wider release. Many large organizations that use firewalls cannot yet provide this selective access capability to their customers though it is available in commercial-off-the-shelf technology. 
• Malicious attack and safeguarding intellectual property. This is an issue since all documents and code in progress are continuously available to company members and potentially to Internet providers and anyone on the net when they are continuously shared across wide area networks. 
• Security over the Internet. The most secure network would be behind a firewall, but this may be unnecessary. With careful security and an ISDN router OBJS can design a good security scheme. This problem is common to the entire electronic commerce and C4I worlds. They will use encryption and so will we. Security sections of the Internet Tools Survey address this area. 
• Unwanted Physical Access. Having physical access to the machine may result in either deliberate or accidental damage to files or to theft. While the possibility of industrial sabotage is reduced in a home office, the chance that careless family members may unwittingly destroy data is increased; however, all major operating systems come with file locks and password protection. 
• Backups in a distributed environment. Preventative measures to insure against loss of technical data include 
• regular, periodic backup of our shared electronic information file system with archiving off site, and 
• backup to tape or disk cartridge (e.g., Zip drives) for each employee's machine with periodic reminders to employees to backup their work.
• Virus scanning. Periodic or event-driven, continuous virus scanning software is a necessity in an Internet world to actively detect a virus when downloading files. Monthly update to virus definitions is needed since there are now in excess of 10,000 viruses.
• Equipment. Instead of having company owned equipment at a central site, it is distributed. Individuals homes may be less secure than central office space. But there is less chance of a disaster (fire, theft) that could affect the bulk of a site's equipment. Purchasing an outlet/phone jack extender with electric shock protection is a wise investment. Lightning storms can destroy hardware and electrified phone lines are notorious for destroying modems. 

Facilities 

• Office furniture. Employees are given an allowance of a specific amount ($700) to spend on home office furniture. 
• Electronic Office Equipment. Basic issue for each employee is a PC or workstation, printer, fax/copier, and ISDN with video conferencing to come. Color versus speed versus cost are tradeoffs different people make with respect to printers. As a guideline, each employee has an $8K equipment allowance but is issued equipment to do their job, whether it is over or under the allowance.
• Office supplies. Employees purchase their own supplies from a standard issue list with a reasonableness and oversight guideline. 
• Phone lines. Two phone lines are needed per employee, one for phone and one for computer connections. With ISDN, both lines may be used for data streaming when the phone is not in use. For the near term, ISDN is our standard. Are there bandwidth haves and have-nots? The bandwidth disadvantaged may feel they do not have the same opportunity to affect team directions as others. This has not proved to be a problem yet but might be with video-conferencing.
• Travel Agent. The Internet Travel Network is free, fairly comprehensive, always available, and infinitely patient. The drawbacks are that it is less instantly helpful and intuitive than a human agent, specials are not always obvious to the uninitiated, and it requires the user to do the thinking and legwork. Some human travel agents are a phone call away and deliver tickets to your home for no fee. Travel planning is up to individuals though there are guidelines for per diems and expense reporting.

Task Related 

• Means of Communication among team members. If we don't routinely see each other physically in the virtual office, we still must maintain comparably effective communications. There are several obvious surrogate media: email, shared file systems, and the web are mainstays. Email and document exchange works well for communications that require composition and thought or that do not require immediate feedback. Also, these media can be broadcast to many as easily as one. And can be ignored or read later according to the consumers priorities. Voice mail provides a second effective delayed communication technology. Fax provides an immediate means of moving physical documents (if not too long) and overnight delivery services do the same for long documents. FTP and WWW do this online and immediately. Scanners can convert documents from physical to electronic form. The phone is used less than one might suppose since it is perceived as intrusive. The fact that employees are distributed and so metered long distance rates come into effect has not so far been a major consideration. So far, we have not made much use of telephony programs and little use of teleconferencing (multiple participants) though we have discovered it can replace travel and that phone fatigue is a factor for long teleconferencing sessions. Protocols are needed if many people participate in teleconferences, both for technology and interaction. Large groups require more formal interaction than small ones. Videoconferencing is coming soon. 
• Distributed Collaboration. Collaborative tasks in our line of work include brainstorming, planning, scheduling resources and tasks, co-authoring documents, designing software in teams, testing, debugging, delivering results to customers, and field service. Technology support can make several of these tasks easier. But tasks are often chunks of work that take hours to weeks to accomplish and so for many purposes coarse grain support is sufficient and checkout-checkin and versioning protocols often suffice to work well together. We have initial experience with brainstorming, planning, and design in distributed environments, plus we have some experience with distributed software support. Discussions and brainstorming are needed when a lot of possibilities need to be discussed quickly. Even here, it is often best if one participant does initial thinking (makes a proposal or sketches the terrain) then gets directed feedback from others. Topics that wander across the map are inefficient to discuss. Advantages of face-to-face communication include insuring that a group is on track and communicating, providing immediacy of give and take, high bandwidth of "being there", and non-verbal cues. In areas like brainstorming, we have evolved effective means of brainstorming via email and annotations. For extended brainstorming sessions and discussions, NNTP or Web-based conferencing allows long, threaded messages and replies with the ability for anyone who's interested to participate. If the topics are confidential, security technology will be needed. In areas like co-authoring shared documents, we can do this effectively by dividing and conquering compound documents. For documents requiring inspection before release, revisions mode is useful so different reviewers can quickly spot changes made by others. For cumulative progress reports, color-coding progress by week makes seeing a month's progress very rapid. In the support area, with remote login, a support person can do much or all support from the desktop. Hardware problems require on-site access but we can hire this done. In fact, third-party system administration support would probably be facilitated by this arrangement to inspect documents and make decisions. We still need to develop more experience with distributed design and with workflow-oriented problems in a virtual office setting. 
• Maintaining chemistry between team members. This is a major intangible issue, since we all intuitively feel that communication sparked by proximity is at least sometimes very useful if not necessary. Also, there is the danger of isolation in the virtual office if employees feel they are losing contact with others co-workers they never see. Some measures can be used to insure we communicate with each other on a pretty regular basis: shared tasks, asking questions, sending status reports. Everyone being sensitive to this danger is one solution approach that seems to work. The problem is temporarily solved by gathering together occasionally, but that adds cost for remote employees. It is important to avoid making anyone a second class citizens. This has not turned out to be a major problem at year one but we need to continue to monitor this area. We do not know the multi-year prolonged effects of physical isolation on a work team. 

