Fall 1995 - LifeUSA

After escaping Ameridata, I was without work for about a week before I got a tip about a gig: One of the tools I had used in my work was something called MicroFocus Cobol DIALOG SYSTEM. This was pitched as the power of COBOL (which truth be told, was actually very good at doing straight up transaction processing) with a graphic user interface. The way it worked was that you wrote your Cobol programs in text on the back end, and it worked just like any old mainframe. But instead of building a screen interface, you put in a single section to "call out" to a paired Dialog program: You built a graphic user interface in a different tool, and you had a single two-way pipe between your screen and your program logic. It was kludgey and you were definitely limited in how the program would operate - you collected the data on the screen, hit a button, and BAM it sent it to the back end where the real work happened. Then it passed the results back up the pipe, and you had your results on screen.

In other languages like VB, Powerbuilder, Delphi, etc, the logic was at the same layer as the screen code... but those tools were only just coming to the fore in the mid 1990s. For an number of years, Cobol and Dialog System was a very slick solution for the big corporate IT shop that was looking to leverage existing talent (years of Cobol programming), but modernize the user experience.

Having this on the resume was like having Java on your resume in 2000. It was HOT.

So I was out of work for a week, and then it was off to a project at a life insurance company LifeUSA - back to contracting. This was a local company, with a strong "ownership" culture - the employees were believers, the bosses were very out and about, and stock options were given freely. By the time it was finally sold in the late 1990s, there were more than a few millionaires made... across the organization. It was like Enron, only the business wasn't a sham and people actually made money.

I learned all of this in the course of the year I worked there, from Fall 1995 to Fall 1996 (when I was recruited to join ValueRX). Of course to start, I was not brought anywhere NEAR the company: Somehow I was put into a room with 3 other programmers, a classroom with tiered desks, a huge whiteboard, and four computers. In a low profile building in an office park 2 miles from the main campus. Our team included Eric, the chain smoker who only knew the back end Cobol parts, Steve, the young(er than me) geek who only knew the Dialog side, myself, who could bridge the gap, and Bob W.

Bob W (I don't often withhold names, but this guy...) was a programmer who I had run across at Ameridata: He had been working on Cobol reports for months and months, and suddenly had to leave due to a family emergency. I picked up his work, and discovered that he hadn't actually done much work, but that the work he HAD done was completely useless... and I rewrote them from scratch in a couple of weeks. My DAD had actually run across this guy's name in code HE had rewritten as well - the guy was a bad penny: If you saw his name on a program you were working on, you knew you were going to have to reach in, rip out the guts, and start from scratch.

So I'm on a new project, and somehow in the middle of this strange scene, I'm in an office with this guy again, who was all chummy chummy. He was supposed to be "the Architect" of the project, laying out the program logic that we were supposed to make happen. As it happened, he was around for maybe 2 hours a day, and out making phone calls the rest of the time. So Eric, Steve, and I whiteboarded the whole thing and got to work. After a few weeks, Bob stopped showing up entirely... and nobody asked us about it. But we suspected he was "reporting his progress" to the boss who was 2 miles away and never stopped by.

The project was a system to track Insurance Agent commissions against the draws they had taken. The data was complex, and we needed to calculate the value of the policy over time (commissions were paid at certain maturation times). Eric's coding was very tricky, and I actually learned some good tricks in the way he did things. We got along well even though his constant nicotine cloud was thicker than the haze inside of First Avenue. Steve, on the other hand, was a programmer who liked to make things complicated. He wrote elaborate error checking routines into the graphic front end that were unnecessary (validation could be done much faster and cleaner on the back end), and by the end of it, the screens were huge unwieldy affairs that pushed the very limits of the computers it was running on.

We had split our efforts because I had a hard time working with him: We were both young and convinced we were right - so he worked on one big screen, and I broke my half of the program into several smaller programs that handed control across as needed. Many small, simple programs that do their job well, versus one huge monster that crashed frequently. To be fair, his side DID work - and it behaved ALMOST like a Powerbuilder application would. But I would have done it completely differently.

