Tales from the jar side: Don't Answer Right Away
Welcome to Tales from the jar side, the Kousen IT newsletter, for the week of June 23 - 30, 2019. This week I taught classes in Functional Java and Managing Your Manager, gave four talks at an NFJS event in Dallas, TX, and made more progress on both my Kotlin Cookbook and my Managing Your Manager book. Rather than discuss those, however, I'd like to tell a story from my academic days.
Back in the late 80's, when dinosaurs roamed the Earth, I was a graduate student at Princeton in the Mechanical and Aerospace Engineering department. I wanted to go there for a few reasons. One was that Princeton turned me down when I applied to be an undergrad, so I wanted to get in and then turn them down. Yes, a petty reason, but it was in the mix nevertheless.
That didn't happen, obviously. I was surprised to get in, so I decided to visit and really liked the place. More importantly, I liked my potential advisor, Oddvar O. Bendiksen, and in a Ph.D. program that means everything.
As an aside: he was an awesome advisor, and I was his first Ph.D. student. He wound up not getting tenure, however. He stepped into a tenured job at UCLA after that and is still listed as a Professor Emeritus there. That tenure fiasco really colored my view of academia, and contributed to my looking for a job outside it. I also learned a very important lesson, which is never to drink with a Norwegian, but that's another story.
I also liked the prestige associated with the Princeton name. Most of my life I've been collecting credentials, and what's the point of acquiring a credential no one has ever heard of? Coming from high school in York, PA, I already felt like I was from the middle of nowhere, but even in York people knew about Princeton.
Some hadn't heard of MIT, the ever popular Massachusetts Institute of Technology. My mother told me that once she was asked where her son was attending college, and when she said MIT, the person said, "there are plenty of good technical schools around here, so why go all the way to Boston?"
(Now that I think about it, that probably bothered my mother more than me. Both my parents really cared about recognizable credentials. I cared, too, but I had to actually go there.)
Another reason I found Princeton attractive was that it was closer to home than my undergrad school, but not so close that I could just pop home on a whim.
(There's a reason I took my first job in Connecticut. That turned out to be just far enough away from home to limit easy access, while being close enough to not have to hear about how far away I was all the time. The drama field emitted by home decayed slowly with distance, but 350 miles reduced it to manageable levels. Barely.)
Another reason I wanted to go to Princeton was that after spending four years in a hard-core technical university, Princeton felt much more like a "real" school and I wanted to know what that was like. Of course, my experience as a graduate student was going to be dominated by technical topics anyway, but the place at least seemed vaguely normal, or at least more normal than MIT.
But the real reason I wanted to go to Princeton was that they had a very structured doctoral program that was designed to get you in and out in four years. I used to hear horror stories about grad students desperately trying to finish up after seven long years when their advisors asked them to stay to do just one more paper. I wanted no part of that. I've always been in a hurry to collect my credentials, and the Engineering school promised four years of funding and no more. It was in everybody's best interest to get you out and on your way inside that time frame.
The structure also appealed to me because I've always been really good at what I call "The Game of School." The Game of School is based on the idea that when a teacher asks you a question, you need to not only know the answer, but be able to present it in the way they want to hear. Some teachers want reasoning by analogy. Some want detailed steps. Some want references to other work, and so on. I've always been good at figuring out not only what the teacher wants, but how to give it to them in an optimal way. Being good at the Game of School is easier if you actually learn the material, but those are two separate skills that aren't always found together -- a fact I discovered at Princeton when my first office mate was an academic superstar who couldn't communicate clearly at all.
The structure of my Ph.D. program involved several steps:
You took roughly ten courses covering a minimum range of topics, which was expected to take about two years. That meant two courses each Fall and Spring semester and one each summer, so you would finish just in time to take the general exams. Strictly speaking, you didn't actually have to take the courses; you just needed to be able to answer questions about the information they covered, but presumably it was a lot easier to know the material if you actually sat in the class. (In the movie A Beautiful Mind, John Nash somehow learns everything without actually ever attending a class. I'd say that was at least possible, but doing it the hard way.)
At the beginning of the second semester of your second year, you gave a seminar on your research to date. That was intended to demonstrate that you might actually be able to accomplish something Ph.D. worthy in the two years after your general exams.
During that Spring semester of your second year, you arranged three or four "interviews" at the pace of one a month. For each interview, you would go to a faculty member's office and they would ask you questions about one of those selected areas, which was supposed to cover the work of three or four courses. The format could be anything, involving a written tests or oral examinations or a combination of both.
At the end of your second year, you took the actual general exam. That consisted of a three hour oral exam that was divided into six half-hour blocks. Each block was run by a single professor, though any professor could attend any of them and ask anything they wanted at any time. When the exam was over, the faculty would argue about you, and if they gave you the okay, you were allowed to continue on and do your thesis. If they said no, you could redo the exam the following Fall (ugh), after which there would be a yes or no decision that you couldn't appeal. They also could just tell you no and let you write up your work into a terminal Master's degree. It now occurs to me that the term "terminal" in that sentence probably wasn't chosen accidentally.
