BigStep.Com Match.Com Libida.Com Microsoft WebTV 3DO U.S. West

April 1995 - February 1996 3DO's multiplayer game console was licensed by Scientific Atlanta and used as the platform for the Digital Set-top box used in the U.S.West Broadband Network Video-On-Demand Trial in Omaha, Nebraska. The operating system for the box was a derivative of the 3DO Opera OS, and an engineering department (called the "cable group") had been formed within 3DO to modify this OS (called NetOS) according to the demands of USWest in collaboration with Scientific Atlanta's needs for their Level-1 system, and Digital Equipment Corporations' standards and interfaces for their digital video servers being developed in a division in Shrewsbury, MA. We were asked to take over the engineering management for this project after 2 previous engineering managers had quit, and the immediate boss wanted no more to do with it. We felt somewhat "set-up for failure", but it seemed like a challenge. The biggest issue was that the integration and program management being done by USWest were extremely poor and a significant factor for the repeated failures by vendors to deliver releases that worked per spec and that failed to arrive on time. Additionally, the morale amongst the engineers, and the working relationships between all 4 companies continued to deteriorate while each company was bound by contract to deliver. With good support from management, we continued to manage the delivery of the next releases, and worked hard to manage expectations externally while dealing with the morale and technical and logistical issues internally. By steadily replacing broken processes with processes that worked, small step by small step and after overcoming a lot of pushback on all fronts (including the upper management), we were able to gain the confidence of the remainder of the engineering team and upper management internally and key members of the ranks of the other 3 companies to be able to influence the way in which the overall program and the integration in particular was being managed. This was a significant factor in the turnaround of the trial, and in the successful delivery of MT5. In MT5, all of the agreed features from the vendors were working, and a number of key latent bugs that poor integration management had left lingering around for months had been fixed by virtue of the now close working collaboration of the various teams who no longer seemed to "hate one another"! Worthy of mention was the continued delivery of NetOS releases despite a cut in engineering staff to less than one-half the original number and a corresponding loss of skill sets and working familiarity with the code. This was a result of 3DO effectively pulling the plug on the project, but not being able to do so formally as they were under contract to deliver. One of the key members of the team who quit before we took over the management left us with a buggy MPEG driver, and under extreme pressure from USWest, we were able to facilitate the complete re-write of that driver in 2.5 months while fully shielding my engineers from that pressure. Thanks to the influence we exerted outside my domain of responsibility on the program as a whole, USWest executive management treated us as management advisory for their program and we were able to catalyse changes in process that were recognized and acclaimed by all of the vendors, who fully supported my efforts. We also conducted a source code escrow alone with great success, undoing the damage inadvertantly done on the prior attempt by I&R.