Scaling to 7, 20, 50 employees 

As we grow, we need to periodically assess whether distributed work is as effective as centralized work and whether to centralize some or all of our operations, including whether to provide optional central facilities or centralize some projects or job functions. Certainly, it seems like is should be easier to double in size electronically easier than in physical space, where there may be contracts that overlap, size changes from one year to the next, where lease life-cycles are 3-5 years. 

Scaling in a centralized setting is hard. Shrinking is worse with long leases. Since we are likely to expand and contract as contracts come and go, a distributed workforce with some contract programmers, temporaries, and part timers may be useful. But then the issue of retaining corporate expertise and loyalty becomes an issue. 

Unless the corporate world becomes distributed very quickly, which is increasingly possible technically but a big step for corporate cultures, the company is going to be responsible for molding new employees into the Virtual Office paradigm every time it hires a new one. Of course, this is simply a reality of doing something new and bound to get easier as time goes on. 

Backup Plan 

If the distributed work plan does not work effectively, we can still centralize, all at once or on an as needed basis. Costs would involve: 

• relocating employees or training replacements for those who choose not to move 
• locating a lease property with desirable features (locations, term of lease, etc.). 
• centralizing a once distributed computing facility. 

Mostly, these involve actions we and our employees have so far avoided taking that are part of the normal cost of doing centralized business. The main financial effect would be to increase our overhead. The dominant cost would be in replacing employee expertise. 

Technology Needed 

Without the proper enabling technology, the Virtual Office would not be possible. A Virtual Office needs, to some extent, to assume in each node some of the responsibilities of a centralized office, such as the ability to print or an Internet connection. The Virtual Office, however, should do more than simply replicate a Spartan workplace. 

Basic Technologies needed at each node 

Fundamental technologies required to make a Virtual Office functional are listed below. 

• Computer and Monitor 
• CD-ROM 
• Printer 
• Phone/answering machine 
• Fax/modem - create/send/receive/print fax
• Modem 
• Internet connection 
• Web browser 
• Web search engine 
• Word processing including HTML editor
• Newsgroup reader
• Graphics Editor 
• Spreadsheet 
• E-mail with MIME attachments 
• Virus Scanner 
• Backup 

What is key is that workgroup software interoperates. Everyone does not have to use Microsoft Office v7 but must be able to view and print Word documents (there is a growing preference for HTML formats since they are smaller and based on a standard readable format).

Improvements 

Technologies which improve the Virtual Office and make it competitive are described below. 

• ISDN/cable modems 
• Central file server 
• Internal Newsgroups 
• Fax/copier/scanner
• Security via encryption, passwords, certificates 
• Virtual LAN 
• cellular phone/beeper/toll free numbers 
• Internet telephony 
• Videoconferencing 
• Whiteboard 
• VRML 
• CD-ROM Jukebox 
• ATM 
• ORBs or other distributed object technology 
• ORB services and facilities 
• Java and JDK 
• Web-enabled workflow 
• Web-enabled DBMS 

Of course, this list is just the beginning and does not list programming environments, domain-specific software, administrative software, and many other company-specific software requirements (e.g., project management, PIM, time tracking software to accurately time tasks). For us, the question of software development is still up in the air (though we have had some experience in this area), for example, how to handle revisions control, making developed software work on UNIX and PC platforms, debugging a distributed application, testing (how to tie our machines together to simulate a distributed computing network), and performance analysis (improving our application's performance). 

