In the spring of 1990 the City of Atlanta held its collective breath waiting for the announcement as to who would be hosting the 1996 Summer Olympic Games. Then, suddenly, Atlanta was it, Host To The World: 30,000 athletes and their delegations, plus two million spectators, press and others. Over the next few months the shock of our success set in while planning for the sports venues and the Olympic Village got underway.
The Georgia Institute of Technology was involved in Atlanta's bid for the Games from the beginning. Researchers from Georgia Tech's Interactive Media Technology Center produced the computer graphic simulation of Atlanta and the proposed Olympic Village that's widely credited as the winning card for Atlanta. The emphasis on technology has continued, turning the Village/Venue planning from a purely property management project into a technology management project. Our goal all along at Georgia Tech has been to maximize the synergy between our needs at Tech and the technological infrastructure of the Olympic Games which could be left behind as a legacy following their conclusion. The coming of the Olympics has become for Georgia Tech a campus-wide technology upgrade project with a particularly visible deadline motivator: the '96 Summer Olympic Games.
The Mission
In a sense, the scope and timing of the Olympics' preparations have required the professional management and organization usually devoted to disaster preparation and recovery exercises. Much of the Olympics' management staff consists of loaned executives from industry sponsors, especially on the technology side. Without these experienced professionals, organizing and conducting the Games would be an impossibility.
To capitalize on several possible technology projects initiated by the Atlanta Committee for the Olympic Games, Georgia Tech started a companion project larger in time frame than the Olympic Games but motivated in schedule by the '96 deadline. The project, which we christened FutureNet, quickly gained visibility within the state and garnered financial and political support from the Chancellor, the Governor and the Legislature, as well as industry. We've been working closely with several of the official sponsors of the '96 Games in an effort to maximize the benefit to Tech where reasonable. This collaboration has provided a wonderful win-win-win for all parties involved. The complexity of contracting arrangements has increased somewhat for everyone, but with few exceptions the working relationship in the "trenches" has been an excellent example of technology experts cooperating toward a common goal in the face of time, financial and manpower constraints.
The Field of Play
The Olympic Village will house more than 30,000 people. Georgia Tech residence halls, fraternities and sororities accommodate fewer than 6,000 students today. To develop this housing infrastructure required building a lot of new dormitories on the Tech campus, plus a considerable remodeling effort for the fraternities and sororities.
The Village will be wired to receive video signals from each sports venue for viewing by athletes and Atlanta Committee personnel. Although the video signal isn't required in every residence hall room, a video feed is specified to a significant number of campus buildings where Atlanta Committee personnel will be stationed during the games. A hybrid fiber/coax video plant similar to what Tech already has installed in some residence halls is being provided by Scientific Atlanta. Some 60 channels will be carried over OC-3 SONET links back to the Tech campus where they will be tied into the local video system. Athletes and others will be able to watch the Olympics from common rooms in the athlete dormitories. In addition, a separate telephone infrastructure will support the Village. There will be an on-site PBX feeding via fiber optics to multiplexers in buildings where significant phone services are needed, and Centrex services where requirements are fewer.
A computer network is planned for the Village to furnish timing and scoring information as well as Village local information. This computer network will run as a 16mbps or 100mbps token ring over a fiber optic network. Because IBM is the official Olympic Sponsor for computing, PS/2 computers running OS/2 are planned for both the Village information stations and the larger mini- and mainframe computers needed for more complex administrative functions. The press will have access to the computers to get scoring results in near-real-time and biographical and historical information from online databases.
All of the above services will run over a large fiber optic network carrying the video, voice and data signals between the central computer room and the rest of the campus, as well as copper wiring to connect the fiber backbone to the individual stations.
Win Some, Lose Some
There have been several instances where the Olympics' technology needs conveniently matched Georgia Tech's technology requirements. In these cases, we could offer permanent installation of the technology, relieving the Olympics' technology group of responsibility for removal and restoration costs. The cost savings could then be funneled into the up-front installation costs and the post-Games management costs. The fiber optic network is one example that has proven to be an exceptionally fortuitous acquisition for everyone involved.
The video network is another instance in which cooperation and requirements coordination has paid off. The existing video network on the campus covered at least part of the need for services in the residence halls. The existing plant needed some upgrading and new services needed to be installed, but the Olympic requirements could be augmented to meet the Georgia Tech requirements and left in place after the installation for significantly less than the cost of installing a temporary video network from scratch.
On the other hand, the Olympic need for copper wiring inside of buildings is so sparse (one TV per residence hall floor) or so exceptionally dense (200 computer drops for press access in one large room) that no benefit could be identified in a permanent install. This was quite disappointing since we had originally anticipated that Georgia Tech's needs would match Olympic requirements, just as they had with fiber. Similarly, the phone switch targeted for use by the Olympics was expected to be usable by Georgia Tech after the games. But due to several different issues, the switch isn't expected to remain after the Games.
