Absolutely No Machete Juggling

Rod Hilton's rants about stuff he cares about way too much...

Spring 2013 Semester in Review

Another semester is over, and it was quite the doozy like the last one. This time, however, it wasn't because I hated the workload (like last semester) but because I hated the material. Or more specifically, half of it, but more on that later.

This semester, I finally took the cross-department course from the Business School that I was dreading. My school's program is a joint-degree, Computer Science and Information Systems, with the CS stuff coming from the Engineering school and the IS stuff coming from the Business school. The degree has a number of requirements but one of them is that you must take at least one PhD-level class in each school. This means that I am able to take nothing but pure-CS courses, except I have to take one PhD-level course in the Business school. I've been dreading this since I started, somehow hoping this requirement would be removed before I had to take it, but this was not the case, and I had to take it this semester. With the credits I am able to transfer from my Master's degree, this actually completes all of my course requirements, but because I was interested I also took a course on Graph Theory, which was a class crosslisted as both an undergrad and master's level class. And once again, I somehow managed to get stuck grading.

I started the semester taking three courses on Coursera as well, but quickly decided that doing so contributed to my sense of being overwhelmed last semester, so I quickly dropped them and decided to download all of the lectures when the classes ended, and hence won't be mentioning them further.

Classes

Applied Graph Theory

Textbook: Graph Theory, a Problem-Oriented Approach

Like I said, Graph Theory wasn't a course I took because I needed it, I took it because I felt like it would be low-key and fun, and a lot of the problems I'm interested in for research relate to Graph Theory in some way or another.

A great deal of the material had been covered in the same professor's Graduate Algorithms class, but not as much as I was expecting. There was lots of new stuff that I learned, and I really enjoyed it. This particular professor is a very good lecturer; I always enjoy her classes, and this one was no exception.

The course is strange because its quadruply cross-listed. It is cross-listed as both a CS class and a Math class, and also cross-listed as an Undergrad and Graduate class. This led to an interesting mix of students in class, many lacking the math background of others, many lacking the CS background of others still. All together, there were 13 CS Undergrads, 15 Math undergrads, and 5 CS grads including myself (no Math grads). Due to the makeup of the class, the material was definitely taught at a level catering mostly to the undergrads, which made for a pretty easy class for me and the other graduate students. After the last semester's overwhelming workload, this was welcome.

Graph Theory!

Graduate students were required to also do a semester project and present it to the class at the end of the semester. As was the case last time I had a similarly-structured course, once I got into my research project I found it hard to focus on the rest of the class material due to being obsessed with my own project. My understanding and retention of the first half of the class material is definitely better than that of the second half. Because the professor drops the lowest homework grade, I didn't even do the final homework for the class, electing instead to put more work into my project.

As the only PhD student in the class, my project presentation was allotted an entire class period for the presentation (the other students each got half of one), which consisted mostly of me lecturing on the background material not covered in the class before presenting my research. For the project, I was given an interesting, NP-Complete Graph Theory problem to try and optimize, which I attempted to find a solution to by implementing a Genetic Algorithm search. The code for this project is here, my presentation slides here. In the end, I was unable to actually solve the problem, but I had a great time trying.

I had an absolute blast with this project, and my presentation went very well. I actually coded it up so that the students all met in the PC lab on campus instead of in the regular classroom, and they each ran their own copy of the search algorithm and uploaded their best result automatically to a server I was running connected to a projector. This turned the search into something of a competition, and the interactive component was a lot of fun. I somewhat regret shifting my focus too far away from the class material once I got invested in my project, but overall I had a great time and really enjoyed this class a lot. I wound up with an A in the class.

Topics in Analytical Research Management

Textbook: None

Like I mentioned, this class is a requirement, the one Business School class that CS students needed to take as part of the degree requirements. I had been dreading it since I heard about it, as I have no interest in anything I could learn from the business school, and even less interest in this particular class subject.