Experience Report after One Year 

The bulk of this paper was written in December 1995 when OBJS was wrestling with the decision of whether to constitute itself as a virtual office. Since making the decision to do so, we have spent a year living the experience and this section briefly describes our experience to date. 

Overall, the experiment has been a success so far: 

• There have been no show stoppers. In many ways, the virtual office is similar to a central office. The work remains essentially the same as do administrative tasks. 
• The advantages and disadvantages we projected for central and distributed work are close to what we expected. We do save substantial money not renting office space. Our overhead is low. Having no commute is great! We do individually control quality of life for our individual offices. We have experienced only limited problems with office-home interference while avoiding problems with bureaucracy in past employer's central offices (though that may just be because our management style differs and we are a smaller work group). We have a flex-work policy that accommodates reasonable office hours and some alternative time. We have timesheets since we work on Government contracts and need to track hours on specific projects. Far be it from being distracted by home doings, we often work as hard or harder and more productively than we did in past central offices, but clean separation of home life and office life is more the issue. We do find some times a year to get together or get away from the home-office to conferences, though not always everyone meets together at once. Human communication seems to go well and people do not currently report feeling isolated but this is still an area we need to monitor. We need another year to know if we will continue to enjoy working in relative physical isolation. So far though, sharing new ideas seems as fun and exciting as ever. 
• We used the Internet to search for employees, found 40 excellent resumes, identified five candidates most matching our needs, and invited them for job interviews which included a description of the virtual office experience. All were excited by the prospect of the work and also the work environment, and all accepted positions. See OBJS Jobs and OBJS People on our homepage. 
• We set up each employee's office, computing environment, and Internet hookup in each respective home. Most setup and support is an individual's responsibility though one of us acts as an expert consultant when problems arise. Our current office environment includes an external web site as well as a shared, secure internal file system and web server that provides us an organized information space for company shared information. Several employees have ISDN access though some are still ramping up. Most communication is via email with MIME document attachments. We use the phone less than might be expected and internal conference calls are so far rare. To date, we have not invested in video conferencing since 1996 VC technology was not cost-effective, improvements seem just around the corner, and we have not felt a strong need to communicate this way but may find it invaluable later (or may it intrusive)-still we are looking forward to the experiment. Overall, technology for virtual offices is now close to comparable to that for central offices. Both can still be improved and can benefit from improvements to software infrastructure, which is what we are working on. 
• We worked our way through developing policies and procedures that match the virtual office. Mostly, these are extensions of a central office but some are specific to the virtual office, e.g., the furniture allowance. We are working our way through office procedures like distributed timesheets, weekly progress reports, accounting and taxes with employees in several states and time zones, insurance issues, and personnel policies. Policies and procedures and required legal postings are placed on the internal web server bulletin board administration page, convenient for all. All these P&P as well as legal notices we are required to post are available on our shared internal web site's Admin hompage.
• Our work progresses well. We have a lot to show for our first year's work in terms of technical accomplishments and have made good progress on our contracts. The fact that we are a virtual office has not seemed to be so very different than when some of us were centrally housed. We have gathered some useful experiences in how to brainstorm over email (an idea generation phase lasting days, an analysis and consolidation phase). Also, we use various annotation technologies in decision making: email threads to comment on comments; color-coded Microsoft Word or Netscape documents in monthly individual progress report to show weekly progress on individual tasks at a glance; Word revisions mode to inspect and comment on documents from internal reviewers. See Electronic Support for Collaboration & Decision Making in OBJS. Shared documents are stored in the shared file space with access by all internally; restricted access is provided for completed research products awaiting release approval from Government sponsors; and released documents can be made available to the public. 
• While we have not yet installed video-conferencing, we have gained some limited experience with commercial services. We recently completed our first recruiting interview with a remote job candidate using Kinko's commercial service. Quality was reasonable -- 1+ second delay in transmission of picture and sound took a little getting used to, it wasn't always easy to tell who was talking when several were in the room, but pan and zoom controls worked well and lighting was reasonable. The use of video-conferencing for our purpose was deemed a success, somewhat expensive at $600 for 2 hours week-day rates but cheaper far than air travel and hotel. Of course, this was not over the Internet.

But the experiment is not over yet. Several new hires joined us in mid to late 1996 and we will need to report back around a year from now with additional experiences from a larger team with a longer term view. 

Conclusion 

This paper lays factors so far identified in a decision to constitute our company as a virtual office. An effort has been made to compare relative costs of central and distributed office space and to identify barriers to this form of doing business and reduce these risks by conscious preplanning. Nevertheless, it is still an experiment, but one well worth trying initially on a small scale. As with any experiment, it must be monitored-we must learn to identify our mistakes, correct them, and improve. Still, continuous improvement is just one ingredient. So is continuous innovation. After all, nothing ventured, nothing gained. 

