Why does or did Elsevier make PDFs grey and therefore unfit for printing?

Question

My university subscribes to various online journals. Sometimes I print off articles that I know I will read over and over again. (I'm all for saving paper, but some things I do prefer reading on paper. I print on recycled paper and I bind the articles, so they will last a lifetime.)

Now I tried to print an article from Advances in Mathematics (published by Elsevier) and it took me a while to figure out why part of the text was blurry: The PDF has seemingly random snippets that are not black (#000000), but in a dark grey (#231F20). On screen the difference is hardly noticeable

especially if it is text:

However, when trying to print it, the printer is trying to make it less than black, which results in blurry text:

As most other publishers manage to produce pure black PDFs, I must assume that this is intentional (it's not just "all italic text" or "all math mode symbols", the colour changes mid-sentence). The question is just why?

I doubt that this is an anti-piracy measure: surely there must be a less intrusive way of marking a PDF file digitally to record the IP from which a particular copy was downloaded. (Some journals add text like "downloaded from [this IP], [university name]", which I don't like, but at least the text itself is not blurry.)

I also doubt that it is to promote their offprint sales: Elsevier sells offprints of their articles, but only 25 copies or more and only to the original authors of the article (or someone acting on their behalf). In other words, it's not to "encourage" me to buy "quality" offprints of a single article of which I am not the author.

(Of course, it can't be to encourage purchasing of physical volumes/issues of the journal. It is entirely unreasonable to assume that researchers buy an actual volume/issue of the journal just for one article, if only because a single issue contains so many unrelated articles and researchers typically live in small offices.)

What benefit do publishers have from mutilating their PDF files?

Making a printable (= pure black) PDF would also have benefits: I'd be convinced that Elsevier is a great publishing company, publishing high-quality research papers in a good format for everyone's benefit and, as an author, would be more likely to consider publishing in Advances in Mathematics.

This also makes me wonder:

1. Just to check, I am allowed to print PDF files (for use in research) of online journals, right?
2. Am I allowed to open the PDF in some other software, modifying the file so that it prints in pure black?
3. Am I allowed to use another method of getting a pure black printout of the PDF (e.g. by loading a special printer colour profile that treats the dark grey as black¹)?

_{¹ Unfortunately, I haven't found out if this is actually possible.}

Note. I have not tried to investigate this issue systematically, but I observed this in 3 articles from Advances in Mathematics (Elsevier) from around 2005–2006.

Answer 1

This often happens when a document uses the CMYK color space and the black is set as (0,0,0,100).

When you go print in a monochrome environment, the document's color information is converted to grayscale first. Because black ink on white paper can't actually create gray, a halftoning process is applied, where the shades of gray are hinted at by using a circular-like pattern.

Because CMYK (0,0,0,100) black isn't seen as being as black as it gets, in the conversion process, it gets turned into something more like 90% dark gray, and the halftone pattern appears.

Probably somewhere along the production process, someone got their blacks mixed up, and if it's consistent with all their PDFs it could be in the scripted part of the process, there's an issue either between color spaces (CMYK vs RGB) or a conflict between the source documents and the imported parts.

Answer 2

guifa is correct in their analysis of why this happens, but I would like to note that there is often an easy fix available once you know what is going on and so long as the PDF is saved as text rather than as a scanned image.

When you go to print from the PDF, click Properties from the Print menu, and on the advanced tab hopefully you'll see something like this:

Under your printer you'll hopefully see Document Options -> Printer Features. If so, you may have a "Print All Text as Black" option (hopefully you do), and this is Disabled by default. Turn it to Enabled and try printing an example of the dotty-page you were getting before.

If the page is actually text and not just a scanned image, you will now hopefully enjoy a print of pure monochrome, crisp black text. This problem is quite common especially with older monochrome laser printers, as they attempted to accurately print the "less than 100% pure black" using the only tool it had available - black toner. With this option enabled your smarter monochrome printers will understand that you don't want an accurate portrayal of the text - you want a readable one!

This option will cause a problem for text effects that use a gradient, but honestly outside of the print-proofing world I've always turned this feature to on by default to produce nicer text prints.

Answer 3

It may be deliberate redistribution control. I recently did work on an eLearning system for a university where the content provider wanted visible per-student watermarking as part of the licensing process. Given the permissible tools (maintenance overhead), document sizes, and the number of students, it wasn't possible to produce a workable solution (too resource intensive when on-demand, too storage intensive to pre-calculate).

In your case, the publisher may be using an old technique for protecting against a lecturer printing one copy, then photocopying (which historically was faster than multiple prints and enjoyed analogue degradation). That's reminiscent of some copy protection mechanisms used in the 8-bit home computer/monochrome copier days, which would need a word typed in from a printed manual. By chosing colours that had very close luminance values, a photocopy became fairly useless making pirated software useless.

Individual academics and their supply chain can be slow to update their thinking - and indeed, since mutual trust is largely respected (both parties trading on reputation), there's no real driver either.

It's equally possible that some of the source material for the PDF came from a low-quality scan of typed or printed materials (or a high-quality scan of a low-quality source).

Answer 4

If the text is changing fonts mid sentence it's more likely to be deliberate. You could convert the PDF to PostScript and then look at it in a text editor to see if there's a systematic and unnecessary change in the font where you are seeing this issue.

If it is deliberate, I can think of why. Elsevier may be wanting to track down those leaking their credentials to sci-hub to shut them down http://www.theatlantic.com/technology/archive/2016/02/the-research-pirates-of-the-dark-web/461829/