Why I left academia

2009-05-24 01:21 by Ian

I decided that I wanted to be an engineer when I was eight years old.
I had found my dad's packed-up box of engineering notes and textbooks. Specifically, a course called "Feedback and Control Systems". I was a confident reader at 8, but the thing that captivated me was the wordless clarity of the schemata. Then one day, my dad picked up a hand full of junk in the garage and showed me how to make an electromagnet, and the choice was made.

By age 10 I could solder, read electrical schematics, and build simple circuits. That year my teachers thought I was dull-witted and had me tested because they couldn't reliably engage me in class. When they got the results back, I was put into the GATE program. As it turned out, nothing the school could give me was beyond my grasp. I was just bored.

By age 12, I was building my own toys and fixing VCRs for neighbors. I used the money I earned to buy my first oscilloscope.

Before I had facial hair, my long-suffering mother had to listen to my explanations of allowable dose limits from the magnetron gun that my friend Matt and I made from reclaimed microwave ovens, and were using to energize fluorescent light fixtures from afar. But I think she was more concerned the day she found out about my arc furnace improvised from a broken CRT television when I showed her the crude glass that I'd made from sand in the yard. She was much relieved when I took up QBASIC and started building custom ISA cards for the 386DX they'd bought me the year prior.
As long as there were no explosions, uncontained fires, or police reports, my dad's philosophy and point of view was simple: "He knows more than me and he isn't dead yet."

By the time I was 14, my parents had come to believe that my misjudgement wasn't going to get me (or anyone else) killed or maimed. And so receiving phone calls from my middle school regarding the confiscation of my home-built helium-neon lasers was something to laugh about.

I was building my own microcontroller projects at 16, and ditching class in highschool to play with my 68K dev-kit. I loved my AP computer science class, but my senior year English teacher was the only teacher that really understood how to engage me (I miss you, Mrs. Popejoy). She gave me all the hardest literature she could think of. In the time it took the rest of the class to read Brave New World, I'd finished not only that, but also The Count of Monte Cristo and Les Miserables. Every week that year, I finished at least one book from her reading list.

All of my life I was a decent student, but I was usually bored senseless. And every year I was told "don't relax, it will get harder", but it never did. I scored a 1310 on the pSAT and a 30 on the ACT without studying for either. I was top 5th percentile according to the ASVAB and Iowa tests. I'm fairly certain someone has a record of my highschool graduation. But having been socially uprooted by a cross-state move in my junior year, I didn't even care enough to go to my ceremony.
Nothing about school ever felt like "accomplishment" compared to what I felt from my own self-directed efforts.


When I finally got to college, I had already taught myself digital design, and had been writing 32-bit assembly for two years. I'd written an ATA driver on the ColdFire, and had a uCLinux build for that target by that point. But I was not permitted to test ahead of CSE120 (Digital Design) nor CSE225 (ASM on 68K).
"Everyone needs to take these intro courses."

Wonderful... CSE120 is a pre-req for CSE225. That's a delay of a whole year before I can do something new. I decided to put up with it, and take my math and elective load. CSE120 was mildly worthwhile. It introduced me to LogicWorks, and designing a simple CPU was a fun exercise. But it wasn't remotely challenging. I walked in the door already knowing how to do K-Map optimization, and constrain glitch. I once had to fight for my grade because a TA who was grading one of my projects didn't understand my solution (which was better than he thought possible). I had used an XOR gate in place of two NANDs. But he disagreed with my truth table (which I am certain I know, and can validate by inspection). I won the argument with LogicWorks.
It took me a few minutes to realize and believe it. But I just taught a grad student something that his teachers hadn't. Something basic.
He had never seen an XOR.
How is that possible?

That semester during finals, I took a break from building this to show up at my final lab practical for CSE120 30-minutes before the end of the window on the final day. Upon entering, the TA running the lab scoffed at me incredulously...
"You think you're going to pass CSE120 in half an hour?"
Misunderstanding his meaning, I replied "Don't worry, it won't take me that long."
The proctor found this funny enough to laugh.
But to his credit, he gave me an appreciative nod when I passed it in 12 minutes on the first attempt.
I don't know whether the proctor noticed. but I didn't use reference cards for the 7400-series logic chips. I had their pinouts and timing diagrams committed to memory years ago.

