I have an author who asked for the Word doc, and when we said no, you need to work with the PDF, emailed back with "Good news! I converted the PDF to a Word document! I will re-convert it back to a PDF before I submit it!"
"Good news, everyone!" Getting changes in the Word doc is a freaking nightmare. It's right there in the instructions; don't do that.
Stephs author is being crazy, of course, but I will say that it is frustrating to do anything other than the smallest editing in the pdf when your bibliographic program and figures are native to Word. It would be better if authors and editors could collaborate on the lossless version of the manuscript.
By the time the paper gets to me, it shouldn't need more than the smallest editing from the author. It would be better if the authors would finish writing the manuscript before they submit it.
I'll tell another one. Some journals will not accept figures generated in Excel, because they are not high enough resolution. However, if you print out the Excel figure and then scan it back in at a (meaningless) high resolution, the figure is acceptable to the journal. When faced with the prospect of learning a new graphics program to finish a paper they long since stopped caring about, people fire up the printer and git er done.
Yep. Git er done. That happens because it's an automated process accepting the figure; it's just checking the resolution. If the resolution is too low, it'll look bad and the author will complain bitterly about it and insist that I redo it, as if I can give him perfect sharp figures when he sent me blurry crap he created at 96 dpi. That's why we insist on 600 dpi from the author. However, if the author sends me a figure that's still blurry at 600 dpi, I have to assume that he doesn't give a shit if it's blurry. I do care that the figure looks bad, but I can't force him to give me a better figure. The resolution is more important with complex figures than line graphs, I guess, but the automatic process can't tell the difference.
One trick I've used to cheat with high resolution is to save an Excel-generated graph as a PDF and then output the PDF from Acrobat as a .tiff with a high resolution.
This is what we usually do, and what I usually suggest to my authors. It's not "subversive", it's a "work-around".
I am back at work from a week's vacation today, and I don't want to be.