Selecting which professors to do the interviews and the half-hour blocks during the general exam was an art highly debated among the grad students. My friend Todd and I used to put it in baseball terms: Prof A threw high heat in fluid mechanics, but preferred ugly curve balls for math, so you should only select him for math in the actual exam because at least that way he was limited to half an hour. Prof B was known to throw bean balls even at his own students, so avoid him at all costs. Prof C asked the same questions every year, so if you found out from the previous year's students what he'd asked, you could prepare without actually learning much of anything. Prof D's arm was dead for years, but unfortunately he was in an area I wasn't covering so I couldn't use him.
Funny story: Like everybody, I chose Prof C for one of my math areas and was prepared for exactly the questions he was known to ask. During my exam, halfway through his half-hour block, he deviated unexpectedly. I was shocked and very nearly protested that he wasn't allowed to do that, but managed to catch myself just in time. Half an hour can seem like forever when you're lost, but I managed to string together enough coherent statements to somehow get through it.
We students developed certain rules to follow during the exam. The biggest one I remember was, when all else fails, conserve mass (or, more properly, momentum). Conserving energy was a nice idea too, but that was always harder because there can be so many different sources. But if you started off by saying the mass coming in equaled the mass coming out, you couldn't go too far wrong.
The idea behind all those preliminary stages (the coursework, the pre-general seminar, and the interviews) was to weed out anyone who wasn't going to pass the general exam. My year started out with eighteen students, but by the time the general exam rolled around in May there were only twelve of us. They scheduled us two students at a time each morning and each afternoon for three consecutive days. As luck would have it, I was in the very last group.
In a way, that was good, because I was using Prof C for math, so I could check with one of the earlier students to verify that he was still asking the same questions. Which he was, until he got to me and deviated. I'm still annoyed about that.
Each year they normally failed one or two students. My year they failed two students the very first day. I rationalized it by saying at least they got both expected failures out of the way right away, and it would be clear sailing after that.
Then the failed another student the second day.
Ugh. I went for a long walk around the Institute for Advanced Study, which is where Albert Einstein used to hang out from the 1930s to the 1950s. I didn't actually go inside -- I'd long come to realize that I would never be qualified for that, not that anyone would have cared if entered the building -- but the area was peaceful and I needed to burn off some energy. There was nothing practical I could do anyway. Last minute studying wasn't going to make much difference at that point. As Tom Petty said, the waiting was the hardest part.
As it turned out, they didn't fail anyone else, including me. I actually did very well on the exam, which makes sense when you think about it. The whole structure of the exam was designed for people like me. I'm good at the Game of School, I'd chosen my professor/interrogators with care, and I'm comfortable speaking in front of an audience. I even managed to throw in a gag at one point and cracked up the room, which I'd like to believe helped. If nothing else, it relieved some of the tension.
No, my problem was that I completely failed my first interview. The whole process nearly ended right at the beginning.
I'd chosen a professor I didn't know well (I was running out of options anyway) and the topic involved a subject where I knew the basics but was still fuzzy about the details. But my real mistake was that I tried to answer all his questions immediately. He would ask a question, and I would give him the first answer that occurred to me, at which point he'd say something like, "Really? Then that implies xyz, right?" I'd stammer and try to agree, and then he'd challenge it with some edge case or some principle I'd just violated. This happened multiple times. It was a disaster. I never even got a chance to show what I knew, because as soon as he realized I understood something, he'd move on to another question.
If you've ever been in one of my talks or training classes, you know that I'm pretty good at breaking down complex topics into segments I can easily explain, but that's only true if I really understand what I'm talking about. If I'm uncertain, or if I'm worried I'm actually wrong, the whole process falls apart. Instead of looking like I know something, I come across as not understanding anything.
After the interview, the professor went to talk to my advisor, who sat me down for a little chat. "When they ask you a question," he told me, "don't answer right away. Write it down on the board. Restate it. Think about it. Work the problem. Nobody is expecting the right answer immediately. They want to see how you handle questions when you don't know the answer."
(And, as I mentioned above, if all else fails, conserve mass.)
How horrifying. But after my advisor assured the professor that I wasn't a complete idiot, he gave me a chance to do the interview again. I performed much better the second time. I don't think I did a great job by any standards, but I did well enough that he let me move on.
All my years of experience since then have taught me how to handle it when I'm confused. I have much more patience now, though still not a lot, and I can tell when I don't really understand something (like Kotlin coroutines, so that chapter is going to be tough). I also know that when I encounter something really new, I have to repeat it at least three times before it sinks in. I'm not happy about that, but it's a fact of life.
I've been thinking about all of that a lot lately, as I'm working on two separate books and struggling with hard sections of each. By now I have enough faith that I'll eventually figure everything out, so I don't quit or throw things or use excessive amounts of profanity (yes, the word "excessive" is doing a lot of work in that phrase), but I still hate not knowing. At least now I realize it's all part of the process.
So if you wind up reading either of my books, know that none of it is as simple as the text makes it appear. There are hours and hours of struggles hidden throughout.
Plus, at my age, all of that conserved mass wound up on my waistline. Sigh.
Last week:
Functional Java online on Safari
Managing Your Manager online on Safari
Prepared the latest Early Access release of the Kotlin Cookbook
No Fluff, Just Stuff event in Dallas, which was really fun
This week:
Spring and Spring Boot online on Safari
Prepare three chapters of Managing Your Manager for their first review
Complete at least two proposals for new training classes on Safari