Fonts and Nonsense: What Bookerly and Literata Get Wrong

Expert publishing blog opinions are solely those of the blogger and not necessarily endorsed by DBW.

ebook typograhy Kindle Bookerly LiterataThere’s been a lot of hoopla lately about two recent typographic improvements to major e-reading environments, Kindle’s new Bookerly typeface and the Literata font for Google Play Books.

Commentators have tossed around celebratory remarks like, “the Kindle finally gets typography that doesn’t suck” and “e-readers rejoice!” But many of us in the cheap seats—that is to say, the production business—have been snorting and expressing our deep skepticism about these typographical “innovations.”

Still, I wanted to check my own curmudgeonly reaction with someone whose eye for type is more sophisticated than mine. Charles Nix is a senior designer at Monotype, and in 2014 he gave a wonderful talk at ebookcraft in which he referred to a new phenomenon called “Ebook Stockholm Syndrome.”

In a marketplace where design and typographic finesse have been largely absent from a depressingly high percentage of what goes on sale, and where well-designed ebooks swim against a mighty stream of poorly QA-ed titles, many readers have come to expect their ebooks to resemble what I’ve heard referred to as “Web sites from 1998” (David Blatner, at PePcon 2013).

To get all the ebook and digital publishing news you need every day in your inbox at 8:00 AM, sign up for the DBW Daily today!

So maybe the bar has been set low, but do Google’s Literata and Amazon’s Bookerly really up the typography game in a revolutionary way?

“The ebook environment challenges the fundamentals of legibility,” Nix says, which it then becomes the task of capable designers to correct.

As mobile e-reading continues to rise, challenges are only growing more acute. Particularly on e-ink e-readers and many smartphones (including phablets like the popular iPhone 6), the available line length is relatively short. “No amount of finessing will make justified type work well on a short measure [line length]. There just aren’t enough characters on the line to play the game of distributing space and hyphenating words,” Nix explains. Yet on the Kindle, where full justification is the default, smooth, agreeable typography is, well, just not possible—new font or otherwise.

Nix suggests that offering a new typeface choice “is a band-aid” solution to these bigger constraints. And default justification combined with poor hyphenation dictionaries (I’ve seen “o’cl-ock” hyphenated just like that recently!) are just two of the most obvious the culprits.

Alternatively, Nix says, setting the type “flush left/ragged right by default and more or less eliminating hyphenation on words less than seven characters long (and then with a minimum of three characters before or after a hyphen) would more directly address the problem of poor text setting” on many e-reading devices. (Have a look at this Fast Company article touting the innovation of Bookerly, and you’ll notice one of the screenshots—presumably supplied by Amazon—shows four hyphens in a row.)

And as for the goal of placing a well-aligned drop cap at the start of each chapter, Nix adds, “Proper alignment of drop caps is a fifteenth (possibly sixteenth) century problem. I have just as much use for a fool-proof recipe for royal icing!”

ebook typograhy Kindle Bookerly Literata

This is what my mother’s e-reader looks like.

I recently had occasion to open my mother’s e-reader to see what she was reading, and I nearly threw it across the room like a hot potato because I was so surprised by the state of it. She had blown up the size of the text—certainly part of her reader’s bill of rights—but this caused terrible gappy holes to appear.

As Nix rightly points out and my mother’s e-reader attests, “justifying on a short measure is the worst of both worlds: terrible word spacing and bad breaks coupled with staccato bursts of words (thought fragments)”—like an accidental e e cummings poem.

But apart from these typographical infelicities borne of shoddy decision-making under hardware constraints, there’s the more broadly problematic idea that one font will work well for every single kind of ebook.

This notion really just throws book design out the window by dispensing with any halfway nuanced appreciation of the content—something that should make any author or publisher recoil as well. Just as all print books are not typeset in the same font, so their digital counterparts should be afforded the same basic considerations.

It is wonderful that Amazon and Google are putting effort and energy into digital typography, and their newest Bookerly and Literata additions are most welcome. Really. But we do need a splash of cold water on the plaudits they’re receiving for doing so. There is still a lot of work to do.

11 thoughts on “Fonts and Nonsense: What Bookerly and Literata Get Wrong

  1. Gwyn Headley

    Don’t be too disheartened Laura — Bookerly and Literara are depressingly ordinary, derivative fonts designed for the lowest common denominator. But it won’t always be that way. I’ve just got a new Mac with a ‘retina’ screen and to be honest I can’t see the dots (well, I am 68 and half blind). Screen resolution is only going to increase, and in five years we’ll be able to use Walbaum and Modern No.20 as text fonts — and see the hairlines. In our ebooks we suggest (in the nicest possible way) that as they were ‘designed’ with the limited font menu available to us, that they should best be read in Palatino, or Seravek, or whatever the designer chose. Neville Brody once famously set a headline in Zapf Dingbats. But your mother needs font size and bright light to be able to read comfortably, and her Kobo allows her to do that; at the expense of rivers. — Gwyn Headley, author of The Encyclopaedia of Fonts.

    1. Richard Fleming

      “Don’t be too disheartened Laura — Bookerly and Literara are depressingly ordinary, derivative fonts designed for the lowest common denominator.”