I hated, hated, hated this class. I hated every second of its 3-hour length. I hated the fact that I had to stay downtown after work and grab food before its 6:00pm start time instead of eating dinner with my wife at home. I hated the complete lack of common ground between the few CS students taking the class and the overwhelming majority of business-school types in the same room. I hated the fact that almost all of the material was built on previous classes that the business school students had taken before but the CS students had not. I hated every lecture, every assignment, every student presentation I had to sit through. I hated the over-reliance on Powerpoint for lectures. I hated all the business-buzzword talk that I try so hard to avoid in my day job. I hated that I never saw a single mathematical proof of anything being discussed. I hated having to use Blackboard. Most of all, I hated how it was the only course I took not because I thought it sounded interesting, but because the department required me to. I deeply, deeply resented having to sacrifice an evening at home with my family every single week to take a course I had absolutely no interest in, that I felt I should not have to take at all, with a bunch of people with whom I had nothing to talk to about. I hated this class.

I did not hate it due to workload, which was relatively minimal. Let me first start out by saying that, at no point during the semester did I actually understand any of the class material. I mean that literally, it all seemed built on classes in the business school prior to this one, classes I did not take, and I never understood anything. Almost every lecture may as well have been in French, I didn't even recognize the vast majority of the words being used, let alone understand the sentences they were in. At no point did my understanding of the material reach a level above a D or D+. By all rights, a D or perhaps an F is what I deserved.

However, the assignments were extremely easy, though occasionally time-consuming. Four of them were to simply read a paper from a list of possible papers and write a review, akin to the kind of review one might write if reviewing for an academic journal. Of course, none of these articles were CS articles, they were all business-type articles. The most interesting paper I saw was on the subject of machine learning, and the opening paragraph talked about how it could be used to increase return on investment for businesses and my interest was instantly sapped. I just have no patience for this money-centric worldview.

I managed to get an A on all four of these articles. I never read a single one of them. I read the abstract, the introduction, and the conclusion section of each. I rarely understood any of these sections, and I didn't even attempt to read the stuff in the middle. Yet they were enough to write my "reviews", which largely consisted of finding a detail and generically referring to it as "under-explained" or something similarly vague. Every one of my article reviews was complete and total bullshit, an attempt to make it look like I understood the articles when I absolutely did not. My lowest grade was on a review of a paper that the professor was a co-author on, where my same generic criticisms were met with defensiveness, explaining that the content I felt was missing was cut out due to the journal's length restrictions.

There were also a number of assignments that involved a lot of calculations in Excel. I never understood the purpose of these calculations, or what they meant. But I was consistently able to look at the professor's example Excel files, re-use the formulas out of one column and put them into my Excel sheet, just changing the cell numbers to use what appeared to be analogues in my worksheets to the cells in his. These assignments were, for me, exercises in pattern recognition, a skill at which I'm quite adept. I was able to get decent grades on these assignments as well, only occasionally did my pattern matching skills fail me.

A month or so before the midterm, I decided that I was being lazy. I decided I was only having trouble understanding the material because I disliked it, and that's no excuse. So I went to the professor's office hour, with one goal in mind: to walk out understanding exactly one thing. We'd covered dozens and dozens of different concepts by this point in the class, and I understood none of them. I decided to pick one that was referenced a lot, and fully understand it. Sure, I'd still be crazy behind, but at least it would be an improvement.

I asked my question and he immediately launched into a tirade of nonsense that baffled me. I asked him to hold on, slow down, go back and explain so-and-so first. I took up an entire hour trying to get him to explain things more simply before eventually the professor shrugged and said "you know, this stuff is pretty basic statistics, maybe I should have had a statistics class as a prerequisite." He then gave me a URL to a web page with hundreds of written pages about statistics and told me to read it. He didn't direct me to a particular thing to read up on, just the URL which had at least two semesters worth of reading material alone. First of all, I have what I consider basic statistics knowledge, I took statistics in high school at the very least. What he was referring to was more like graduate-level statistics, which I'll admit I don't know much about, largely because I have little to no use for such a thing studying theoretical computer science. Secondly, this course is already a course I'm required to take and don't want to, the idea that he'd add another one as a prerequisite was infuriating. And finally, it certainly didn't do me any good right now, so thanks a pantload.

Synergize our opportunity platform to facilitate tiger team knowledge management

So I gave up. I was mentally done with the course. I felt like I gave it an honest effort to understand something, anything from class, and the professor basically told me to fuck off. From this point forward, I resigned myself to the knowledge that I'd fail the midterm, but with the decent grades I got on the assignments and the fact that I'd be able to work with a partner on the final project, I'd probably manage to get a C in the class, which would be enough for it to count. I spent every lecture from this point on working on my Graph Theory project on my laptop, I never looked at another one of the professor's slides again. I honestly don't even know what many of the lectures were even about, I just used the class as a 3-hour work session for my project.

