I’ve had my Kindle for almost four months and have read hundreds, maybe thousands, of pages on it, from blogs and newspapers to magazines and books both short and long, both classic and fresh. And now I must rant to the world about the surprisingly low quality of digital print. Not the font—the Kindle’s display is so much like paper, visually, and the font so smooth and clean, that I relate to it like paper print, and judge it like paper print. Because I do, I’m much more aware of the fairly severe and systematic typos that seem to slip past whatever digital proofreading process is in use, if any. Here are the worst problems, and some possible solutions:

Punctuation
This is the big one, the greatest difference between paper print and digital print. I can’t do any better than this excellent post on ten typographic mistakes. So many publications make quote and dash mistakes that it’s almost more efficient to keep track of those that do it right. (The Kindle edition of The New York Times is one.) A fiction series I finished recently by Robin Hobb (Fool’s Errand etc.) was almost perfect, but insisted on doing ellipses as period, space, period, etc., which occasionally produced the distracting spectacle of an ellipsis broken by word-wrap.

I want to collect the publishers of these offerings together in a locked room with a pile of paper books from my study and roar at them that on paper, nobody makes any of these mistakes! I would hurl yellowed $4 disposable trade paperbacks and pulp magazines at their heads, screaming that not one of these throwaway tomes confuses tick marks with quotes and asking them what they have to say for themselves. What’s more astonishing is that the very same publication will be perfect on paper and flawed on e-paper. In particular, my Asimov’s sci-fi magazine, though cheap and tawdry on paper, makes zero punctuation mistakes, whereas its Kindle edition uses tick marks for apostrophes (it does get double-quotes right). What is the deal here? Does something about the digital medium make everyone forget what print is supposed to look like?

Word Wrap
Unfortunately, even those publishers that produce perfect source material for the Kindle stand a good chance of having it mangled by the Kindle’s word-wrap algorithm. The Kindle takes an unnecessarily timid approach to the problem, deciding that lines can only be broken by spaces. Not hyphens. Not dashes, em or en. Not ellipses. In other words, once again, none of the line-breaking conventions that have been used consistently and correctly over the last century or more by every single paper publication.

This leads to some weirdly clumped lines on the page. When an author indulges in something like the following (from The Geography of Bliss), the result looks ridiculous on the Kindle:

“There is yim cheun chom, the I-admire-you smile, and yim thak thaan, the I-disagree-with-you-but-go-ahead-and-propose-your-bad-idea smile.”

Because the Kindle won’t break at hyphens, it gloms everything into one “word” until it is forced to break “ahead” into “ahe” and “ad”. Although a huge hyphenated stream like the above is more rare, it’s common to run across a sentence using a couple of em-dashes and an ellipsis, and instead of wrapping correctly, the Kindle will make one big unbreakable word out of it, warping the justification.

Hyphenation
Since we’re talking about hyphenation, the Kindle could use a good hyphenation dictionary. Now I realize that the Kindle’s processor and memory might not be big enough to warrant such an investment, but couldn’t there be a way for authors to insert a “potential hyphen”, something that would tell the Kindle that it was allowed to hyphenate here in order to justify a line better? It would help.

Character encoding
The Kindle has glyphs for all of the typical Latin characters, and will render them when presented as HTML entitydefs (like “í” to mean í). The problem is that there are a lot of other ways to render these same characters, all of which are perfectly valid and equally popular, including UTF-8, Windows-1252, and ISO-8559-1. But if one of these alternate schemes shows up in a document presented to the Kindle, it simply chokes, rendering it as garbage or not rendering it at all. Since the most common ways of getting a document into the Kindle involve bouncing off a server of some kind (e.g. emailing one’s own document to it via Amazon) you’d think that there’d be some extra cycles on that server to spend on interpreting these alternate encodings. There are programs that can do this, like, oh, let’s see, every single web browser written in the last decade.

Conclusion
The good news about all of these flaws is that they’re correctable without having to mess with the device itself. A little attention on the server side, maybe a tiny update to the word-wrap algorithm, and we’re all good to go. Stay tuned.