We handed off version 1 to the client, who liked half of the program.... and wanted another project done. They gave me the whole project this time, and Eric and Steve were stuck working on Version 1.1 of the monster. Every so often I'd stop by to see Steve's handiwork. He was always defensive and I quickly learned not to comment on it in ANY WAY. We saw Bob W from time to time, and were never sure just what project he was working on.

Now, 2 months into the project, we were moved into the mothership: Goodbye remote facility with no privacy, hello cubicle. And it was here that I learned about the culture. For the most part, it was a pretty positive experience, especially after the slightly depressing start in corporate America at Ameridata and the Telephone Billing place. They had spirit days. People brought food to share. People went out to lunch together. It was at one of these spirit events that I witnessed the horror of one of the VPs cooking up RAW bratwurst as though they were pre-cooked. 3 minutes a side, soft and pink on the inside. I ate nothing but potato chips that day.

I was sitting in my cube one day in January 1996 when the Director of IS came by and asked "Have you met Mark? He likes music - you should go talk to him". So I wandered over almost immediately, met Mark Loesch, and we became immediate friends. A couple of weeks later, he invited me to his birthday party... but not being a particularly good listener, I came a week early. And we sat with a bottle of wine and knew that we were of the same tribe. The relationship with Mark was what brought me to ValueRx in the fall of 1996. On my other blog are more stories about Mark, who I miss every day since he died in 2007.

In the meantime, Mark and I were regular lunchmates, heading off the the "Cafetorium", where the neon sign promising "Fresh, Tender, Juicy Wieners" brought a laugh daily. There were other characters there as well: Mark's sister in law Molly was just beginning her programming career there. Our friendship there was wonderful, but by early summer, Mark had taken off to a new, mysterious gig, and I was stuck with the Cobol Guys.

Over on the "old Cobol Guy" side of the fence, a guy named Gene was never seen to be doing any actual work, but was always too busy to answer a question... and I overheard enough heated discussions about "Beany and Cecil" that I'm sure that once he left, he probably uploaded himself into the alt.tv.beany-and-cecil newsgroup and haunts it to this day. There was Tom, who resembled a pekinese, and wore clothes two sizes too tight - and vintage from the 1970s, which is when they were first bought. Another programmer, whose name is lost to time, had Solitaire up on his computer EVERY SINGLE TIME I WALKED BY.

I met more "Business Analysts" and was still none the wiser on exactly what their role in the organization was supposed to be. As I worked, I developed a work style to cope: Meet with the business analyst. Go to the client, review what the business analyst had told me, get the REAL requirements, and build according to the client's specs. This worked pretty well, until one of the analysts figured out I wasn't doing what he said. He tried to raise the issue with the client, but she was happy with my work. So he went to the project manager, who took him off the project, and put a different analyst in. Even today I have a suspicion of the pure "business analyst". They rarely understand what the client wants, nor what the developer can do. The best BAs I've worked with have a realistic view of their role. The worst actually believe they add value.

In addition to Bob W, there was one other "prize" coder at work at LifeUSA. One of the guys from the Telephone Billing gig was doing mainframe programming... and I was called in to finish his work on something when he was out for a week. I already told part of this story in another post, so you know that I fixed his 2 months of work in a few days.... and when he returned he wasn't very happy with me. But for a time, given my shared history, this guy, Eric the smoker, and I were friendly. It was in that time that Eric dropped perhaps the most crass line I've ever heard (outside of some of Phil's gems) - In talking about doing Y2K work for $70/hr, Eric paused and said "Man, for that kind of money, I'd blow dogs".

By the end of my year there, things were looking good: I had a good reputation, people wanted me on their projects... There was some tension with the Cobol Corps, but I didn't pay it much mind. In the early Fall, the IS Manager asked me into his office for a meeting.

He wanted me to consider dropping the contract, and coming on as an employee. He made the pitch for the culture, the people, the opportunities... but in the same breath he stressed something else which made me shudder: He wanted me to know that this was a slow moving place, and that I could expect a long, stable career there, but I'd need to slow my work down a bit, because I'm shaking up the team and people feel threatened by my work pace. We're a TEAM here, not a superstar shop, but it's a good family. So slow down, come on board, and I can have stability and an easy life.

I thanked him profusely for the offer, and within a week was interviewing with ValueRx.