Appendices 

For More Information 

There are now millions of people with home-based businesses and millions more that telecommute so our experience is far from unique. But it is still not common for an entire organization to be a virtual office. More and more information and experience reports are available in this area. In this section, we list just some of the web sites and a few of the articles on the topic. 

World Wide Web Sites 

Promoting Telecommuting: An Application of the National Information Infrastructure, September 7, 1994. 
This interesting report was prepared by the Committee on Applications and Technology of the Information Infrastructure Task Force and covers telecommuting, one of eight areas in which NII applications can enhance the quality of life, including society benefits ranging from synergistic benefits for traffic congestion, urban air pollution, national petroleum dependence, increased worker productivity, improved quality of life and benefits to the family, the Government's role in being an exemplar, industry's role, issues of labor unions including monitoring leading the home-workplace leading to invasion of workers' privacy, the benefits for employee morale of increased employee trust, less micromanagement, and more. 

Nomadicity in the NII 
This paper lays out a conceptual framework for nomadic computing. 

AREVO (Advocates for Remote Employment and the Virtual Office) 
An good starting point for information on Virtual Offices, AREVO quickly explains the basics of Remote Employment and the Virtual Office, the benefits of both, and how to promote remote work. This page also links to many other sites which provide more in-depth analysis of telecommuting; however, AREVO and most of its links concentrate on the human aspects of remote work and are light on technical insight. 

Gil Gordon Associates: FAQ on Telecommuting/Telework and Alternative Officing 
This FAQ covers a broad range of issues with brief answers to basic questions, although it is constructed from the viewpoint of a centralized corporation considering telecommuting options. It covers who should telecommute, how to manage telecommuting employees, how to equip telecommuters, getting upper management buy-in, human/social issues, and the Telecommuting'96 Conference. 

SOHO Central
SOHO (Small Office/Home Office) Central is the home page for the Home Office Association of America (HOAA). The site contains advocacy information about remote work and working at home, including legal and tax considerations. HOAA itself is an organization which represents the interests of SOHOs, both by lobbying and offering its members discounts with service providers. 

Jala International, Inc. 
Jack Nilles' coined the terms telework and telecommuting in 1973. The homepages provides a cost-benefit analysis of telework centers and home-based telecommuting, as well as references to many publications. 

Successful Management in the Virtual Office 
This masters thesis by Bernie Kelley and Bruce McGraw researches how management techniques are altered by telecommuting employees. Although the paper is geared towards semi-distributed work models (employees are out only so many days a week) as opposed to fully distributed Virtual Offices, it still makes several interesting points about how one manages across geographic separations. Kelley and McGraw surveyed 28 managers about the qualities of a telecommuter and his manager. They concluded that the key to successful management in a Virtual Office was a paradigm shift from a "participatory style" to one which allowed independent work and was results-based, de-emphasizing procedure. 

OBJS Internet Tools Survey 
The Internet Tools Survey describes a number of emerging technologies enterprises, large and small, will need to be Internet-enabled. It is being completed as part of the OBJS contract Scaling Object Service Architectures to the Internet. 

The National Industrial Information Infrastructure Protocols Consortium (NIIIP) 
NIIIP is a consortium of organizations which, in association with the U.S. government, are developing open software protocols to allow rapid assembly and operation of Virtual Enterprises. We are co-architects of the NIIIP technical architecture. 

Articles from Magazines and Newspapers 

"What to do if your home is your office," Laurent Belsie, Christian Science Monitor, March 28, 1996, p. B1. 
Belsie reviews Alvin Rosenbaum's book The Complete Home Office: Planning your Workspace for Maximum Efficiency, which suggests designing a workspace that is good-looking, comfortable, and productive. The book covers zoning restrictions, parking, getting a separate business line, furniture that meets both office functionality and home decorative requirements. 

"Net revolutionizing the way we work, live," Scott Burns, The Dallas Morning News, April 16, 1996, p. 1D, 4D.
Burns quotes statistics which show that telecommuting has exploded (16.7 million people work at home regularly). He then lists some personal observations about telecommuting. 

"Contract work satisfies choosy corporate nomads," Diana Kunde, The Dallas Morning News, April 24, 1996, p. 1D, 11D.
Kunde interviews contract employees and recruiting experts to explain why contract employees have suddenly become a hot commodity in the computer world. With telecommuting technology, contract employees could Virtual Enterprise with companies across the country or even the world. 

