Web dev at the end of the world, from Hveragerði, Iceland

Books and Print Showcase

In their own words:

Books&Print Sandbox is supporting eight ground-breaking collaborations between creative economy partners and academic researchers to explore Books and Print as historical, contemporary or future phenomena.

In other words, a sandbox for trying crazy things, where failure is relatively safe for everybody involved, something that’s a rarity for many of the ‘creative economy’ partners (I always find the term creative economy weird for some reason).

All of the projects were interesting. Some more so than others.

Nico Macdonald opened up with a statement that is both true and not true (paraphrased slightly as it’s from memory):

In thirty years we’ve gone from ebooks to… ebooks. Which isn’t a lot of progress.

It’s true insofar as progress in ebooks has been slow and in many ways has been (and is being) held back by misguided elements in the ebook industry.

It’s not true insofar as it implies that iterating on improving ebooks isn’t a worthwhile endeavour or a potential source for innovation. There is a lot of value in nailing the process of creating and presenting long-form readable text in an attractive design.

But that’s a nitpick on my part because I’m sure his point was that after thirty years of digital reading the only thing we’ve got to show for it are ugly websites and half-arsed ebooks; surely we can do more?

It’s an excellent point that I’ve just intentionally hijacked to elaborate on a pet peeve of mine. Sorry, Nico.

First project up was Book Kernel. As I understood it, this is a project to make it easy for event organisers to create books out of their events, offering crowdsourcing, easy editing, and simple creation of ebook and print book files.

Each of the audience members should be able to create their own memento of the event; a custom ebook or even print book.

More in their own words:

We make books from a live experience and get them to you before that experience has ended.

We can reprint them for you too.

Its focus (event organisation) is well out of my range of expertise or familiarity so I don’t have much to say about their project. They’ve gone for a niche and, to an outsider, seem to have done it well.

The Next Time(line) is a project involving Alex Butterworth and Bradley Stephens that attempts to present all of the various (very) extensive data on Wordsworth’s The Prelude in an interactive and graphical form.

For anybody unfamiliar with the history of The Prelude that is a lot of data, a lot of versions and iterations, over many years, with copious annotations, by an author who read a lot and annotated a lot of what he read, connecting them with the whatever he was working on.

It’s pretty much going to be impossible to present in a manageable form.

Their interface work in places an didn’t work in others. They have very clear ideas for how to try to address the bits that didn’t work. More importantly, the interface they came up with looked like it would scale down very nicely.

So all in all, they came up with a data interaction GUI prototype that almost but didn’t quite work for the dataset they chose, but will work on almost any sane dataset you find.

But, of course, I’m biased because my first degree was in comparative literature and this sort of tool is what a lot of my professors back then fantasised about.

The Secret Lives of Books was the first real disappointment of the evening. It’s a technical marvel but I have no idea what it’s for.

This is what they claim it does.

Visualising the unexploited data sets of our public libraries to reward bookish curiosity with unexpected connections

Yeah, not really.

What they demonstrated in the showcase was a motion-tracking system for books (that was the cool bit that could be put to a near infinite variety of uses) which you used to navigate a crap 3D library UI of books by picking a title up and waving it around. Most of the data visible was a faithful copy of the stuff you can read off the book’s covers.

The problem is we already have a 3D library UI for books. The library itself. What we need is a detailed and flat 2D overview of the various metadata the book has. If their project does offer useful data, then they did a horrible job of presenting it in the showcase.

You could accomplish a much more useful system by embedding NFC chips in all of the books (more on NFC later) and putting up a big honking screen in the corner of your bookstore or library. When anybody is curious about a book, they pick it up, bring it up to the screen or a pad next to the screen and get a screenful of metadata.

You could even go low tech and do it with barcodes. (Which, as it happens, is what Amazon is doing with some of its smartphone apps.)

I do have a bias against 3D UIs so that in and of itself may have put me off the entire project.

—But, Baldur, how about all of the data that motion tracking generates? You could know exactly where all of the books go? That’s cool!

No, it isn’t. Data only has value if you manage to preserve its context.

For example, judging by my analytics data, most of you come from Twitter. That’s useless. I knew you’d come from Twitter because I spend too much time there.

What the analytics data is missing is what the link’s context was on Twitter. Was it a retweet? Was it thrown into a conversation? Did the tweeter disagree with me? Agree?

I don’t know. Data without context, such as what we get from website analytics only serve as point-scoring incentive systems driven by mimetic desire. Can I get as much traffic from Twitter as blogger X? Unless you observe people taking specific actions in a specific context on your site, you have no real data you can learn from.