Sure enough, when it came for the midterm, I did poorly. There were 4 total questions, 2 of which I actually understood relatively well due to reviewing the material with another student in the class (who was able to explain, in about 10 minutes, what the professor could not explain to me in an hour). The final question -- worth the most points -- was, once again, barely English for me. I didn't even understand the words in the question, let alone how to answer it. So once again, I just wrote garbage down. I wrote 6 paragraphs, each exactly one sentence long. Each sentence was a statement that was true, which I knew from having memorized random sentences from the slides from class. Did any of these statements answer the question in any way? No idea. I just knew they were true statements about things, and that they were stated in class. I got the second lowest grade on the midterm in the class, but when I compared my answers with the other student I studied with, he was dismayed when he saw that he and I got the same number of points on the final question despite the fact that I clearly did not understand it. I got a B- on the midterm, not because I understood the material at a B- level, but because of how it was graded.

There were only two more graded artifacts left. One was to read a paper of our own choosing and present it to the class, similar to the paper reviews from before. I did extremely well on this one, because I was able to select a CS paper and I'm pretty good at giving presentations (he actually said it was one of the best he's ever seen). The final assignment was a project, and I was able to work with a partner (the one I studied with before). We selected a very CS-oriented project and did a very cool experiment, which utilized a fraction of classroom material. He was largely responsible for the research and the paper, whereas I did a lot of the gruntwork, writing scripts to generate data sets as well as all of the slides and the project presentation to the class, which is here. This division of labor worked well for us because he understood the class material better than I did, but I was better at giving talks. In the end, we got a low A- on the project. We were the only group who not only designed an experiment, but actually performed it and got real results. Everyone else designed experiments that they couldn't actually do, so it sort of drives me nuts to know they got a higher grade for utilizing more material from class and doing absolutely nothing with any of it.

These assignment and project grades were actually enough to make up for my midterm grade, and I wound up with an A- in the class. In a way, I actually wish I'd gotten a B, since my project partner, who scored only a few points higher than me on the midterm and otherwise had the same grades, got an A. This means my grade was damn close to the A, and is now the lowest grade I've received in all of my classwork, and this is the one class ruining my perfect 4.0 GPA. This kind of enrages me since I never wanted to take this class to begin with, but I have to try and keep things in perspective. I did not understand the class material at an A- level. I still do not understand much of anything from this class, and by all rights I should have failed. But due to the structure of the class and the grades, I managed to get an A-, so I don't have to retake the class and it counts towards my degree, which is really all I needed.

Grading

Believe it or not, I once again got roped into grading. The professor for Graph Theory asked if I could grade her class. I said I didn't really want to, but I didn't want her to be in a jam, so she should go ask everyone else on her list and if she can't find someone, come back to me. She found another grader, but the other grader could only do it part-time, so she wanted to know if I would be willing to be a second grader and grade part-time as well.

I'd only have to grade half of the six quizzes, and I'd have one office hour per week instead of two. I figured this was tolerable, so I reluctantly agreed. Then after the semester started, the other grader got re-assigned by the department to grade another class, and I was the only one left for Graph Theory, so I had to double my office hours and grade all of the quizzes. It was too late to get anyone else, so I was stuck with it.

This was the first time I was grading written work instead of compiled programs, which meant I had to deal with handwriting, lack of descriptions, poor English, errors in proofs, and crazy, off-the-wall answers from left field. I also had to figure out for myself how many points I'd take off for particular errors, whereas with my previous grading effort I was given a rubric by the teacher. Grading this type of work is much more time consuming and much more difficult. I hated it. I also had to have office hours every week so students could come by and argue with me about the points I took off. I hated this more than anything else - nobody ever came to me and asked why they got the answer wrong because they wanted to understand the material, they almost always acknowledged they got it wrong but felt I took off too many points. This type of crap annoyed me to no end, particularly considering how obscenely lenient I was with points. I was giving 50-75% of credit to answers that I'm sure the professor would have given a 0 to. I know this is the case, because the only grading feedback I ever got from the professor was that I was being too generous with points.

You may have also noticed the name of the course I graded was "Graph Theory". Wait, isn't that the name of the course I was also taking? Why, yes, yes it is. PhD students in the department are allowed to take and simultaneously grade the same class, the professor grades the TA's work, and the TA grades everyone else. If that sounds a bit strange to you, let me assure you that you have no idea how strange it is.

