04 Jun 2015
Every year, when spring comes, I get pretty excited to get outside and do some naturalizing. Well, mostly botanizing. I don’t consider myself a particularly good naturalist. I can’t walk through the forests close to home and spout off the names of all the plants around me - common or binomial. I can name the most prominent ones. I can usually identify the habitat I’m in. But overall, I have a lot improving to do. And each spring brings the promise that I can practice, and get a little better. However, I’ve been a little short on time this spring. I’m trying to wrap up several projects I’ve been working on for my postdoc, because I’ll be ending that sooner than I had expected to take a new position this fall (more on that later). But last week I had the chance to get out for a bit, and I was actually pretty happy with how much I remembered from years past! More importantly, I got out into the northeast (North American) forests, and had time to just look around. I wasn’t running or hiking, just looking, identifying a few plants, refreshing my memory, and thinking about forest structure, associated species, etc.. In years to come, my research focus will be moving back to these forests (though I’ll still be working on CFR projects as well), so it’s good to be reintroduced, and get that spark to open up Flora Novae Angliae and just poke around the pages a bit.
25 May 2015
I moved over all of my old posts from my old Blogger blog. Old posts are tagged in the title with the word (archive). Personally, I have nothing against blogger, and has I wrote elsewhere, I simply wanted to move the blog over to GitHub/Jekyll because I thought it might facilitate more writing. I’ll have to tweak some formatting on the old posts, but most things came over easily enough thanks to the scripts I found here. Another example of the awesomeness that is code sharing. Thanks to coolaj86!
15 May 2015
I’ve written a lot of R code over the course of my two years as a postdoc here at UConn. Or at least, I feel like I’ve written a lot of R code. This is not surprising - after all, I think of myself as a quantitative ecologist of sorts. I got curious about just how much I’ve written a few days ago, so mostly as a way of “productive procrastination” I used a few command line tools to figure out how many lines of code I have currently in my project directories. (For the curious, I mostly just used
wc `find [directory name] -name "*.R"`, with
*.Rmd.) In total I have 23,779 lines of R or Rmd code. Assuming I’ve written this code over the course of 520 to 730 days, that’s about 46 to 33 lines per day. To be fair, the Rmd files contain a lot of lines of text as well. However, this is only what my current directory structure contains, and doesn’t include all of the delations, edits, etc..
I really don’t know if this is a lot or a little. I know it’s about on par with my dissertation work. But I imagine that software engineers blow these numbers out of the water. Actually, I’m sure there are plenty of other ecologists who blow these numbers out of the water. Ultimately, I take two things away from this little diversion. First, personally, I’m happy with the amount of code I’m producing, but there’s a nagging sense that I wish I had more to show for it in terms of publications or R packages. I feel like that’s coming though. Second, you can get a lot done if you think about it from a daily perspective. 30 to 40 lines of good code is pretty tractable to do in a few hours or less. I imagine that if I took the same attitude toward writing 200 - 300 words a day, I’d have a 6000 word manuscript in under a month. Yes, I’m aware this is not at all a novel thought, but it did drive the point home for me.
07 May 2015
At some point while writing a manuscript, I inevitably resort to killing trees for the sake of making progress. That is, I print out a physical copy of my manuscript, and put pen to paper. I like to think that someday I will be able to skip printing out the last couple of drafts, but I have a hard time seeing that happen anytime soon. Perhaps when I buy a bigger screen? When I started graduate school, I would print out what ever papers we had to read for class (and there were plenty of them!) so I could physically underline and write marginal notes. Years of reading physical copies of material with a pen in my hand had me convinced that I wouldn’t retain as much information if I read papers on my computer. But I trained myself to read completely digitally, sparing myself the inconvenience of lugging around stacks, or boxes, of papers when I needed to move offices. Though I should point out that I often still hold a pen in my hand when I read. The transition for writing seems more daunting. One of the biggest reasons I print out my draft manuscripts is so I can lay it all out - it lets me look over several sections at once. This is really critical when I need to move things around. But more fundamentally, there is something about physically touching the pen to paper that stimulates my thinking. Sometimes, just looking at a section and starting to write the first bits of a new sentence will send me back to my computer to restructure a whole paragraph. It just doesn’t feel like that stimulation comes by staring at my computer screen.
05 May 2015
I like doing peer reviews. It makes me feel like an active member of a community of scientists. Over the last year, I’ve probably averaged one a month, which seems like a pretty ok number. (It’s definitely more than I’ve published, for what it’s worth.) I’ve asked around, and it seems that by some standards this is a lot and by others its pretty modest. In any case, it’s what I’ve been doing.
Now, every time I do a review, I go through a little dilemma of whether to sign it or not. I feel like it’s pretty important to be open, and I never write reviews that I would be concerned if the authors did identify me. Not all of the manuscripts I’ve reviewed have been accepted, but to be honest, I can’t recall suggesting that any of them be rejected out right. This may be because I have yet to receive a manuscript to review that I really believed was irreparable, or couldn’t be shaped into a good contribution. Or perhaps it’s because I’m just to nice. Whatever the case, I like to think that I offer good criticisms and comments, and that I help the authors to make a strong paper. But still, there are times that I flinch, and submit without signing. I think as I gain more confidence and knowledge this will fade away, but we’ll have to wait and see.
With all of this in mind, I was intrigued when I recently got an email about a service called Publons. Essentially, Publons is a way to make public your involvement in the peer review process. That doesn’t necessarily mean you have to make public the content of your reviews, though that is an option, but it is a way to show that you’ve done reviews for various publications. If you do elect to make public the content of your review, or if you submit a post-publication review, it seems that there is a way for others to cite your contribution as well. I was sufficiently intrigued to make an account and go through the process of adding a few reviews, though not to make them open at this point. A have few random thoughts on this kicking around in my head:
- It seems really promising to have a formal way to see open reviews (sensu what is done at PeerJ and soon PLoS ONE).
- I’m curious to see if post-publication reviews take off. When I was at Stony Brook I really enjoyed following Faculty of 1000, so I think it could be great.
- I can’t help but wonder what the business model for Publons is? How are they going to stay afloat and pay bills? I don’t really see them having a lot of valuable information to sell to marketing firms.
- Oh my! How do some people have time to do 1000s of reviews a year! That’s crazy.