Listserve to Blogger...

Wow.

I'd had this blog sitting idle for as long as I had my Google account -- the big G tacks on a Blogger account automatically once you set up a credential with them -- and I kid you not, last Friday night I was mulling over using it to track my field change. Sunday morning I had an invitation to post on the Listserve. (If that ain't kismet, I don't know what is.) I spent a day deciding just what I wanted to say, and I figured that I could possibly get some eyes and opinions on my (evil?) master plan by telling my story on the list. As far as the timing of these two independent events go, I'm pretty pleased. (For the hat trick, tomorrow's my birthday. Happy quirks like this always give me a smile.)

What I posted to the list read as follows:

I just graduated in May with a MS in biology, focused on animal physiology and biomechanics. I learned many things about myself during my graduate education; in particular, I learned that I truly don't want to continue in biology. Suffice to say, killing my research animals made quite an impression.
Throughout my studies, I had a growing interest in pointing my biomechanics knowledge less toward animal systems and more toward robotics. My interests grew until finally I broke biology ranks and took an introductory robotics course through the computer science department at my university the semester before I graduated. I didn't have the background everyone else in the room did, and I spent a lot of time outside of class catching myself up on unfamiliar material. I loved every second of it.
My new goal is to develop this interest to the point that I can pursue a graduate degree in robotics and make contributions in that field. Since I'm now out of school, a lot of this development is must be done informally. I'm documenting the process on my blog, robotfrogblog.blogspot.com. I'm hoping to use this record as a way to keep myself motivated, to follow my progress, and especially once I have something concrete in my new graduate school search, to set an example for anyone else out there who finds themselves wanting to re-chart their course. As I've been told, it's never too late to start over... or to change.
Cheers!

Now, what exactly do I intend to do with any of this?

I readily acknowledge that what I'm trying to do isn't something that most people going into robotics would do. Not long before my final year started, I sat down to ask one of the CS faculty about how I might best go into robotics, and his advice was to finish a PhD in biology and to collaborate with someone in robotics. Again, that's well and good for someone who wants to stay in biology. I don't plan on raising and destroying another cohort of animals any time soon, so my immediate concern is finding another way.

That aversion to killing research animals is a double-edged sword. On the one hand, it's quite motivating -- I never, ever want to go through a euthanasia day again. On the other, it's a booster rocket at best. I can't let that issue alone guide me the whole way, because it doesn't have anywhere near the power to keep me going convincingly through the application process for a degree, much less a program. I need to move on to some concrete interests in the field itself, based on learning as much as I can however I can, and then -- and only then -- will I be able to start exploring the next stage, which is getting my new degree and really, when I get down to it, a new life.

The way I see it, I have a phenomenal amount of catch-up learning to do before I ever express interest in a lab anywhere. In particular, the following general deficiencies come to mind:

• Mathematics. To pursue biology, I needed to have Calculus I and Statistics... and really, that's it. When I took Introduction to Robotics, I spent a lot of time just teaching myself the ins and outs of linear algebra enough to get by on the homework assignments. While I did manage to keep my head above water, I don't want to try functioning flat-footed in a similar situation again. My undergraduate mathematics experience was largely of the "plug-and-chug" variety, and while that may have been the expedient approach favored by some, now that I'm working through a more detailed teaching text, I can't help but think that even my calculus needs some significant shoring up. Beyond that, I need to learn a lot of mathematics before most robotics texts make sense.

• Computer science. To put it bluntly, I don't have too much of anything here. I know my way around LabVIEW, and to a very limited extent, MATLAB. That's less science than science tools, though, and they won't get me very far themselves. I need to get myself through oh, so very much more... algorithms, data structures, operating systems... I keep feeling as though I essentially need to learn a full CS degree's worth of material. Any best books out there that people would recommend, and in what order?

• Fundamentals of robotics. The intro class covered a considerable portion of Spong et al.'s Robot Modeling and Control, but not all of it. For me, a lot of that material isn't accessible yet because I don't have the math and/or CS background to work through it properly. There's not much I can do until I nail all of these down.

• Current research. This is a biggie. I don't know the field much beyond what I read in highlight blogs, because I'm not into the literature very much yet. Without an understanding of current research in the field, I can't say anything about what I might want to do, much less in what lab I might want to do it. I don't even know enough to produce a credible guess. I need to build up the previous three ultimately to be able to make headway here.

Answering these deficiencies depends heavily on resource availability, of course.

• Courses. If I work at a university, I can take a limited number of credits for free each semester. That's no small thing, considering that a single math course can be four credits, and nowadays, credit hours ain't cheap. (Community colleges around here don't tend to offer the math classes involved, so that's out.) Coursera, Udacity, et al. are very useful for learning the material, but I'm not quite confident yet in the acceptance of their credentials in applying for something else. The ideal would be to get the courses at a brick-and-mortar institution, with an official academic transcript to attest. 

• Self-study. This is how I'm meeting my mathematics needs right now. I'm working through Gilbert Strang's Calculus, second edition, and I enjoy it immensely. I could conceivably answer all my math and basic CS needs this way, but the problem is finding a way to convert that learning into an acceptable credential. At this point I figure that I may end up taking the GRE subject tests in math and CS so I have some objective evaluation of my knowledge available to present.

• Library resources. I need to read, read, read, read, read. There's no way around it. However I can go about using campus library resources to borrow books or read articles, I need to do so. Again, this becomes considerably easier with a university job. 

• Hands-on experience. I have here a (supposedly) crippled Roomba, given to me by my advisor before I became quite so blatant about wanting to pursue robotics full-time. It's a Discovery model, which I understand is reprogrammable to other tasks. Such experience would be invaluable to me. Additionally, I have an Arduino that I'd like to apply to some robotics projects. There are other things I wouldn't mind learning along this whole process, such as how to assemble and operate a RepRap machine of my own and how to use ROS (maybe, hopefully on my Roomba). I'm operating under the assumption that I'm on my own for this, because I don't really have the qualifications to get a robotics job right now.

While I'm writing all of this, I keep thinking all of my plan is just pie in the sky. I mean, who the hell has ever done anything like this? My answer is that I just need to keep my head down and stick to my plan, even though it may well look like trying to bite a basketball. (As Linda Ellerbee once noted, it's absolutely possible to bite a basketball, so long as you know where to start.) The bones are there -- now they just need some localized remodeling and fleshing out. Individual items, like the ordering of mathematics or CS subjects I pursue, might be fluid and flexible, but the ultimate pathway feels pretty well set.

This blog is here, as noted above in my Listserve note, as a means of tracking my progress. As I go, I'll be posting what I'm doing here. I'm hoping that as I develop my knowledge, I'll have more useful things to say. (In particular, I'm looking forward to writing up how I do my hands-on work.) Those items may be useful to someone else trying to do the same proximate tasks, but what I would love for this blog to do is prove helpful to someone else who might want to transfer professionally from one STEM field to another. By the time all is said and done, I want to be able to say that I did make that type of transition successfully, and I want to be able to tell someone else that they have the power to do it, too. I still believe that it's never too late.

It all starts with a plan.

"It's me plan! I plann'd it!"

--Dave Lister, Red Dwarf: The End