"Is your home office adequately insured?", Grace Weinstein, Investors Business Daily, June 4, 1996. 
At least 12 million Americans operate full-time home-based businesses and millions more moonlight from home. Homeowner's insurance policies vary but commonly on site equipment is covered to $2500, equipment away from home is covered to $250, and the definition of "incidental" businesses that some homeowner's policies do cover varies widely. Furthermore, you may be on your own if a co-worker or client is injured in your home office or you suffer lost income or assets from home emergencies (fire, water damage e.g., from flood or water heater) since homeowner's policies do not cover business-related liability. The Insurance Information Institute offers a free booklet called "Insuring Your Home Business," 110 William St., New York, N.Y. 10038. Commercial policies tend to be overkill; now some insurers like CAN, RLI, IBA write specialty policies for this growing niche and the Insurance Services Office in New York is developing a standard form for home-based businesses. 

"A risky weigh to work." Scott Boeck and Marcy Mullins, USA Today, USA Snapshots, date not recorded, 1996. 
Of Americans who work from home or run home-based businesses, 32% gained weight, 17% lost weight, and 51% remained the same. (C'mon, its right after the Christmas holidays!) 

Virtual Office P&P 

Object Services and Consulting, Inc. has adopted the following company policy governing the virtual office. 

OBJS Policies and Procedures
OBJS as a Virtual Office 

Object Services and Consulting, Inc. is constituted as a virtual office. That means that employees are geographically distributed and generally work from home-based offices. 

History and Rationale. The decision to be a virtual office was made in the first months of the company's history and provisionally approved in our six month review with DARPA and ARL pending continued successful execution of the contract work. Some of the considerations used in making the decision were: 

• Our company expertise involves research on large-scale distributed software architectures that can be used for collaboration-at-a-distance, e.g., by military planners or information analysts that are geographically distributed or by command and control operations. There are also many industrial applications. 
• Our first large contract DAAL01-95-C-0112 is funded under the DARPA Intelligent Collaboration and Visualization Program. Our project's application involves demonstrating virtual office and virtual enterprise scenarios and developing technology to enable remote collaboration. 
• We believed staffing would be easier and this has proven to be the case. In response to our initial position posting, we received over 40 high quality resumes, selected the top five candidates, interviewed them and hired all five! All viewed the virtual office as a substantial plus and OBJS was able to recruit best-of-class experienced employees. 
• Other advantages: no dislocating move, no commute time or air pollution, savings in office space rental, self-managed employees, employee controls working conditions. 
• Overall, we believe that distributed collaboration might be our best chance of competing with much larger corporations in the open systems area. It uses electronics to break the space barrier, keeping us on the leading edge, giving us an interesting competitive advantage. 

It is recognized that virtual offices may not be for everyone (they require experienced and self-motivated professionals) and OBJS has taken care to screen and also advise prospective employees of the implications of this new way of working. 

Implications for Policies and Procedures. The decision to constitute OBJS as a virtual office affects some OBJS Policies and Procedures to some extent: 

• Office space rental - our accounting firm Ernst and Young confirms that we are not required to reimburse employees for rental of home office space and we have chosen not to do so. Employees instead may, if they so choose and qualify, take Federal Income Tax home-office deductions. 
• Furniture Allowance - our Home Office Furniture Benefit (see separate policy) provides new OBJS employees upon employment a one-time taxable compensation of $750 for improvements to their home office. 

It has some smaller operational effects on other OBJS policies and procedures including Time Accounting, Weekly Reporting, Benefits, Books and Subscriptions, Expense Statements, Drug Free Workforce, Drug Free Workplace, Time Bank, Charge Numbers, Basic Office Supplies, and Hardware and Software Purchasing. See each separate policy for details. 

Implications for Back Office. The decision to constitute OBJS as a virtual office affects some OBJS back office procedures. We have developed internal procedures for dealing with these. 

• state taxes for states employees live in 
• I-9 information is sent to us via mail or fax and verified the first time a face-to-face meeting takes place after acceptance of employment 
• liability insurance for the distributed workplace 
• on-time reporting of time and expenses for billing purposes. 

Employee Notification. As with other OBJS policies and procedures, the Virtual Office policies and procedures are communicated to employees before hire and changes to these policies are communicated in a timely way as they occur. 

Glossary 

This glossary is not intended to be comprehensive but is intended to cover several of the main terms people use in describing virtual offices and related concepts. 

alternative work arrangement. Any flex-work arrangement where a worker works part-time or compressed schedules (flex-time) or at a remote site not owned by the company or a site of the worker's choosing (flex-place or flex-space or personal harbor) or job-sharing. 

boundaries (our own term). A logical abstraction of a division or distance metric that separates two things. Often some technology or technologies can be used to bridge the gap. For instance, 

• physical distance may separate communicating participants. Cars, airplanes, telephones and videoconferencing reduce physical distance. For knowledge workers, it may no longer be necessary to live in cities near co-workers and it may be possible to reduce time and costs spent in transit, relocating, commuting and business traveling. 
• a temporal distance in communication may be reduced, e.g., by email. Versions may separate different views of an entity through time. 
• an organization chart may partition the control and information flow of knowledge in an enterprise. 
• differing media, formats and information models may require mediation technology to allow use or conversion of remotely prepared information sources. 
• security perimeters like firewalls partition information. 
• legal boundaries like intellectual property rights and international law similarly partition information or function. It may be cheaper to create some forms of intellectual property in countries with high education and lower costs of living, affecting the global economy. 

