Revision history of a scientific manuscript

neanderthal_progress

We just published a paper on the nonlinear dynamics of Neanderthal extinction in this month’s PNAS. Once we started writing the manuscript, I started keeping all of the files under what I’ll carefully call “manual version control” (…we copied and renamed each revision). This makes it easy to use Python’s difflib to track the changes in the paper from one iteration to the next. The graph above shows changes that brought the manuscript closer to the final manuscript version, with some of the major events labelled. You can see that, while the reviews caused us to rewrite many portions of the final paper (about 40% of all of the characters in the file were touched), these edits were concentrated in two bursts: the edits to get the paper accepted, and then an equal number of edits in the proofs stage (the last data point). A slightly more analog version of this data set looks like this:

IMG_2419

Eight months of paper drafts of our manuscript, ordered from left to right. Penny for scale.

Another interesting question might be the degree to which revisions altered the text, replaced the text, or reverted previous changes. The graph below shows changes in the number of characters from each revision to the next (this is sort of like taking the time derivative of the graph above). By comparing the red and blue curves, it can be seen that most long-term oscillations involved us adding and removing entire sections to the manuscript, but shorter term oscillations (like the zig-zags in July) can be attributed to us adding and then removing the same set of text.

manuscript_history_1_01.png

The code I used to parse the file names, analyze the LaTeX source files, and extract the dates is all here on GitHub. I added axes labels and annotations using Adobe Illustrator.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s