A self-indulgent look at this blog

I was flipping through the stats Blogger automatically keeps on the blog, and it made me feel really proud of the writing I do here. This blog has probably been read more than anything else I have ever written, which is an odd feeling, considering that few people know that I do it. My most read posts have almost 28,000 views between them.

Two of the three most popular posts are about non-academic jobs (my first post on this, and my link aggregator page).  The third is a Mendeley review from 2011, which is still fairly popular (and actually still reasonably representative of my thoughts on Mendeley, which I still use 5 years later). Rounding out the top 5 are posts on how I got my National Lab job (hint: don't expect what I did to work for you--it was luck!) and how search committees sift through applications (still true). After my top 5, views per post drop pretty steeply (50% drop between #5 and #6, for example). Interestingly, my posts on job searching in various forms are more popular than anything else. After that, it is actually academic misconduct posts which get a lot of views, followed by more detailed interviewing advice in the top 10. From that, I would guess that most of my audience consists of students and/or postdocs.

The most likely path to my blog is through Google, which sends 5X the traffic of the next most common entry point, which is Xykademiqz's old blog (Thanks!).

Most of my audience looks like it comes from the US, with Switzerland a surprising second (at least surprising to me, since I write in English about issues primarily of interest to North American readers). Only 58% of my audience is using Windows, with 26% using Macs. I wonder how that has changed over time. My readers primarily use Firefox and Chrome, which is not too surprising.

The blog averaged about 2300 page views per month while I was on hiatus, which is just astounding. I pretty much started this blog (and keep it going) for my own entertainment. I tend to do advice posts and commentary, since I would have liked to see that sort of thing when I was struggling with the issues I discuss. It is actually really cool and quite surprising to me that so many other people have found it interesting/helpful.

Interdisciplinary science

This study in Nature looking at grant success rates for the Australian Research Council's Discovery Programme over a 5 year period confirms with data what has long been suspected--interdisciplinary projects are less likely to be funded, and the effect is stronger the more interdisciplinary the proposal is. This is an issue not only for those of us who do a lot of interdisciplinary science (who of course want to be funded), but also for science in general, since more and more modern science is at the interface between disciplines.

Anecdotally, what is true for proposal evaluation also appears to be true (for me at least) in getting manuscripts published. My more interdisciplinary work needs to be shopped around a bit at different journals, sometimes 3 or 4 times before even going out for review. In contrast, my work that fits into a "traditional" discipline is apparently easier to match to a journal, since it tends to go out for review straight away (even if it isn't accepted at the first journal).

It is much harder to find appropriate reviewers for interdisciplinary work--I often end up recommending a list half in one field, half in another. Even so, referee reports often come back with serious misconceptions about the parts of the manuscript that are obvious out of the referee's area of expertise. The system of using 2 referees means that if I am lucky, I will get referees with 2 different areas of expertise. Alas, more often both reports are from reviewers in the same general "traditional" area, who then either ignore out of area issues, or don't appreciate the novelty, difficulty, or significance of the out of discipline results. It is hard to know how much of this is caused by my issues (writing style, not a strong enough introduction, lack of clarity, too much in love with the data, etc) and how much is lack of core knowledge in the referee at times.

If this is the case for manuscripts, it is almost certainly also the case for proposal evaluation. Moreover, I suspect that interdisciplinary proposals have a harder time attracting a strong advocate who can sell the research to the rest of the panel. Having served on panels myself, if a proposal does not attract a champion, it can be easily overlooked even if the science is top notch and the writing is clear.

Unfortunately for me, the problems I am interested in and the methods I use to solve them are highly interdisciplinary in nature. I often collaborate with colleagues in other subfields and departments. A long time ago, I realized that many people would be extremely skeptical of some aspects of my work. I am confident that the research I do is exciting and important--the problem is in getting others to see it, of course. To counter out of hand rejection of some of our admittedly very unusual research combinations, a good cover letter is crucial. I also find that it is extremely important to attend conferences and have student attend conferences so that potential reviewers see the work presented BEFORE they get a manuscript or proposal to review. Hopefully, that little bit of familiarity helps establish enough benefit of the doubt for people to read the work with an open mind, rather than dismissing it out of hand. With the increasing number of manuscripts and proposals submitted, there is less and less time to consider a manuscript/proposal, so snap judgements are important. As an aside, a side benefit of doing unusual research is that we don't worry much about being scooped on our very interdisciplinary stuff no matter how much we talk about it prior to publication.

I am not sure what to do about the problem of evaluating interdisciplinary work, but this will be an ongoing problem for the scientific community. Anyone else have strategies for helping others appreciate the novelty/beauty/significance of their work?

The importance of writing

Why is it that no one ever clearly articulates to young scientists that being able to write clearly and quickly is a key skill? I spend probably half of my time now writing (sometimes more!) and I have never been more grateful for a liberal arts background that included lots of writing (scientific and otherwise).

Of course, like anything else, you only get better with practice. My own first drafts tend to be full of long, difficult to parse sentences that need to be shortened and simplified in subsequent revisions. After years of experience, I know this about myself and can set up my workflow to accomodate. I am trying to get my students to write more so they can figure out their own writing habits BEFORE they have major time pressure on their writing. I am surprised by how much resistance there is to simply getting words on the page. All of them would much, much rather set up Powerpoint summaries than write things up more formally, and don't see why they should change this.