We also found we had to compromise when it came to network connectivity among the fraternity and sorority houses. The Atlanta Committee is using these residences to house National Olympic Committee members and plans to have data, voice and video services available here also. Tech doesn't currently provide any of these services to the fraternities or sororities and so doesn't own the infrastructure. We are investigating providing this, looking at costs and legal issues. The major problem today is lack of conduit space into these houses. The Olympics' technology management has decided to use central office and temporary facilities here and because of time constraints we really haven't been able to find a better way to retain the technology infrastructure after the Games.
The '96 Games have been touted as a high-tech Olympics and rightly so. But this statement prompted some unrealistic expectations in our minds early on. ATM (asynchronous transfer mode) was still in the pre-standards stage when the majority of the technology standards were set for the Games and so could not be considered as a possible technology for mission-critical and one-time-only functions.
This is one place where the technologies of FutureNet and the Olympics have diverged. FutureNet has chosen to use this technology in further stages of deployment after the Games. Also, FutureNet has become a testbed for new equipment in multi-vendor trials of high-speed backbone technologies, digital video services and switched distribution of network connections into classrooms and offices. Georgia Tech will not be carrying mission-critical traffic on unproven technology, but as the technologies demonstrate their robustness, we plan on deploying them as appropriate.
Just Sign on the Dotted Line...
One lesson that took me a while to learn is that having two parties in a contract is fine but three's a crowd. In our attempt to contract with Olympic sponsors for both fiber optic cabling and installation, we realized that we had to contract with each vendor separately and with the Atlanta Committee separately as none of the parties wanted to take responsibility for any other. In retrospect, this makes sense, but early on, it seemed somewhat excessive. Also, preparing each contract for modification during the execution proved to be an absolute necessity. Change planning has become a way of life.
And for me, the biggest lesson learned has been that no contract will be executed until six months after the absolutely last drop dead date that you needed it. I have great respect for lawyers who work on contracts, but I'm glad I'm not one. Issues that everyone agrees to around a table informally become very complex when put on paper, and without excruciating attention to detail, don't make good contract content. Even worse, if the people originally involved get reassigned during the working period, new parties may not have the same basic ideals, creating conflict that can only be solved by reviewing the contract again.
Nevertheless, we successfully negotiated contracts with AT&T for fiber optic cabling, BellSouth for fiber installation and Scientific Atlanta for the video plant. We also put together a contract with the Atlanta Committee for shared use of the campus and much of its technology.
What Do We Do With the Students?
One of the stickiest issues that we've had to deal with is how to handle the campus functions that are to be displaced for the six weeks of the Olympic occupancy. The summer class schedule was the most controversial since it meant students would have to be housed off-campus with faculty driving to a remote campus to teach. Another related issue has been the possibility of a telecommuting trial for faculty and staff. Each of these issues could be the subject of an article in itself so I won't go into all the details, but the outcome is that Georgia Tech will have a late summer session and push the start of the fall quarter forward a few weeks. A telecommuting trial is still in the works and we are trying to ramp up our technology infrastructure to allow faculty and staff to network from their homes at least some of the time.
The Future of FutureNet
Georgia Tech is considered a powerful force in engineering education in the region, the country and even globally. Our plan for FutureNet has been to create an infrastructure that could act as a model for an infrastructure that would support the State of Georgia in about five years' time while supporting instruction, research and administration at Georgia Tech now. This infrastructure would provide puzzle pieces that could be connected together providing flexible video and data connectivity for collaboration, education, conferencing and research.
With this in mind, we are beginning to experiment with remote educational concepts and ubiquitous, flexible connectivity scalable to a region-wide area. We are planning to pilot content distribution to a classroom, from a classroom, to and from residence halls, to remote locations within a few hundred meters or a few thousand miles of the campus. We have begun brainstorming with others such as Zoo Atlanta, Fernbank Science Center and Trickum Middle School to create pilot projects that will demonstrate new and innovative ways to use technology to augment education.
Debriefing
I guess the old adage is true: be careful what you ask for, because you might get it. When we began negotiations with the Atlanta Committee and some of the Olympic technology sponsors more than a year ago we didn't know exactly what success would mean or how much visibility the project might attract. We have felt like a startup company whose sales have outstripped the existing resources but whose ability to grow is limited by many factors. We have exceeded our more realistic expectations in most areas and have been disappointed in only a few. We have gained valuable experience in large project management (and built an in-house project management team) and have learned valuable lessons in quickly growing an infrastructure to support different technologies on a one-square-mile campus and beyond.
Also, it has been interesting to witness the clustering of a critical mass of interested parties around a project. This critical mass has formed in spite of problems that we have had in getting information out to industry and other parties who might be interested in collaborating. We've also found that getting clear and complete information to the campus community is very important even though it's sometimes difficult. Keeping our customers happy and prepared for the problems that inevitably happen during times of rapid change is one of the most critical issues that we have faced.
And last, another old saw: work will always expand to fill all available resources plus ten percent more, no matter how many resources you get.
Ron Hutchins is director of engineering in information technology at the Georgia Institute of Technology. [email protected].