On Academic Tenure

I’ve recently become engaged with many of the intricacies of the academic tenure process. I found out last week that Brown University’s Tenure, Promotions and Appointments Committee had voted to deny my advisor tenure, all-but-guaranteeing that he will find himself without work in a year’s time.

I’ve spent much of the last week organizing a student petition to convince the University to reconsider their opinion, and consequently, have developed some opinions on the process of receiving tenure (within computer science) as a whole.

Let me outline the process usually followed by assistant (non-tenured) professors: Find as many grants as possible within your research area. Use these grants to acquire a large set of qualified graduate students. Spend your time supervising these students as they publish, sometimes achieving ten or more papers within a year, and attempt to get these papers into the most prestigious publications available. Speak at conferences whenever possible.

I have nothing against professors who follow this model. It’s almost certain to produce a positive recommendation for tenure after five years, and I’ve taken many good classes with assistant professors of this type. However, I do think that the nature of the tenure review process precludes many other forms of professorship, and that universities would do well to consider the types of academics that will (and will not) make it through this trial.

For reference, at Brown, the computer science department is responsible for generating a review (positive or negative) for the faculty member seeking tenure. It’s a confidential vote taken by already-tenured faculty, and is submitted to the Provost, as well as the aforementioned Tenure, Promotions and Appointments Committee. The Committee, which is comprised of professors from across the university (with no particular expertise in computer science) reviews the tenure dossier of the professor, and further generates a review, which they submit to the Provost. The Provost then makes a final decision, which in most cases is just a rubber stamp of the Committee’s review. (There are a few other steps in the process, but they don’t involve reviewing academic work, and hence are superfluous for the purposes of this discussion.)

I believe the nature of this process produces systemic biases against professors who meet any (or all) of the following criteria:

  • They choose to work on questions where success, let alone rapid publication isn’t guaranteed.
  • They opt to have fewer graduate students, but spend more time supervising each of them.
  • They invest a significant portion of their time in undergraduate education.

The first of these biases – against professors who work on “harder” problems – is present for several reasons.

First, and most obviously, working on harder questions inherently risks having fewer publications. Some professors will happen across a paradigm-shifting gold mine, but many won’t, and as a result they’ll be left without many papers to their names when they go up for tenure review.

Second, tenured faculty (the only faculty involved in reviewing assistant professors) are biased against people who follow a different pattern than they did. Whether it’s called a rite of passage, trial by fire, or something else, many professors won’t be keen on “cutting out” the portion of one’s career where they struggled to churn out papers and managed a massive research group, because they themselves also had to struggle through the pain.

Third, there is an inherent risk in trusting a committee (and Provost, for that matter) who aren’t experts in computer science to evaluate the importance of papers. It’s a far safer bet to simply have dozens of papers to your name, because it’s more straightforward to trust in numbers.

These biases work to prevent professors from working on big-picture questions until they achieve full professorship, a process which takes upwards of a decade in many cases. It overlooks the fact that quality of papers is not additive: A single groundbreaking paper cannot be compared with dozens, hundreds, even thousands of incremental improvements. I’m not advocating that every faculty work solely on “moonshot” projects, but there is some value to aiming big, even if the payoff is far from guaranteed.

The second of these biases – against professors who have fewer graduate students – is present for many of the same reasons as the first. Yet, it has an additional pernicious effect: Graduate students leave the program without much face time with the professor, and moreover, professors are turned into glorified administrators, responsible for securing grants and reviewing papers and little more. No idealistic undergraduate seeks a Ph.D. for this to be their future. If we want academically talented students to pursue doctoral degrees, then the process of attaining professorship shouldn’t force them to spend a majority of their time supervising graduate students.

The last of these biases – against professors who invest their time in undergraduate education – is harder to explain. Brown, in particular, claims to prize undergraduate education, and has many times more undergraduate students than graduate students (unlike, say, MIT.) Hence, the professors that are attracted to the department will clearly be those who would prefer to work, and conduct research in, an undergraduate-focused department. If these same professors are asked to commit their time to research, fitting lectures in around meetings and leaving most course administration to teaching assistants, it stands to reason that this may not be the reason they came to Brown in the first place.

Many professors manage to teach excellent courses and produce dozens of papers, but they are, well, uninvolved. They set the standard: Two office hours a week, occasional emailed responses, and otherwise management is left to the rotating cast of undergraduate students who perform much of the grueling work of grading assignments, explaining concepts during clinic hours, and handling student complaints. After a few years, during which the assignments and lecture notes ossify, the course essentially runs itself.

This is what I generally expect of new classes I take. So, when I encounter the rare professor that bucks this trend, that chooses to spend late nights helping his teaching assistants with the hordes of students needing help, or who actively solicits advice to improve the course (even if that advice causes drastic changes), it makes me remember why I came to this school in the first place. Yet, this type of professor does not just go unrewarded, they are actively punished for committing their time to education (and in this case, will be fired.) It’s a cruel irony that the professor most-admired by students, a professor whose door is always open (until 3 a.m. in many cases), who is always willing to give honest advice and hear real complaints, is also the professor most likely to be denied tenure.