Quickie Mendeley review update

For anyone following my Mendeley experiences, I just updated to the latest version, and the integration with MS Word is much, much nicer. There is now a hotkey to use (instead of just the toolbar), you can search in a toolbar that pops up within Word, and it is even easier to add multiple citations at once. I am revising a paper this week, and I am much happier with this new version.

Choosing a journal when writing a paper

We are about to start writing a manuscript in an area that represents a change in direction from all of my prior work. My usual journals are not really a good fit for this work, so I am thinking hard about where to send it. Even in the age of Google Scholar and Web of Science, I think journal selection is important (especially for those who are TT and earlier in their careers).

At National Lab, where bean counting was becoming more and more important, the impact factor trumped all. All raises and promotions had to be justified to non-scientists and/or former researchers in a wide array of fields. To make things easy, they would rank papers by impact factor, regardless of field. This is hugely distorting, since a great specialty journal in physics can have a smaller impact factor than a decent specialty journal in biomedical research (just due to the number of researchers in the sub-field). To get ahead, many people would find the journal with the highest impact factor that would publish the research.

My preferred style is to find a journal where my target audience is likely to read it. So if I am doing physics-type measurements on a protein, I would prefer to put it in a physics-type journal (where others might pick up on it and find it useful) rather than a protein-type journal where most readers could care less about this physics-type measurement, even if the impact factor is lower. Even though my bean-counting impact might be lower, I am counting on the actual impact being higher. What is the point of a high-impact publication if no one reads it? That said, when DrugMonkeyasked people about their citation practices, many commenters said they prefer to cite papers from higher impact journals, so clearly impact factor must be balanced with target audience.

This type of selection is much easier when I am intimately familiar with the sub-field. In that case, I know well which journals I read and cite, and can select accordingly. In this particular case, I don't have any knowledgeable colleagues or collaborators I can talk to, but that is generally my first stop in deciding where. How do you all pick where to submit your work (especially work in a new area for you)?

Long delayed Mendeley review

Way back in August, I started playing around with Mendeley after getting annoyed with Endnote. I said that I would write up a review after I had been using it a while, and I think I have enough experience with it to compare it to Endnote.

There are many things I really like about Mendeley. I really like reading papers in it, since I can save notes about the paper that I can search and share if so desired. Searching inside pdf files also works very well, and it is much, much easier for me to use Mendeley than to try to remember which paper I was thinking of and then open it in Adobe Reader. I never really got into Endnote's pdf features--the pdf search and import features were buggy and irritating when first introduced, so I never got into the habit of using them.

The pdf import feature in Mendeley works very well for me. I was able to convert over most of my library directly from my pdf files. This also works well with my new literature review system. Now that I am trying to use RSS to keep up, it is very easy for me to download pdfs into a directory that Mendeley watches. I also use Web of Science for searches and download pdfs through there. This is a huge improvement from my old system of exporting Endnote data, then importing the reference, since bibliographic imports happen automagically. Correcting errors in imported files is really easy, and can be done manually or via Google Scholar search. I spent a bunch of time initially correcting entries, but now I don't really have to. There is a pretty decent time investment in importing hundreds of pdf files at once, especially if you want to be able to properly search by author (since the author names MUST be imported properly for this to work). I did not try importing my Endnote library, but that is another option.

There are a few downsides I've found to Mendeley. I find the cite while you write interface a little clunky. It is not unusably so, and it is really easy to find the paper I want to cite using Mendeley's search functions. I thought the same about Endnote's interface, so this wasn't a major downgrade for me. One minor quibble--it is not possible to do unrestricted edits in the bibliography generated by Mendeley. This is something I have only just started playing with, so I am not sure what exactly I did to cause Mendeley to complain about it.

One big issue I discovered is that Mendeley and Endnote DID NOT play well together in one document I was working with. The references got corrupted and I had to strip out all the references and redo it all using just Endnote (since that is what my collaborator uses), which was really annoying. This is hopefully just a minor glitch for me, but may become a problem when I am writing with collaborators who still use Endnote. I haven't had this problem in some of my own documents (updating old proposals, for example). I plan to avoid on using two reference management systems in the same document as much as possible, but be careful if you do!

Mendeley was very disappointing in one area, and that is in keeping a library synced between two computers. I used to use a laptop and a desktop, and I tried to use Mendeley to sync my library between them, but this caused me lots of grief. I had lots of duplicates in my library, and Mendeley sometimes lost track of where the pdf files were on my laptop. I have solved this problem for myself by switching to using one computer (a laptop with a dock), but this seems like a feature in Mendeley that needs a lot of work, but could be very useful.

I don't really use the online features of Mendeley much, but in theory, you can grab citations while surfing and also have access to pdfs through the Web interface. I am not sure how well this works, since I always download pdfs when I want to cite and/or read something, but this is something that might be useful to people who use Google Scholar for searching and/or don't always work on the computer where their pdf library resides.

In short, I am keeping Mendeley. That it is free is only icing on the cake!

UPDATE: Thoughts on the latest update here.