I had to sit, every class, with students whose grades were lower than they wanted because of my grading. I had to hear the comments about what "bullshit" their grades were as their quizzes were handed back with me standing right there. Worst of all, I had to present a full lecture on my semester research to what I can only describe as a maximally hostile audience, who I was certain would have enjoyed nothing more than to see me stumble and fail as karma for taking 10 points off that one time when I "totally should have taken off like 2 or 3, max."

I had no friends in the class, and no other students I could talk to when I was having trouble with the material myself. I had to figure everything out early, before my office hour the day before assignments were due, which was inevitably when everyone would show up to ask me questions. The sense of being completely alone and an absolute outsider in the class transformed what would have otherwise been an awesome class into something awkward and uncomfortable. It placed far more stress on me in terms of needing to understand material out of fear of humiliating myself or telling a student something incorrect. This alone turned my "low-key, easy class" into a stressful burden.

The value I got out of this was that it has made it abundantly clear that I do not want to teach. I actually really enjoy lecturing and teaching people, but I detest with every fiber of my being grading their work and having stupid little grade-grubbing conversations with them where I spend 30 minutes listening to them explain why they deserve 2 extra points while in my head thinking how they really deserve a 0 and I was already being nice by giving them partial credit. The lecturing part of teaching is fun, but the evaluating/grading portion has made it crystal clear that teaching is not for me.

I know I've said this before, but this time I totally, seriously, super-duper mean it: I am never going to be a grader again. Dammit.

Conclusions

This semester was tough. I hated one of my classes more than any other, even Artificial Intelligence (where I at least liked the material), and the one that was supposed to be low-stress and fun wound up being highly stressful and unpleasant due to the grading situation. It was my second semester in a row where my overall level of enjoyment of school was low due to overwhelmingly negative reactions to a particular class. I had two semesters of awesome school, and then what wound up being another two semesters of unpleasantness. Considering that the only reason I'm in school is because I thought it would be fun, you better believe I was wrestling at least once a week with the question of just dropping out. Usually in the minutes before the Business School class started.

I stuck with it, largely because the one constant in the courses that I didn't enjoy was that, when I was doing a project or research for the semester, I was loving it. As much as I hated the business school class, I'm really proud of the experiment my partner and I designed and I really had a lot of fun working on it. As stressful as Graph Theory was due to my "other" status, I absolutely loved my project, and really enjoyed presenting it, even to a hostile audience. Last semester, I hated the A.I. workload, but I had a blast doing the semester project. And even in my Computational Complexity class, which I really enjoyed but was disappointed with due to the infrequency of class, the highlight was my independent literature review.

Since the end of this semester actually marks the end of all classwork for me, I slogged through the rough semester hoping that the next semester, where I'll be doing research completely on my own, will be a lot more of the thing I liked from the classes that I did not. I came to the school looking forward to actual classroom material, sitting in a physical class instead of taking classes online like I did for my Master's degree. I enjoyed that experience most of the time, but I'm definitely at the point where I'm ready to be done with classes and doing research.

I had hoped that this semester would be low-key enough that I could spend lots of time figuring out what I wanted to research, but unfortunately both classes wound up being more time-consuming than I had hoped (particularly at the end where I was juggling two large projects). I'm not as far along in the "what do I want to research" process as I was hoping to be, so I'm hoping to catch up a bit this summer.

I wish I could have enjoyed all of my classes as much as I enjoyed some of them, because I think it would have been possible to do so for all but the Business School class if not for particular circumstances. In a way, I'm sad to see the chapter of classwork for my PhD close with a little under half of that time marked by disappointment in some cases and outright hatred, but close the chapter has, and it's time to buckle down and start doing my own research.

This period will be marked by self-doubt, I'm sure. Even now I don't really know what I want to research (though I've developed quite a reading list), and I don't really know if I'm up to the task of coming up with something truly original and novel. I know that I enjoy the process of research, as it's been the only aspect of more than one course that I actually liked, so I'm hoping I can find a problem or a set of problems that I'm both passionate about as well as confident I can solve. I hope next school year is better than this one was, if I reach the point where 2/3rds of my time in school has been unpleasant, I don't think I can possibly justify sticking with it.

comments powered by Disqus