There are many more kinds of boundaries. In general, boundaries serve a purpose, to partition a space, but they also create the need to dynamically cross boundaries as seamlessly as possible to reduce costs of boundary transitions. 

business relationship. There are many kinds of business relationships between business entities. An individual or a legal entity like a corporation can participate in many of these in varying roles. Most of these roles imply something about a shared though not necessarily legal contract between the parties. Examples are: customer, supplier, director, owner, partner, officer, employee, stock holder, consultant, program manager, office manager, principal investigator, student, exempt, non-exempt, consortium member, subcontractor, contract, grant, part-time, etc. There are IRS rules guidelines for distinguishing employees from contractors. The Internet provides a rich new way to identify, establish, and maintain business relationships. 

business processes. The patterns, rules, and recipes that govern how a corporation, workgroup, or other enterprise operates, including but not limited to 

• internal communications (organization charts, roles and responsibilities, policies and procedures, reporting, time accounting, human relations, technical work), 
• customer communications (information gathering and market analysis, image creation, closing orders for goods and services), 
• supplier communications (purchasing, accounts receivable and payable, legal, accounting, forecasts), 

Only some business processes are affected by employees being distributed geographically. Many do not change at all. Many benefit from automation, whether workers are centralized or distributed. 

collaboration. Working together to achieve some result. Goals of the DoD DARPA Intelligent Collaboration and Visualization (IC&V) research program include developing infrastructure for rapid assembly of high performance teams and teams of teams to solve problems that arise quickly in large-scale military command and control settings and even more often in "operations other than war," e.g., disaster relief. Often, collaboration is across organization boundaries, involves intelligence gathering, secure communications, sharing diversely encoded information models on a need-to-know basis, continuous planning and scheduling, and coordinated action to fulfill a hierarchy of mission objectives while satisfying a variety of constraints or business rules. 

composition of an organization. A virtual organization or virtual team has a mission, resources, and members. Onto this basic structure is imposed a collection of business relationships, business processes, and boundaries that together are composed to form a description of the detailed instance of a working organization at some point in time. In any organization, evolution occurs and the organization over time can be reconfigured. Static organizations undergo change infrequently; the alternative is to provide for dynamic reconfiguration. Operations that change an organization are called "moves". Some moves are welcome and expected, e.g., giving out raises; others may not be, e.g., drug testing. Moves must maintain company invariant constraints, often called business rules, e.g., don't overspend your department budget. Virtual teams are in some ways simpler than virtual offices in that they are an abstraction that may ignore some of the possible rich set of choices in an organization, e.g., preserving intellectual property boundaries, which they may do because they are wholly within some context where that is a non-issue. That is, they project out (ignore) some of the possible abstractions that an organization might be composed of and invariants that it might try to maintain. 

corporate memory. The cumulative experience and intellectual property of an organization. Corporate memory loss is the decay of that knowledge over time as employees fail to record or maintain known information, the information grows stale and loses currency, or employees leave taking expertise with them. Not all such memory loss is bad; the term semantic garbage collection refers to purposely removing old records and corporations usually have policies for record retention for some kinds of records. A related command and control notion is situation assessment, where a model of a problem is created and maintained, often shared by many people with different kinds of expertise, and used for recording the current state of a situation, alternative courses of action, changes, and continuous planning. 

encapsulation. Encapsulation is the logical hiding of these details that only affect the functioning of the organization and not its results, products, or services. The customer may not need to know or care if a supplier is organized as a virtual office as long as high quality products and services result for competitive cost. 

gratuitous communication. Hallway conversations and socialization, coffee and lunch breaks, all these interactions lead to a feeling of community and may lead to the exchange of good ideas. This form of communication is often missing from the virtual office, except via email levity. 

home office. An office located in a worker's home. May qualify for an IRS tax exemption. 

home workers. A broad swath of people from mothers who work part time to professionals locating full-time offices in their home. 

infrastructure. Technology needed to support the information processing and communications needs of an organization. Until recently, many organizations purchased low level infrastructure (hardware) but created custom software systems in-house. In recent years, more modular software with standard interfaces is providing higher levels of interoperability, making it increasingly possible to interoperate with customers and suppliers in environments other than an employer's proprietary legacy environment. 

