I was the last hire-on from a major staff-up to service their biggest (and only) contract. The trouble was, they didn't really have a contract. They had a letter of authorization that got renewed on a monthly basis. When their customer lost interest, they had about two weeks warning before all the work went *poof*. Things were a little confused when my start date arrived.
Still, the company gamely soldiered on. We took over a product called Unitree. Once, long ago, Unitree was one of the original hierarchical storage systems. Killer stuff out of Lawrence-Livermore. Fifteen years of hack jobs by various vendors had not helped it to age well. Nonetheless, we prepared to shoulder the burden and perhaps roll out a new version. Until the corporate bosses learned that a major development effort cost money - while products don't pay off until they're finished. After that, we just did maintenance.
From there, we went on to something much bigger. Namely Canada. As contractors (excuse me, "technology partners"), we flew back and forth to Toronto. Some of us every week for months. We were laboring to save the Apollo project which was being done by Prism (a subsidiary of Nortel). It was an interesting sounding project: they were building a monster system to control their entire telephone market. The trouble was, it had around a hundred people, no clear direction, warring managers, and basically no idea what it was doing. If it weren't for the fact that Prism was ISO 9000 certified, I would have thought they were hopelessly disorganized. In the end, despite a lot of hard work by a lot of people, the project was a disastrous failure. Or was it? Titan made a lot of money. The Nortel employees all found internal transfers. Nortel wrote the whole thing off as an R&D expense. So what if nothing was produced and millions of dollars were spent? Everyone seemed happy (except for the jet lag).
In Spectrum's last gasp, we started working on a product of our own. The project code name was "Merlin" which narrowly beat out the other two contenders: "Rainbow" (last time we let Jenelle into a status meeting) and "Dingo" (because it comes in the night and carries your babies away). It was supposed to be a remote caching system for ultra-high bandwidth cable modems. Or maybe it was a pretty front end system for @Home or someone. We never did get a clear answer. While we were busy trying to find one, the company decided to shut Spectrum down. The engineers (all except for Phil) got transferred to the sister division across town. Everyone else got the axe. Oddly, it was the folks who got the severance package who were happy and the ones who kept their jobs who got depressed.
While at Client/Server, we helped finished up a program called CARE for GTE. If you get your phone service from them, and you phone them up about something, they're probably typing into our program. Of course, it's not really ours since the program was finished before I arrived. I got to help maintain it - which in this case largely involved dealing with the hassles that come from a poor definition of what the program was supposed to do in the first place. Let's hope my corporate masters learned from this little lesson before they get to do more free work on the next contract.
In the end, C/ST found itself with no more work to do than Spectrum had had. They started sniffing around after Year 2000 work for some reason. (Given only one guy knew COBOL, we're not sure how they planned to actually pull it off if they got a contract.) When that didn't pan out, they put us on mandatory vacation ("Come back when we call you or you run out of vacation time. Starting now."). Most of us didn't go back.
The DarkMage <barkley@cirr.com>Last modified: Tue Apr 7 19:18:12 CDT 1998