On the first day of CSE225, the professor (actually the professor, not a TA) addressed the class thusly.
"This is a weed-out class. The department tells me that I must fail 25% of you. If you do not pass this class, you cannot enter the upper-half of the engineering program."

Ok... that's a... strange way to educate people. Fortunately, I've been doing this class in my sleep for the past several years. But a few weeks into the class, I noticed that my grade was suffering because of daily attendance check quizzes. I went to the professor's office hours and explained that I was working full-time with a 21-credit hour load, I showed him my Coldfire-based hard drive controller, and asked if I could take all of the quizzes for the semester all at once there in front of him so it wouldn't hurt my grade. After all, is it really important that I come to every class if I already understand the material?
He looked at me like I was an unwelcome and unimpressive intrusion into his open office hours and said simply:
"Everyone needs to show up every day."

This was mathematically incorrect.
I went to that class a total of eight times to take tests and turn in projects, which were always scored better than 100% and passed CSE225 with a C+. About 25% of the people who went everyday failed.
Thanks, professor. I'll see you in CSE240 (Survey of computer languages) next semester. Maybe I'll actually need to show up for that one.

Which, as it happened, I did for the two-thirds of the time we covered Scheme and PROLOG, both of which were new for me. But the third that was devoted to C++ was more review. I was already acquainted with the imperative and procedural paradigms in my primary high-level language.

CSE210 (Data structs and algos) would have been great for me if it had been triple the pace, or skipped the first two-thirds (which I took in an AP class in high school). As it was, I ditched class to do things like this or like this rather than sit through the tutorial-tier (sophomore-level??) class that I attended six times and passed with a B+ (10% of the grade was attendance checks).
Meanwhile, BitTorrent was becoming a thing around that time (2003), and I was reading about distributed hash-tables when I wasn't being graded to sit through review. I realized one day that my syllabus wasn't even going to get me to where I already was when I started my first semester.

And so it went for my time at ASU. I legitimately needed to take the math classes, and a handful of other things. But after two years it made little sense to me to pay so much time and money for some bureaucracy to tell other people that I know how to do the things that I am already provably doing.

I left the engineering school out of boredom and disillusionment, and took a semester in the business school to learn some practical skills that I knew I would want later if I wanted to run a business. Once I learned them, I dropped out of ASU to contemplate life for a year.


Perhaps my problem was that I went to study something that I knew better than most of my teachers. I went back to community college to take every science and math class at which I wasn't already an expert. I was pre-med during this period, but also taking all the psychology on offer. My teachers were often retired practitioners, rather than lifelong academics. And I had a far better experience of college.
I actually wanted to show up for Human A&P, and scored 95th percentile on the ACS organic chemistry exam. I never bothered to take the MCAT (I didn't really want to be a doctor), but I'm sure I would have passed it. I was always a top student.

I made a good impression on my (most excellent) biology professor, and BIO181 (General Biology for majors) radically changed how I saw the world. I was invited to join his undergraduate research team, and I accepted. I had a wonderful time doing it, despite the long-term damage to my health I knew I was incurring by burning as hard as I was.
Two years flew by. I got two associates degrees on accident, and ran out of things to study,

I never considered returning to ASU to finish a BSE.
One of the capstone classes of my abandoned degree path was operating system design. Having written an operating system de novo, and read the source code of many others, the idea of paying someone who knows less than me to bless my knowledge for the sake of credentials that are destined to rot was... too unappealing.

So I would stick to what my 8-year-old self decided he wanted to do, and make a profession of engineering without the degree (before it was cool). Whatever college was or said about a person in my grandfather's day (PhD, economics) or my father's day (BSE, mechanical), it was gone by the time I arrived. But if the point of going to college is to become educated, I succeeded. And it hasn't stopped me from consistently being a top-earner in my field.

On the final day that I walked off campus, I had enough of a portfolio that my employers were more interested in what I have actually done versus what a degree says I can do. Moreover, I prefer the company of people whose values are so tuned, and have been blessed to work for many of them over the years.

But sometimes, I still miss working with Dr. Nagy.



Side-note: Somewhere in the world, there is/was an ornithologist and truck driver with four doctorate degrees named Tom Gaskill. When I was 19 and earning money for school, he helped untangle my thinking over the course of a few months in which we worked closely together. He is possibly dead by now, had no family, and wasn't the sort of man that wanted to be remembered or even thanked for his good deeds. But if he or his friends ever read this, know that his influence is still being felt.

Previous:
Next: