Originally Posted by slex
ScanTailor doesn't handle PDFs directly. It is a program to postprocess scanned pages. You should disassemble the pdf into images first, then convert them into jpeg/png/tiff and only then add them to ScanTailor. Under linux:
pdfimages filename.pdf f
mogrify -format png *.pbm
And the files are ready for use in ScanTailor. Then you assemble the output files of ScanTailor with another program - e.g. gscan2pdf or djvusolo. I would prefer to process the files from archive.org in this way, even if they were ready out of the box for Onyx M92. The procedure allows you to trim unnecessary white margins, center text and straighten up tilted pages. The new file is usually more neat and reader-friendly than the original.
You can, of course, skip all these steps and download djvu files directly from archive.org, if you don't mind the formatting. To download don't click on "DjVu" in the "View the book" box on the left. Instead click on "All Files: HTTP". This will open a page with a list of file formats. Then right-click on the djvu file and select "Save as".
Slex, with your final suggestion you have made my day. This worked and I am absolutely delighted. I can now read materials from the Internet Archive on my M92. I'll raise a glass on you tonight. Many thanks.