Jan. 19th, 2017

randy_byers: (Default)
DARS Gang.jpg
With Tom and Kathy at a Washington State DARS conference in Ellensburg probably in the late '90s (Photo by the fourth member of our team, Susan)

Yesterday I cleaned out my desk in Schmitz Hall, and amongst other things I discovered this photo from a long ago work conference. December 31st was my last day as a University of Washington employee. I'm now officially retired, and I've specifically applied for a disability retirement, although that hasn't been approved yet. This, however, is the story of how an English major ended up working in the Academic Data Management Office. Not that my trajectory is all that unusual for the early days of the Information Revolution.

My first job in the Office of the Registrar, which is currently located in Schmitz Hall, was a temp clerical job in the Graduations Office in 1988. The supervisor there, Virjean, liked my work well enough that when a credentials evaluator position opened up in the office in February 1989, she hired me. The cred evals processed graduation applications, which meant we determined whether students had completed their degree requirements and could be granted a degree. So in four years in this position, I became thoroughly familiar with the University's undergraduate degree requirements.

A few years after that, around '91 or '92, the U bought a license to the Degree Audit Reporting System (DARS), which was a software package that allowed schools to encode their degree requirements, feed a student's classes into the system, and let the program determine whether the degree requirements had been fulfilled. Since I was familiar with the degree requirements and was considered a pretty smart guy, in 1993 I was given the job of implementing DARS from the degree rules side. Susan (who took the photo above) was the COBOL programmer in charge of installing DARS on the mainframe and figuring out how to feed the requirement "encoding" and student records into the system.

The mainframe version of DARS came with screens for entering the requirement encoding, but the mainframe team was short-handed in those days. They didn't have the bandwidth to implement the entry screens. The first stage work-around was to have me manually create text files in which each line and each position within each line was mapped to the DARS data structure. Needless to say, this wasn't a very user-friendly solution. So they decided to have me develop an Access database with forms that allowed me to enter the data in a more intuitive way and then export it into a flat file like the text files I'd created earlier. I don't think I had any knowledge of Access at the time, or if I did it was just a couple of entry-level training courses that introduce you to the concepts of tables and queries, forms and reports. I'm not sure why they thought I'd be able to figure Access out on my own, other than they thought that I was a smart guy with good analytical abilities.

So I spent six months learning how to use Access, including how to write procedures in Visual Basic. This was definitely one of the strangest periods in my working life, because I was essentially being paid to learn. I spent all day, every day, reading Access manuals, trying to figure out how to do what I needed to do. Eventually I developed a database with data entry forms that allowed me and others to encode the degree requirements for DARS in a relational database and export them into flat files for upload to the mainframe every night. If there was any kind of error in the data, the upload would abort. However, it worked well enough that eventually we were able to hire two more encoders to begin the job of putting all of the UW's undergraduate degrees into the system. I also developed a diploma back-order database for the Graduation Office, and I was pretty darned pleased with myself.

The first woman we hired to encode turned out to be mentally unstable. She had scars on her wrists from previous suicide attempts, and she tried to commit suicide while she was working for us too. The story she told us of that attempt is actually pretty funny in a morbid way, because everything she tried failed, including closing the garage doors and starting up the car, only to have it run out of gas. Anyway, it was less funny when she accused me of emotional abuse, and we had to go through a long, painful process to determine that I wasn't actually being cruel to her.

I think by that point we had hired Tom, who was a gay man from Minnesota. He was tighter with his money than anybody I know, except for maybe my brother's friend, Steve, who funnily enough is another Lutheran-raised Minnesotan. Tom's partner loved opera and had hundreds of CDs that he loved to listen to at top volume, which got on Tom's sensitive nerves. So, like my Mom, who insisted that they add a room on their house in Crooked River, so that she didn't have to listen to Fox when my half-deaf father had it on full-blast, Tom and his partner had a grandmother apartment separate from their house where Tom's partner could listen to loud opera to his heart's content.

Eventually, much to everyone's relief, Kimberly moved on to another job, and we hired Kathy to replace her. Kathy was a much more down-to-earth, no-nonsense person who was also taking care of her sick mother. Things in DARSland stabilized for a while until Kathy's mom started going downhill and Kathy had to look for a less demanding job so she could spend more time caring for her.

I'm not sure why I went into such detail about these folks, other than to give some context for the photo. In the meantime, because I needed to pull in representative students to test our requirement encoding against, I learned how to write queries against our student data and started to learn the structure of the relational data warehouse of the mainframe flat files. I even took over the creation and maintenance of the official degree codes for the university, because I had become so familiar with them through using them in DARS. Also, once we hired Maggie to replace Kathy, we were well into the maintenance phase of DARS, and I started to lose interest in the project. While we were in the implementation phase, it was the first time that I had ever felt I got my greatest sense of fulfillment in life from my job rather than from my hobbies and pastimes. This was also the point at which a client-server version of DARS came along, and all my Access work was scrapped.

By this point, my knowledge of student data structures and Access query writing were good enough that in 2007 I was moved over to a job in the Academic Data Management Office essentially writing ad hoc queries as well as running stored queries and processes that more knowledgeable people had written. Eventually through a process of attrition through death, retirement, and post-Great Recession layoffs, I became the last person standing on the data side of the office, which also included some non-data functions such as desktop support. I always felt like a total imposter, because I didn't know how to write SQL from scratch, but I kept reminding myself that nobody knew the underlying data structures better than I did. By the end of my career in that department I was *the* go-to guy on the campus for questions about which tables had which student data and how to join the tables. If I didn't know the answer to the question, I knew who did know the answer. Needless to say, this stuff wasn't written down anywhere, and our data dictionary was always a work in progress. So I guess I earned my keep despite my lack of SQL proficiency.

And that's the long-winded story about how an English major ended up in a semi-technical job. Aside from my knowledge about the data structures, accrued over time, my other important skill was the ability to problem solve in a methodical way when things weren't working. I was good at analyzing where things were breaking down and then working my way toward a solution by a process of elimination. At least one person I worked with who was far better at SQL than I had no ability to trouble-shoot, because when she started getting bad results, she always jumped to the idea that there was something wrong with the underlying data rather than accepting the more obvious possibility that there was something wrong with her SQL. Of course she was also mentally unstable, so there's that.

Anyway, needless to say this career path was not anything I had in mind when I earned my English degree and started looking for work. But I wandered along the way, going with the flow, and found my own idiosyncratic path.

Name Plate.jpg


randy_byers: (Default)

September 2017

10 111213141516

Most Popular Tags

Page Summary

Style Credit

Expand Cut Tags

No cut tags
Page generated Sep. 26th, 2017 12:43 pm
Powered by Dreamwidth Studios