Intranet. Refers to a company's private Internet or more loosely to the computing infrastructure within an organization. For many centralized organizations, this usually consists of high speed communication lines, local area nets, workstations, PCs, or dumb terminals, all supporting TCP/IP and encapsulated in a physical security firewall, which allows only some communication protocols like email and also allowing modem-based dial-in password-based connections from employee's homes. Intranets based on Internet technology are much cheaper to maintain than proprietary corporate networks. Firewall-based solutions make it very difficult to permit finer-gained or logical information access in a controlled way, which would permit some customers to "see inside" the organization but with limited access, for instance. Virtual Private Networks and certificate-based security use encryption on top of standard Internet and can be used to accomplish this. The Intranet: A Corporate Revolution is a comprehensive list of the advantages of intranets and links to other Intranet sources. 

location-independence. In this context, location-independence is the property permitting a worker to move about freely from place to place while carrying or having immediate access to his/her personal environment, including work environment. 

middleware. Monolithic "stovepipe" systems that do not interoperate or share information well are increasingly being replaced by more modular, standards-based technology. Middleware refers to a layer of communication-bus software that provides standard communication data models, interchange formats, services, and facilities that together constitute a standard framework for rapid assembly of distributed applications (see Object Management Group). Today, much middleware is domain generic but increasingly standard middleware exists for interchanging domain-specific technical and business data electronically (e.g., Electronic Data Interchange, PDES/STEP, OMG business object facilities, standard interfaces to reduce accounts receivable cycle to improve cash flow in supply chains). 

mobile workers. Workers who travel and work at the same time. Includes field workers (e.g., insurance assessors), business travelers with laptops, and anyone who moves to where the work is (e.g., home repairmen, soldiers in the field). This may imply a need to carry tools and information to the work site and possibly communicate sometimes in an untethered mode but often in a permanent remote site (e.g., via ISDN), either continuously (e.g., using wireless connections for email, for instance, from RadioMail or Datalink International) or discretely (e.g., via dial-in lines or periodic sneakernet), possibly away from conventional infrastructure (e.g., in a jungle) or connected to normal or abnormal infrastructure (e.g., via foreign plug adapters and telephones) or connected only by long-distance to an Internet provider (e.g., using satellite links like Hughes DirecPC Turbo Internet Service (see www.direcpc.com - using a DirecPC Satellite Dish, one gets 400 Kbps on download) unless you are an information provider when uplink bandwidth is too low), and often in band-width constrained circumstances where quality of service varies widely (<1Kbs, 2.8Kbps, 28.8Kbps, 128Kbps, higher). Infrastructure required includes wireless modems (spread spectrum (2 Mbps) or infrared modems (3 Mbps)), routers to modem banks, phone jacks, adapter plugs, telephone, ISDN, or high speed lines, phone, fax, videoconferencing, white boarding, and palmtops or laptops. 

mobile office. Tools, technologies, and workspace that a mobile worker needs to get his/her job done. 

MUD. In on-line game-playing, Multi-User Dungeons provide complex reconfigurable environments, roles, and habitats. 

nomadicity. As defined by the Cross Industry Working Group (XIAT), nomadicity is the "the ability of people to move easily from place to place, retaining access to a rich set of services while they're moving, at intermediate stops, and at their destination. A person is a nomad vis-á-vis the NII if she moves as little as from one desk to an adjoining one or as far away as across the continent." 

personal environment. As defined by the Cross Industry Working Group (XIAT): "The many parts of our lives -- our family life, business life, personal social life, business social life, etc. - each frequently involves a different physical location. People move between these places as they move between the different aspects of their lives. Doing the appropriate thing in the appropriate place is key to the way many people organize their lives." 

remote employment, remote work, telework. Any working arrangement where the employee performs some significant portion of his/her work from some work site other than the employer's central office-typically from the employee's home, thereby substituting information technologies for commute time. Remote workers include a wide variety of knowledge workers, e.g., accountants, architects, attorneys, bookkeepers, claims adjusters, computer programmers, engineers, estimators, graphics artists, journalists, technical researchers, technical writers, telemarketers, transcriptionists, and many more. Depending on the conditions of work, the business relationship between the company and worker may be employer-employee or employer-contractor. Contractors generally provide their own equipment, communication lines, have different performance evaluations, and set their own hours. 

society of experts model. A step beyond a virtual office where individuals provide expertise to many organizations and the line is blurred between employee and contractor. Certainly, this model requires distributed work. Agreements on compensation, statement of work, ownership of intellectual property all must be made explicit in this model. 

tele-. At a distance. Examples: Television, telecommunications, and more recently telemedicine, telelibraries, … 

telecommuting. Communicating, collaborating, and working while geographically separated from the central office via electronic devices such as faxes, videoconferencing, or modems. The term was coined over twenty years ago. Telecommuting is more relevant to periodic remote employment, where the employee spends time in an central office and occasionally "telecommutes" there, than Virtual Office. See Gil Gordon Associates, a center for telecommuting information. 

telepresence. Just like being there. Several technologies enable varying degrees of telepresence, from mail to phones, video conferencing, and virtual reality. Smellivision is still a research area as are Star Trek holodecks. 