I would trade one professor who has a real commitment to undergraduate education for a dozen who don’t. “Senior lecturers” don’t count–the title, unfortunately, implies that the holder couldn’t hack it as a real professor, whether the department intended it as such or not.

If Brown, or universities in general, can’t recognize that research en masse isn’t the only attribute that makes a professor valuable, then there is no reason for them to even enroll undergraduate students. If the University doesn’t reconsider their decisions, I’ll have a hard time recommending the school to high school seniors. The truly formative experiences that are a hallmark of a liberal undergraduate education can’t be had with professors who spend their lives running from one research meeting to the next, fighting their way up the tenure ladder, with undergraduates left to wonder what, exactly, they are paying for.

On Traveling Respectfully

I recently participated in a weeklong photography course held in Oaxaca, Mexico. For a week, the six students (myself included) spent dozens of hours working with families from surrounding villages. We ate meals together, played with their kids, and helped with housework–all under the supervision and assistance of bilingual teaching assistants who both acted as translators and instructors.

It was an incredible experience–I’d never been to Mexico, and had never had the opportunity to practice documentary photography. I returned with a broadened sense of the world, a large portfolio of images crafted during the week, and several questions hanging on my conscience.

The instructor for the course, Stella Johnson, has been traveling to Oaxaca for 35 years, and has known these families for much of that time. She’s also (perhaps obviously) fluent in Spanish. Yet, the six participants in the class were essentially tourists: We didn’t speak more than a few sentences of Spanish, though some of us tried to learn more during the week itself.

Now, these families were hardly blind to the dynamics of the week. They’d hosted students from classes like these for many years (decades, in some case), and perhaps appreciated the opportunity for their kids to interact with Americans that weren’t on television or the news. The kids, for their part, were largely too young to identify the oddness of the context–they were more interested in playing basketball or climbing trees than questioning what had brought us to Mexico.

The result was an odd mixture of extraordinary hospitality (by the families) and deep unease (personally). I’d like to think that I impacted the family in some way: By the end of the week, another student had given a spare laptop to the kids, and I’d loaded it with educational videos and translation apps so that they could learn English, math, and science to complement what they did in school. I helped to split firewood, pick up around the yard, and teach English to the kids. Yet, throughout the entire week, I felt like every interactions was passed through the filter of photography. I put my camera away at times, or let the kids use it themselves (resulting in hundreds of blurry shots of random objects), but even simple play was tinted (in my mind) by being an observer of their life, to whom a camera and international flight weren’t prohibitively expensive, and who would jet out of the country in a matter of just days.

I appreciated the trip immensely, but I think it is incumbent upon any photographer (or more broadly, human) to think about what responsibilities one has in situations like these.

I’m not sure what makes someone transition from an intrusive tourist to an involved documentarian. I’d say that committing an honest effort to speaking at least some of the local language (Spanish, in this case) is a critical first step. I also think that spending with the families before even taking a camera out might reshape the dynamic: If the first two or three days had been spent helping with housework or homework, and only then had I brought my camera out, I’d imagine I’d feel less like an outsider looking in.

Of course, much of this was constrained by the brevity of the course. In the span of a week, we had to acclimate to a new city and different language, all the while finding time to hone our skills as photographers. I give great credit to the instructors for crafting an excellent course despite these limitations, and I’m deeply grateful for how welcoming I only wonder what more could have been accomplished had we, the students, had more awareness of these challenges ahead of time.

On Bootstrapping a WordPress Server

I recently decided to (re-)launch my personal website, given that recruiting season is starting in a few months. I previously ran a static site through Github Pages, but after several years of alternately building my own frontend and trying out (usually buggy) open-source themes, I decided I wanted to take the simpler route and launch my own WordPress installation.

After launching a DigitalOcean virtual machine, running Debian Stretch, I loaded it up with standard system administration tools. I obtained certificates via Let’s Encrypt and started up Nginx and MySQL, not to mention an FTP server and a private Jupyter notebook server. Then, faced with the decision of how to configure a web server, I opted to run with WordPress. I’d used their hosted option previously, but had avoided it for the last several years given my general lack of interest in PHP (something that hasn’t changed.)

Tired of weeding through npm modules in search of a working NodeJS blogging engine, I decided to offload the work of mobile optimization and frontend scripting to WordPress’s pre-built themes. I give them a lot of credit for their ease of set up–aside from some quirks specific to my Nginx configuration, I had the administration panel running well within their five minute benchmark.

I was skeptical of their commenting system, at least in contrast with newer alternatives like Disqus. However, times have changed, and Disqus seems to automatically include “sponsored links” (read: advertising) for free users, which nixes it as an option. I’ve left commenting disabled as a result, though I may revisit that going forward.

Last, mostly out of curiosity and not necessity, I decided to put my entire domain behind Cloudflare’s CDN and DDoS protection, which can be done for free (though with limited configuration options.) It took no more than five minutes to get onboarded and configured, and that was mostly spent waiting for GoDaddy to transfer nameservers.

In total, from launching the server to configuring Cloudflare, the process took no more than ten hours, and much of that was spent tuning configuration options and reading the guts of various documentation pages. For all the managed blogging options out there, self-hosting seems more viable than ever, due to the ease of installing and configuring most requisite plugins.