Updated Monday, March 9, 2014
I'm a huge fan of The Setup and have always found it helpful to see what tools people in similar fields are using to get work done. This is my setup.
My name is James and I'm a grad student in the UCLA Department of Statistics. I work with my advisor, Rick Paik Schoenberg, on topics relating to geophysics and statistical seismology. Part of our work focuses on developing methods for geophysicist to use that measure how well earthquake data fits to their theories on seismicity. The other part focuses on developing/refining models used in earthquake forecasting.
Outside of my dissertation work, I also work as a graduate researcher for the NSF funded Mobilize Project. We're developing a year-long course called Intro to Data Science and implementing it, currently, in 10 LAUSD high schools. We're also developing 3-6 week mini-courses that fit into high school math & science courses. Our coursework aims to help students develop their sense of data-literacy. To this end, my officemate, Terri, and I have written quite a few labs for students to work through.
The areas of statistics that spark me most are probability models, spatial-temporal methods and most recently non-parametric methods. I'm not a fan of the Oxford comma but I am a supporter of the Gunners!
I've been a MacBook user since I bought the first-generation aluminum MacBook way back in '09. I still use this 13-inch model as my day-to-day weapon of choice, though I've upped the memory to 8gb and had to replace the LCD panel after a pretty hard drop. I find that the smaller form factor is a little easier to lug around everywhere I go.
My laptop (and notes & academic papers) usually travels in one of two bags. When I'm out and about I'll toss it into an Incase Slim backpack. It's padded, it's got tons of storage space for all sorts of computer related accessories without being overly large and I love it. When I'm on my bike, computing to/from my office, my stuff travels in a Topeak Explorer bicycle bag.
Besides my 13 inch MacBook, I have a late 2011 15-inch MacBook Pro with 16gb of ram and a 2.2 GHz Intel Core i7 processor. I keep this laptop stashed at home and connected to a 24-inch ASUS monitor. I find that Apple peripherals are tough to beat for day-to-day use, so I use a wireless Apple keyboard and an Apple Magic Mouse when I'm working from home. This also conveniently leaves my USB ports open for back-up drives or the occasional flash-drive.
The stats dept. has supplied me with a mid 2009 20-inch iMac which was recently upgraded to 8gb of ram in my school office. Connected to this computer is a 27-inch Samsung monitor that I nicked from my friend Josh EmBree's office when he wasn't looking (He works at home from a pair of red gym shorts. I doubt he's even noticed it's missing).
A real benefit to having multiple computers is I can run multiple instances of code without having to setup/maintain/figure out how to do it via servers. The downside is file management, but hey, that's why they invented the cloud, right?
Other hardware includes an Apple iPhone 6 which goes with me everywhere and an arsenal of coffee makers at home.
When I updated to OS X 10.10, I started using most of Apple's built in applications with the notable exception of their e-mail client. Instead, I find that I really like the interface and usability of Mailbox which syncs my e-mail between my phone and computers. I also love the Remind me later feature to clean stuff out of my inbox that I won't care about until some later time.
I've enjoyed the switch back to Safari as well. I'm a huge user of its Reading List feature to keep track of blog posts and other articles I'd like to read but don't necessarily have time for at the moment.
Almost all of my work is done using R and my main interface with R is via RStudio. Using R has become a lot less cumbersome since starting graduate school. I even do much of my data scraping & cleaning using it now thanks to some of the packages by Hadley Wickham & Jeroen Ooms.
For other things, like writing HTML or Python scripts, I use Sublime Text. Using SublimeREPL has made writing Python code almost as easy as writing/running R code. I've tried a few times, unsuccessfully, to integrate my R workflow into Sublime Text but that hasn't happened yet partly due to the fact that I can't seem to find a decent way of integrating R-flavored Markdown into it.
There's a few other editors I use for specialized circumstances. For writing LaTeX I like to use Latexian which is awesome because it auto-compiles your document as you write it. Mou has a similar feature when I'm writing something in Markdown, which I wish more e-mail clients supported.
Other software that has become essential to my work for is Caffeine to keep my computers awake while they run code and f.lux to stop my eyeballs from catching fire when coding late at night. Spotify and more recently Noizio are how I drown out other sounds. Jekyll and Cyberduck help me manage my website and I use git for version control.
On occasion, and usually much to my chagrin, I'll find myself playing old Nintendo games on OpenEMU, my number one way to waste time that needn't be wasted.
Finally, Dropbox solves many of the problems with working from so many computers and RStudio's Projects system also helps keep my projects organized.
I often find myself longingly looking at the latest 13-inch MacBook Pros, but in my dreams I would have a server with a bunch of ram and processors to do all of my code-crunching and a new MacBook to interface with it. I'd also have some sort of charging system that could power my laptop for a few days but would still be light enough to put in my backpacking bag. Accessing wi-fi from the wilderness would also make for a lovely dream.
It'd also be nice to live in a world where data was easily accessible and always came pre-cleaned, but I guess then I'd be out of a hobby.