Boox NeoReader Annotation Export Is Meh

When you use the built-in “NeoReader” on a Boox tablet, you get the best pencil input and quite good highlighting and annotation support.

If you don’t have a Onyx Boox eink tablet with that app installed, don’t bother looking for it on the Android/Google Play Store – that app is not available anywhere else, it seems. And the app of the same name on the Play Store is a QR Code Reader.

The built-in NeoReader app’s annotation export is very disappointing, though.

Here’s the one highlighted section from “Do Androids Dream of Electric Sheep” I have. It’s the TXT export of one highlighted section from the ePub version of that book:

Reading Notes | <<Do Androids Dream of Electric Sheep_ - Dick, Philip K__33d2f489-4536-43a1-ab54-5b581fb9efb1>>Dick, Philip K.
2023-01-19 21:50  |  Page No.: 148
How can I save you," the old man said, "if I can't save myself?" He smiled. "Don't you see? There is no salvation."
 "Then what's this for?" Rick demanded. "What are you for?"
 "To show you," Wilbur Mercer said, "that you aren't alone. I am here with you and always will be. Go and do your task, even though you know it's wrong."
 "Why?" Rick said. "Why should I do it? I'll quit my job and emigrate."
 The old man said, "You will be required to do wrong no matter where you go. It is the basic condition of life, to be required to violate your own identity. At some time, every creature which lives must do so. It is the ultimate shadow, the defeat of creation; this is the curse at work, the curse that feeds on all life. Everywhere in the uni

It’s nice that there is a timestamp (2023-01-19 21:50) per highlight.

But this page number is useless – this is an ePub, which means the amount of pages 100% depends on the font size. At least in this case, because there’s also the convention of denoting positions as blocks of 1024 characters in the absence of a page-map file.

For example, when I use the Calibre server to read the ePub edition of the book, the cover page is denoted as Current position: 1.0 / 291. When I jump to position ‘148’, I’m no where near the highlighted section. If I do go to the highlighted section, Calibre reports this as 213.3 / 291. (The decimal point is a bit confusing to me; why is that kind of precistion even needed?)

This means:

  • Annotation export is useful for quotes. I can live with not knowing the location in an ebook for quotes. Just copy the part into a note in my Zettelkasten, reference the source, done.
  • But the annotation export is useless for ‘academic workflows’. I can’t use the export to get to the highlighted location in another ebook reader app, so I need to use the device to check what’s on the page and around it for context.

So I’m stuck with processing information in an ‘academic’ workflow by using the device itself. Just like I would have a paper-based book open next to me.

I did a whole series on processing David Epstein’s book “Range” if you’re interested.

That doesn’t defeat the purpose of the reader. It’s perfectly fine to prop up this digital book the way I’d prop up a physical one. But it diminishes its potential.

If only the export used character offsets, I could imagine writing my own scripts to jump to highlights in .epub files. But since I have no clue how to make sense of a page location, the TXT export doesn’t help at all.

None of this is an issue with PDF files, of course: they have fixed page contents and thus page numbers do make sense. (But print PDFs don’t fit on the tiny Nova Air2’s screen.)