The last paper response

Well it’s been a few weeks since I got the response back for “last paper” since I have roughly three weeks left to finalize all the edits they requested, this weekend I plan on tackling most/all of it so that we can get the ball rolling with my collaborators so I can (hopefully) get the response back to the editors before the deadline they set. Because there is a deadline and I do not want to be late, mostly because it will make more work for me if I am.

Quick recap for those who are new or don’t remember. “Last paper” is called last paper because it was the final paper I had to do in my long list of backed up papers. In total I had four first author papers I was working on… for years. It wasn’t four at once, it was one, then two, etc. Last paper was the last, but hardest on that list and finally the dam broke, papers started going out, and now last paper is the final paper out of the four to be published. It was a stressful time, I don’t recommend it, but we’re through it now and even though “robot paper” (this one) is still not open access yet (sooooooon), I’m happy that three of the four papers are done and out of my hair.

So now that we know why last paper is called such, let’s talk about where we are now. Last paper took years to do, it was an intensive and difficult dataset. I collected the data for this experiment solo. Then I wrote completely custom code, from scratch, then rewrote it, again and again. Performed analyses that I had never done before, wrote custom code for that as well. Did things that no one in our lab had done, then finally wrote the paper. It was a long process, but we submitted it back in October(ish?) I think… I’ve completely lost track at this point.

A few weeks ago we got word that the paper was close, but they asked for major revisions. That sounds back, but it’s not. It’s part of the process and I’ve never had a paper where they requested minor revisions the first round. Bad would’ve been desk rejection or reviewed rejection, the first wouldn’t be too bad, the second would be embarrassing (both happened with robot paper multiple times). Since that initial decision I’ve been incredibly busy with other work, so it’s just sat there.

I was hoping to have the revisions done last week, but with the other new first author paper I’m working on (we’re scientists, we’re always writing something), the news from DARPA (here), and my first dissertation data collection (here), it’s been a little busy. Welcome to research life, you get used to it scary enough. For those who haven’t published or don’t want to publish a manuscript I’ll explain what happens now.

From the beginning, we submitted our manuscript, the editor sent it out to reviewers (unpaid… sad face) to look over the manuscript and provide feedback. Normally you’ll have anywhere from two to four reviewers, at least in my field. We had two this round, both liked the paper, but had suggestions on how to make it better. They put those suggestions in the form of a letter, sort of. It’s hard to describe. They summarize what the work was about (showing they actually read it), explain how well it fits with the journal, the advances that the manuscript makes, and then critiques of the paper itself. That takes roughly one to two paragraphs if the reviewers are through, I’ve had single sentence reviews before though.

They also give “specific” feedback. Think line number and what’s wrong. My job is to go through all that feedback and either (1) make the changes accordingly, or (2) explain why the reviewer was wrong or why the change shouldn’t be made. Option 1 is always the easier and better option. Fighting with reviewers doesn’t end well, mostly because they are the final say, so you can argue all you want, it just won’t do anything but piss off the reviewer and make your life harder.

Out of all the suggestions I’ve ever gotten from reviewers, we’ve never gone with option 2 even though one reviewer was clearly trying to pad their publications by forcing us to cite several of their works (some of which didn’t even remotely apply). Which to me feels slimy, so I wanted to fight it or cite other similar works just out of principle.

My job now is to make the changes to my manuscript, typically we will change the font color to red to further highlight those changes, then provide a response letter. The response letter is literally copy and paste their critiques and offer a point by point response with either option 1 or option 2 (always 1). Normally we will add the exact change we make, so if I add a additional text I will thank the reviewer, then say line XXX now reads “… this is the additional text.” Or something along those lines. We always thank the reviewers because again, they are unpaid. The system is shitty, they get nothing for reviewing other than a pat on the head (almost literally).

So that’s not so bad and I can do that in the course of a few days, sometimes not even a full day depending on how through the reviews are. We had awesome reviewers this time so we have a lot of points they brought up. I prefer through reviews because it makes the final product better. I can read a paper 100+ times, but it will never get any better with just my input because I know the work better than anyone since I did it. Stuff that makes sense to me may be confusing to others.

A good example of this is some custom videos I made from our data. They are super cool, I’m really proud of them, they are one of a kind, but the reviewers were confused. Specifically about the time scale and a few other things that made me realize that things needed to be added to them to provide further context that I knew so it was obvious to me, but it was not obvious to them. Luckily I have the code, the data and everything I need to remake them semi-fast so not a huge deal.

Some of the other suggestions will take more time, things they would like to see like additional experiments won’t be done, but we’ll justify what we did and suggest that future experiments take their points into consideration. All this won’t be too difficult to address, but the problem is it’s not just me that needs to sign off on the work.

I have several collaborators that need to look over everything before we submit it for the next round. They need to add, modify, or flat out reject my changes before I can send it off. Their names are on the work too, so they need to have a say in how we make the changes and that takes time. In some cases a lot of time. Like me they are all very busy, so I need to be courteous and provide a few weeks to let them read over it.

Once we’re all in agreement, I will incorporate their suggestions, and send it off for the next round of reviews. If I miss the deadline, they will treat it as a first submission, meaning new reviewers, and far more work for us because every reviewer will have different feedback and preferences. However, if I make my deadline, the same reviewers will look over our changes and either approve our manuscript for publication, or they will request further changes. I’ve had it go both ways, but if they ask for another round of edits, it usually isn’t too terrible. I’ve ever (yet anyway) had it go more than two rounds of edits before we publish and in most cases one round is enough.

So that’s the whole process. Once it’s accepted we go through typesetting, which is some work and a whole other post. Which I’ll share eventually, probably once this gets accepted. For the next couple of days however, I’m going to be somewhat relaxing and making the requested changes. With some luck, Monday I can pass along my suggested changes to my collaborators and get feedback so we can get this paper published!

Finally.