      Exactly what I was going to write. I just saw an article that claims research shows that Baskerville is a “truthful” font because it came across better than the worst mass-market fonts ever made like Comic Sans and Georgia. Bookerly is extremely pedestrian. I just saw an example page (albeit pixelated) of Literara and it looks like the typographical equivalent of a Precious Moments figurine.

      The day I publish my work in anything resembling Baskerville is the day I get a lobotomy. I looked at the list of Kindle fonts and immediately realized I just can’t subject my poems to it, regardless of how poorly the fonts are displayed. I spend huge amounts of time customizing fonts in InDesign. Going to standard fonts is horrible enough but the standard font needs to be somewhat decent, like Bell MT or Adobe Garamond Pro.

  2. Caleb Woodbridge

    It often surprises me how large people set the text size on ereaders – I always turn the font size to the lower settings to get a decent number of words per line, but many people use large settings, and ebooks need to display well on these settings too.

  3. Ben Denckla

    Thanks for this article; we need more such counterbalances to the overly-adulatory (even sycophantic?) tone that plagues so much tech journalism.

    There is a damaging cultural divide between people who grasp the techniques and aesthetics of paper typography and those who grasp the techniques of software development. Thus software that renders words to screens is rarely good and often terrible (e.g. Kindle). It is perhaps too much to expect a single person to grasp both paper typography and software development. Don Knuth comes to mind as an extraordinary counter-example that only illustrates the general point. But humans are supposed to be able to collaborate to accomplish more than any one human could alone.

    Yet, software that renders words to paper well has existed for a long time and continues to improve. Everyone seems to have their pet peeves with Adobe InDesign, for example, but these pale in comparison with the crippling limitations of, for example, Adobe Digital Editions. (A recent major release of ADE has improved things, but, as with the overly-touted Kindle improvements of Bookerly and hyphenation, it is too little, too late.)

    Are the challenges of rendering to screen really so much greater than rendering to paper? Undoubtedly the challenges of rendering to screen are great. But I wonder whether there is another, more subtle cultural divide within software people, where the experience of people who have written paper renderers is not shared with those who are writing screen renderers. For instance, at Adobe, a big company, it is easy to imagine that there might be little or no contact between the InDesign people and the ADE people.

    At Amazon, unlike Adobe, there is no culture of respect for words anywhere. So Kindle’s failings are not failures of inter-organizational communication. Kindle e-Ink hardware is impressive, but it appears its software engineers are operating in a vacuum ignorant of the techniques and aesthetics of typography of the last 400 years and the ignorant of the techniques of word-rendering software of the last 40 years. The value of its beautiful new 300 dpi e-Ink display is seriously mitigated by a fanatic insistence on full justification. It will be an improvement if, as promised, later this summer the apparently-crappy hyphenation algorithm appears on e-Ink devices (currently it is only on iOS?). If full justification is insisted upon, even crappy hyphenation is better than none.

    But why are we forced to make such \lesser of evils statements\ like that? It is a testament to the success of the software business in general to have trained us to expect such mediocrity. Amazon has the early Microsoft, among others, to thank for that. A certain older generation of users has been trained to accept any functionality short of crashing as pretty good, since crashing used to be standard. In this light, the unsightly gaps of full justification seem like a pet peeve: just be thankful it doesn’t crash (usually; I have crashed my Kindle)!

    Finally, all typographic concerns pale in comparison to my particular area of concern which, in our age of diminished expectations, is seen by many as a pet peeve: typos (usually OCR errors) in Kindle and other ebooks. Granted, these typos are not Amazon’s fault, at least not directly. Perhaps one could argue that Amazon strong-armed publishers into creating rushed, cheap ebook editions so as not to have the \buy\ button removed from their print editions. But there’s plenty of blame to go around. How could the publishing industry not see the \writing on the wall,\ that books would become digital some day? All due respect to Bezos and others, but it did not take a business genius to see ebooks coming. Paper publishing was a staid, self-satisfied industry that, to some extent, got what was coming to it when Bezos and others stirred things up. Their cries of foul play against Amazon ring somewhat true, but also sound a little like the sour grapes of an industry caught by surprise.

    On the other hand, Amazon could have done a lot more to help publishers make the digital transition gracefully. They could have provided software, and technical hand-holding, that would have made high quality ebook editions possible. They could have created the collaboration between the publishers, who grasp paper publishing, and themselves, who grasp software.

    Let’s return to Microsoft, but this time, in a positive light. They identified that most of the crashing problems in early versions of Windows were due to driver problems. These are nominally not Microsoft’s fault, but Microsoft realized that, to the consumer, a blue screen is a blue screen, i.e. the consumer is going to see them as Microsoft’s fault. So it launched a rather expensive campaign of technical hand-holding to improve the quality of its driver ecosystem. And it worked.

    The problem is, neither poor typography or, more seriously, typos, seem to bother Kindle readers much. I can only imagine this is because they’ve been trained. The attitude towards ebooks reminds me of a famous line from the movie Chinatown:

    Forget it, Jake; it’s Chinatown

    This can easily be re-worked as

    Forget it, Jake; it’s an ebook


    Forget it, Jake; it’s software [which inherently sucks, right?]