telework centers. When employees work remotely from a central site but not at their home, instead at a satellite center. The center may be owned by one employer or space rented to permanent, migrant, or occasional workers (called hoteling). Executive office suites are a variant where businesses share secretaries, meeting rooms, etc. Incubators are variants of these where new business are provided additional services including business advice when getting off the ground. Airlines provide a variant for nomadic workers via a membership fee at airports (e.g., Admiral's Club). 

videoconferencing. When individuals or groups meet together at the same time but at different sites and can see and hear each other. Two way, broadcast (as in M-Bone), and multi-way conferences are supported but for differeing technologies or costs. Videoconferening is related to audioconferencing which only involves groups of people hearing each other and which is far more pervasive today. See section of Internet Tool Survey on video-conferencing.

virtual. A logical abstract surrogate or simulated function. Opposite of real or physical. Examples include virtual memory, virtual reality, virtual environment for training, virtual classroom, virtual laboratory, virtual space, virtual clipping service, virtual meeting room, virtual pet office building, virtual tour of Ireland. In the case of a virtual organization, it almost always involves physical distance between employees or operating teams, but sometimes it involves security boundaries (e.g., firewalls) or legal boundaries (e.g., ownership of intellectual property co-produced by member organizations in a virtual enterprise). Some virtual organizations are aggregate composites of a number of real organizations. 

virtual office, virtual company, virtual corporation. (1) A permanent corporation or partnership containing a significant number of remote workers. The Virtual Office fulfills all of the roles of the traditional, centralized office (e.g., it has corporate officers, owns intellectual property, has employees, pays taxes) although the employees work at home offices and collaborate for the most part electronically with occasional to no physical contact with other employees. See Advocates for Remote Employment and the Virtual Office (AREVO).Virtual offices are typically corporations (legal, logical entities) and corporations typically are not defined with respect to geographic locality of employees. Virtual offices are a matter of degree since even in conventional offices, many business relationships are necessarily maintained across distributed environments, for instance, customers and suppliers are located at different sites, project co-workers are often located in different divisions, and the CEO's speech may be via videotape. In both traditional and virtual office's the organization mission remains the same, but some business procedures change in the latter to accommodate collaboration at a distance. (2) A dynamic, interactive VRML model of an office where drawers of filing cabinets pull out, the calendar or clock can be viewed, and the phone rings. An example. 

virtual enterprise. An assembly of best-of-class geographically distributed individuals and/or organizations assembled an "enterprise" for the purpose of solving a specific problem or creating a product. The Virtual Enterprise may disassemble after completing its mission (but often does not). See NIIIP definition of Virtual Enterprise. Anti-trust rules apply to limit unfair competition and some virtual enterprises limit technology they produce to pre-competitive technology or reference implementations leading to standards, available to all member organizations. Generally, technology needed by a virtual enterprise is similar to that needed by a large central enterprise except in some respects: 

• virtual enterprises are often virtual offices (distributed environments) 
• they face the added requirement of controlling the sharing of intellectual property across organizational boundaries 
• they may be ephemeral 

virtual organization. Includes both virtual offices and virtual enterprises. This term could also be applied to standard bodies, consortia, research projects like those at MCC, contractual teaming arrangements (but usually is not). 

virtual team. By extension from the term virtual office, a virtual team is one that is constructed for some mission where members are geographically distributed. Both virtual office and virtual team are similar in that they both involve distributed membership, will need similar infrastructure, and will involve planning and executing a mission in a resource-constrained environment. Where a virtual office is relatively permanent with a long-term mission, relatively stable membership, relatively stable resources, and a shared context (corporate culture) that may take months or years to put in place, a virtual team may add of lose members fairly dynamically, the shared context must be transmitted as quickly as possible, and the situation is subject to rapid change. This might at first seem like it requires breaking down communication barriers (boundaries) and putting in place a collaborative environment quickly, but it more likely means having a fairly well understood environment that is modular in structure, rapidly configurable to meet widely differing needs, and evolvable as situations and priorities change and available technologies become available. 

This research is sponsored by the Defense Advanced Research Projects Agency and managed by the U.S. Army Research Laboratory under contract DAAL01-95-C-0112. The views and conclusions contained in this document are those of the authors and should not be interpreted as necessarily representing the official policies, either expressed or implied of the Defense Advanced Research Projects Agency, U.S. Army Research Laboratory, or the United States Government. 

© Copyright 1996, 1997 Object Services and Consulting, Inc. Permission is granted to copy this document provided this copyright statement is retained in all copies. Disclaimer: OBJS does not warrant the accuracy or completeness of the information in this survey. 

This page was written by Craig Thompson in December 1995 with updates supplied by summer student Shaun Joseph in June 1996 and some additions made by Thompson to reflect our experience after one year of operation in January 1997. Several people at OBJS contributed comments. Send questions and comments about it to: Craig Thompson.

Last updated: 2/10/97 sjf

Back to Internet Tool Survey -- Back to OBJS