A minor note to referees

I frequently receive referee reports which reference particular lines of my document as “Page 13, line 20 …” Of course, I appreciate all of the careful reading. But this strikes me as a very useless way to refer to a location in my file.

It is extremely unlikely that, after responding to the referees other more general comments, page 13 will still have the same line breaks it did when I submitted it. I do have the original submitted file (advice to young authors — save a copy of the original submission!), so I can figure out where it was, but I then have to locate where the corresponding line is in my modified file and find the corresponding line to edit. It would be a lot more convenient for me if referees gave me a few words from the line they are referring to, so I could directly search the file for them. I try to do this; if you are a referee, I encourage you to do the same.

12 thoughts on “A minor note to referees

  1. Another thing I do, also when implementing my own edits (I’ll print out a draft of a paper I’m writing, mark it all up, and then have to implement the changes on the electronic version) is to implement all edits from last to first. This _usually_ makes for no line/page changes.

  2. I agree that extra context (in the form of surrounding text) should be given, but how else are we supposed to do things short of annotating the PDF or a hard copy?

    (Grumbly ex-copy-editor, even more grumbly referee.)

  3. Yemon- Why not annotate a PDF? There’s lots of software for things like that now. I’d use my tablet pen, but I fear my handwriting would be too recognizable.

    Dylan- Because editors never send us the TeX source? I’m sure that there is some very clever way of doing this, and if we ever switch, I’m sure we’ll wonder why we were so dumb for so long. But inertia is a powerful thing.

  4. I’m currently too tired to respond cogently (or coherently) but something about Sweep Them Into The Dustbin Of History tendencies doesn’t chime with me – especially when it comes to things connected with publishing. But then I have to admit I am mentally if not physically from a previous generation, and for me there is no substitute for getting a printed hard copy and taking a red pen to it, with a visual rather than semantic approach to some of the editing if not the reading.

  5. I also don’t really see the problems with page & linerefs as a quick way to signal the desired changes…

    (Perhaps you guys all read and write a better standard of paper than what I see or do, but trying to edit my TeX source would be a nightmare for the referee/editor and an embarrassment for me, and I don’t think I’m alone.)

  6. I’d much rather have the page and line numbers than a vague reference that doesn’t give me enough context to find whatever it is. And I worry that with other formats, such as annotated pdfs, I might easily miss some of the comments, or find that the annotations are incompatible with my local setup. But your warning to keep a copy of the version as it was submitted is good advice.

  7. Dylan @ 3: I’ve put my referees comments directly into the TeX source before, in the form of footnotes. I got the source files from the arXiv, and I didn’t modify the text itself. I thought it worked well for a paper where there were no extended comments.

    David @ 7: I’ve used annotated PDFs in communicating comments to my graduate students, through not for refereeing. As long as you can extract a plain PDF file, there shouldn’t be an platform issues. You’re right in thinking that some programs store the comments in meta-data of some kind or their own custom format. E.g. with Skim on OS X you want need to do Print and then Save as PDF to get a vanilla PDF.

    Another thing one can do is mark things up with pen/pencil and then scan it — my department’s photocopiers do a good job of this.

    I used to use the “lineno” LaTeX package to print line numbers next to each line to ease the refereeing process, but now that submission so often involves pointing the journal to the arXiv I’ve basically stopped. (Even if one does this, sometime the referee will just use the arXiv version instead…)

  8. I’m with Yemon on this. I edit on with pen on printed paper copy, and for referee’s comments of the kind that does not take substantial new writing to address, I prefer to mark up a paper copy before typing it in. When refereeing a paper (or seriously reading it, as opposed to just looking up a statement of a theorem), I have to print out a copy and mark it up.

  9. Being brought up on ancient pen and paper methods, I have refereed hundreds of papers using references to pages and lines in my printout (whether mailed to me in the old days or printed by me from a PDF file more recently). Actually, I would always prefer to point to sections, subsections, or such, but many papers are not tightly enough organized to allow for that. There now tend to be arXiv postings, but these are tricky when they don’t quite agree with the version sent directly by an editor. And I’m not inclined anyway to prowl around in the often messy source files which typically employ long lists of private TeX abbreviations.

    Aside from technicalities, I’ve always preferred to go through a printed version with a red pen in hand, making and then sometimes unmaking marginal corrections or comments before deciding what to put into a formal report to an editor (usually with appended corrections and such intended for the authors as well as editor to see). And whatever the methodology, it’s more important that a referee report be fair and informative to all parties concerned while tailored appropriately to the journal or conference proceeding involved.

  10. I’m sure this is a slightly obvious thing to point out, but surely just going through a referee’s comments in *reverse* order removes any problem of the form “after responding to the referees other more general comments, page 13 will still have the same line breaks it did when I submitted it” since any amendment will only change parts of the file that you have already looked at.

  11. Easy, instead of page&line refs, give Section and Paragraph number. These are easy to find in the TeX, won’t change much (or are easy to track down if they do) and don’t depend on the formatting.

Comments are closed.