  4. Brett Merkey

    I have read your posts for a while and really enjoy the insights. Thank you. Your insights got even more insightier when recently, fed up with criminally bland e-book layouts, I started hacking my very own personal editions. After all, if Amazon is charging almost hardback prices for aethereal Kindle azw files, why shouldn’t we expect e-books pleasing to the eye?

    My new Kindle Paperwhite v.3 arrived yesterday. Could not wait to load books onto it and see if Bookerly would transform my user experience. Well, no revelatory experience. I will keep it in trial and see. The default justification seems to be smoother. I have to admit that in my career as a Web application designer, I am not prone to get orgasmic over fonts. We use fonts to communicate and in this multiple device world, our options are limited. Bookerly is a plus. One more option.

    We should grasp all options with a sense of optimism. The e-book industry has a bright future—but there are obstacles. In my brief career as a hacker of my own e-books, I notice two things:

    1) e-book construction tools and processes are primitive or clumsy. They remind me of the state of online application documentation tools in the early 1990s. The e-book construction process at the moment requires too much knowledge of underlying code tech to be accessible for an experienced \book person\ who is trained to know how a book should feel.

    2) The larger publishing houses / distributors invest almost no effort in using the primitive tools we do have to produce something which gives a better user experience. It is not that hard, not that expensive to produce something qualitatively better.

    I recently performed an experiment in this regard and published the result on Amazon, cleverly disguised as a mordant review:

    Brett Merkey

    ps: let your mother read as she pleases on her device. No hot potato, no jihad…

  5. Brett Merkey

    Hmmm. Apparently this blog’s software is not the type to treat posts kindly…or rapidly.

  6. Mary

    The Kobo has a gazillion settings you can play with. For one thing, there are many more font choices and way more sizes than on Kindle. Also, you can make the text darker if you wish which allows some of us to read with a smaller font than we would otherwise. And as to your mother, you can turn off justification on the Kobo. I read on mine with a larger font, darkened and ragged right.

  7. Ben Norland

    Interestingly, Apple now allow the user to turn off full justification and auto-hyphenation. I don’t know when the feature was introduced, but I am thankful for it. Annoyingly, the options sit in the iOS settings app and not in the the iBooks app itself, but at least they are there.

  8. Amelia

    Your blog post isn’t easy to read with the auto-hyphenation. How can you be serious about typography – and legibility – with that going on?

  9. elgarak

    It does not appear that you have tried it out. The changes amazon introduced with Bookerly did indeed improve reading greatly. They did not simply changed the font; they changed the hyphenation tables and spacing rules for justification, precisely to address some of your critiques, and avoid big splotches of white. For instance, when it cannot justify and cannot hyphen, it tends to cut the line short, and keep the spacing within this line consistent with the rest of the paragraph. This may not be very nice — but it works to keep readability and avoiding ‘stumbling’ blocks. Justification is still on, and I can’t say that I disagree with that. It keeps a clean look, and if the spacing is the way it is now, it just works.

    And yes, the font itself was necessary. Case in point, I got an e-book recently where I was greeted with Use the publisher font to see how it is intended at the beginning. I did, and was horrified. The publisher font was junk, which is the politest way for me to say it. The vertical lines were too thin, so that the text blurred on small size settings. To be legible, I needed to set the size to setting 5 or 6. The low side of the sizes was unusable, and to get the letters really nicely, I needed even bigger size, close to what you saw on your mother’s reader. In fact, in my observation, a lot of print fonts (and in particular the serif fonts that are the most elegant looking) are plaques by problems like this on (the relative low resolution) e-ink screens), which is, I suspect, one reason people like your mother tend to move to large size settings. Bookerly improved on that. It provides readability in all sizes/weights, and across all screens (after all, amazon caters to e-ink readers, phone screens, tablets and desktop screens, via its kindle reading app. You may not believe there’s a font that works on all, but Bookerly comes very close). It also avoids having visible pixels and ‘stepping’ on low resolution that I see a lot on print fonts.

    Lastly, I think you miss the point that this is new technology. It’s not print and paper anymore. The rules have changed, and typography needs to evolve.

  10. Desolo

    There’s a reason I spend extra time grabbing book content (sometimes downloaded from an author’s website, sometimes downloaded from Project Gutenburg, sometimes copied from a PDF) and setting it up to be read digitally on my own. I open the downloaded file or copied PDF text in Notepad++, use the find and replace feature to re-write or write the HTML, and include a link element in the head that points to my own library.css. That sets up the general layout, including text alignment to the left, centered sticky navigation by chapter at the top, line-height, etc. Then I add internal CSS for the font family, font size, colors, and borders.

    In the end, it’s so much nicer to read a book I’ve style in my local digital library, even though it’s a good bit more work than just downloading to a Kindle. Throw in cloud storage as a virtual bookshelf for the individual book files I’m not currently reading, and it’s a perfect (end-state, anyway) solution.

    Clearly the current page design in Kindle isn’t ‘doing it’ for me if I go through all that before starting to actually read the book.



Your email address will not be published. Required fields are marked *