We recently wrote about the book Into the Sciences by Frederick J. Ross. Our readers will enjoy the following interview with the author. Almost three years back, Fred’s parting message to bioinformatics went viral through hacker news and reddit. So, part of this interview includes discussions about life after saying goodbye to academia, and whether you should follow his footsteps.
Please tell us about what inspired you to write this book.
Fred: I got tired of repeating parts of the material in conversations. I had the model in my head for quite a while before I decided to write it down, and I kept trotting it out when I was trying to explain some aspect of one of the fields I know something about to someone outside of it. Why a practitioner of a particular field would do something a particular way, or why a piece of work was problematic, can be very hard to explain unless youre in the field, and the model gave me a vocabulary to do that. After the umpteenth time of explaining some part of it, I decided to write it down so I could hopefully explain it verbally less. I figured out would be a few pages long and take me a few weeks to get on paper. Two years later…
Who is your intended audience?
Fred: I was aiming for a level around advanced undergraduate or beginning graduate student. I hope it saves someone some time and confusion in that path. But as I wrote it and started talking about the book with people, Ive been more much excited by the reactions from folks on different paths such as professional engineers, lay scientists working on conservation efforts, and high school science teachers. My little sister, who teaches high school biology, has been doing a lot of work around satisfying Common Core standards, and a lot of the concerns shes been grappling with seem to play right into model.
Is this your first book? How was the experience of writing and publishing?
Fred: Its my second book. I also wrote a novel, called Monologue: A Comedy of Telepathy, which you can find on Amazon and the usual places. Writing a novel and writing a technical work were utterly different. By the time I finished writing Monologue I had a clear process for producing fiction which carried me happily through a bunch of short stories and a novella or two. It completely failed when working on Into the Sciences. This was also the first technical writing Ive done where the material was too extensive to fit in my head at one time.
At one point I was reduced to writing at least 500 words a day on some part of the material. I did that for a couple of months until I finally ran out of material to put down. Then I started trying to organize the mess. It got rearranged three more times before I was done. Then I went and filled in all the citations that I half remembered and had referred to. And rearranged. Then I edited the chapters, and sent out a first draft to some people I trust.
It was incomprehensible. Well, unless youre my friend Jeff Gaynor, whos even more of a polymath than I am. By this time I had reached the phase of loathing the manuscript. So I went back and expanded it out. The text tripled in length, got rearranged again, and then I started on another full edit. And rearranged it again in the process. That got me a manuscript. I happen to know how to typeset, so I set up the ebook and print versions myself and put them up for sale.
And then felt utterly bewildered that this enormous project was out of my life.
Please tell our readers about your scientific background and experiences.
Fred: I suppose my first field was really computer science. When I was thirteen, I was grabbed by the local Internet Service Provider, run by an old Burroughs engineer, and put to work. Five years later I had experience with everything from mucking around with Linux kernels to working in Standard ML. Thats paid the bills ever since, to be honest.
I studied physics as an undergraduate at the University of Virginia. I cant say enough good things about how that department treats its undergraduates. They let me blow through my courses as fast as I wanted, and then I spent a couple years working independently with a student of the Landau school, Eugene Kolomeisky. I also joined the HyperCP group in the particle physics lab in October of my first year. Then in my third year I got frustrated with the math in my physics courses not being rigorous, so I went and did a bunch of graduate mathematics. In the end I learned that all the stuff the physicists were brushing over didnt matter. On the other hand, the doctoral course in probability has probably served me better than any other course I took at university.
Then I went to graduate school in biology at the Rockefeller University. Im still not sure how that happened. I grew up in the woods, so I was a fairly decent de facto field biologist, but I had no academic background in biology at all. Still, they admitted me, and taught me what I needed to know. That was a rough transition. I finally ended up in a tuberculosis laboratory. Now, Rockefeller is an interesting place. Everyone there who is not a biologist or a biochemist is called a physicist. You could be an electrical engineer or a pure mathematician and they call you a physicist. The upshot is that youre expected to know any field that involves mathematics. So I found myself doing image processing, statistics, random processes, and all kinds of other stuff way outside of my previous experience. I also went up to Columbia to take graduate electromagnetism, which is when harmonic functions finally came together for me in a deep way. I still have vague plans of writing an epistolary romance novel involving Bessel functions.
Then, two years in, my lab moved to the Ecole Polytechnique Federale de Lausanne in Switzerland. We spent a year waiting for the laboratory to finish being built and then setting it up. Meanwhile I studied thermodynamics and the old literature on antibiotics, which is what my thesis was on. And started getting a steady stream of people coming to me for help with statistics and programming. That served me well a couple years later when I was thrown out of graduate school (probably the best thing that ever happened to me) and rehired as a core facility member. My commute changed by about a hundred feet, my salary increased by 50%, and I got way from my advisor.
I spent a couple very happy years working in the Bioinformatics and Biostatistics Core Facility. Then the Swiss government decided that my visa was ending, and I had to find something else. NIH offered me a position working on the PubMed search engine, but as I was about to sign the contracts Congress put in a hiring freeze. I waited for a couple of months, and then took a job in laboratory medicine at the University of Washington. That was a comfortable environment, and I loved the people, but it was pretty clear that I was at a dead end. I was as senior as they could ever make me in that system, and I was an assistant with no influence on the direction of the work. So I jumped ship to industry.
I spent a couple of years at Splunk building developer tooling and being an internal resource on statistics and data analysis, and then went to my current job at SignalSense, working on cybersecurity, which has tied together a lot of threads for me.
When you left bioinformatics/academia, you wrote a colorful letter criticizing some of the practices that went viral. Please tell us about your life after academia. Are you happy about the decision? What do you advise other PhD students to do?
Fred: I can honestly say that the years since I left academia have been the happiest of my life. I had a happy childhood, and Im still happier. Part of that is that I now have a wife and two small children, which suits my temperament extremely well. Industry has rewarded not only my willingness to learn now material and to do modeling and statistics but my interests in engineering and programming. I am sufficiently skilled that firing me isnt really a credible threat, since I could have another job in a couple of days where I live, so I am much less at the mercy of random and inscrutable forces than I was in academia.
This morning I advised a friend of mine in a PhD program in political science to quit. She wasnt having fun, it was locking her into a miserable place in her life, the one person in her department who was interested in the work she was doing was becoming unavailable for a year, and she didnt need the title for what she wants to do next. Why stay in a job under those conditions? On the other hand, I have friends who would be fools to quit their PhD programs. Theyve got great conditions, get to be near their family, are working on fabulous projects, and are making quick progress.
Part of the problem with PhD training is that it is so uncontrolled. Two students in the same subject in the same year in the same department can have utterly different experiences based on factors totally external to themselves. So Im unwilling to make blanket statements.
You decided to leave professional academia, but your book shows that you did love interdisciplinary research and was highly qualified to think within the context of multiple fields. Do you think your decision is a loss to science, or will you be able to continue science outside the NIH-funded structure better?
Fred: Well, Im doing cybersecurity research now. Is that a loss or have I just moved elsewhere? Im certainly getting to do security work here that simply wouldnt be possible in academia. It took us about ten engineer-years (with really good engineers) to build the infrastructure necessary to do what were doing. I wouldnt even have access to other engineers in academia, much less that much of their time. What were working on requires deep knowledge of software engineering, statistics, networks, and security. Basically no one has all four, so weve been tackling the same problems of interdisciplinary work that I was facing before.
Now, Bell Labs it is not. Its very applied and focused work. Someone more interested in blue sky research would find it quite uncomfortable, but it fits my tastes quite nicely.