Ebook analytics are similar. What does a high ebook abandonment rate mean? If we know that the ebook in question is fiction, that people bought the series in one go, and that after giving up on the first one they never touch it or the sequel again, then yes, you might be able to draw some conclusions from that.


Generalised ebook analytics lack most of the context you need to draw conclusions with much confidence from data points such as abandonment rates.

Do the readers all stop after the first section?

Maybe the remaining sections of the book have no value, sure.

But it also could be that a university has assigned just that section in one of it’s mandatory courses.

Or, it could be that the section addresses what has become a hot button topic, drawing a one-time traffic blip just for that section.

Or, it could be that the author has updated the remaining sections for free on their site, but telling people to buy the ebook for the first section that is still up to date.

Or, maybe the remaining sections are so conceptually difficult that most of its readers will read it in batches with long gaps between sessions, months even, and the title hasn’t been in your system for long enough for that data point to appear.

More time is often the answer to a lot of these questions. Analytics for one week is noise, but for several months might show an actionable pattern.

Data doesn’t tell us as much as we think it does. It’s fun to play with (hence my approval of Alex’s project which presents data in a playful way) but you generally cannot be confident of the conclusions you draw unless you are confident that you aren’t missing any important context. And context that’s missing is usually an unknown unknown, to quote Rumsfeld and therefore something that we don’t know we don’t know.

Little J is a hyper-local news project. Again, slightly out of my field of expertise.


Most projects of this kind are community-building exercises whose viability hinges on how active and cohesive the local community in question is. And that’s an issue you can’t really solve with software. Good software doesn’t make an apathetic community less apathetic.

Good software, however, can remove hindrances and roadblocks from projects based on active communities.

You can have a look at it at littlej.org and make up your minds on it yourselves.

(I’ve known Tom Abba for years and years so I’m definitely biased on this next one.)

These Pages Fall Like Ash is a story of two cities and two books. The two cities are Bristol and an alternate version of Bristol. The two books are a print book and texts and other media distributed over several hotspots in Bristol city centre. We wander around our Bristol and catch glimpses of the alternate Bristol through the digital and print experience.

This project got a lot of press because it hit a geek trifecta:

  • Geek fav writers Neil Gaiman and Nick Harkaway helped with the plotting.
  • The hotspots were implemented using Raspberry Pis.
  • Locative media.

With bonus points for alternate universes and the like.

I really like what Tom and Duncan have done here. The print book is gorgeous and weird. The book is one of those things where, if you know Tom and Duncan, but didn’t know of this project and somebody handed it to you, your first thought would be:

This looks like something Tom and Duncan would make.

The concept is fun. And the weather has been very nice. I’m looking forward to finishing up the last bits of the project this weekend with a series of walks around the centre.

Although I’ll be using my iPad, since I don’t carry a smartphone.

Which brings me to…

Writer on the Train:

Taking train travel as a starting point, award winning agency Agant, author James Attlee and Fabrizio Nevola of Bath University will play in the fixed, linear space of the Bristol to London mainline, using smart phones and GPS to deliver a new literary form. Their prototype app will respond to the readers’ journey in real time, delivering elastic pacing, video, audio and new writing relevant to the train’s location.

It’s a story for commuters. When you first open the app you tell it which bit of the line you commute on. Then after that you will get one piece on your way out and another on your way back home, each keyed to a specific location and context on the train route from Bristol to London. The app has a fixed number of entries.

Even though I do commute on that route regularly I can’t try it out because I don’t have a smartphone. The project sounds like a lot of fun, though. Pieces written while on the train delivered to you when you’re on the train.

I did make the very mercenary suggestion that they could offer an in-app purchase that unlocks the entries for those who are either impatient or don’t commute. Which made Dave Addey laugh.

The app is well designed with a layout and colour scheme that echoes the London Underground map style and the text is nicely set.

On the whole, projects like this one make me contemplate upgrading to a proper smartphone.

The final project, Digitising the Dollar Princess, is the one I’m the most critical about. I even was so evil as to ask nasty questions in the Q&A session of the showcase which prompted a long conversation with one of the project partners later on.

In their own words:

Lady Curzon’s journey as the Vicereine of India was dictated by the rhythm of the Raj, recorded in intimate detail through letters, diaries, in clothing and photographs. These material remains illuminate her life – but which stories should we tell? Through the lens of this fascinating woman, Nicola Thomas of University of Exeter and Bow Software are setting out to break new digital ground in the genre of biography, creating a compelling non-linear reading experience led by curiosity and rich interaction with source materials.

I have a thorough dislike of fake book chrome and page curls so that may well be what set me against the project upon first impression.

I get what the project is about and that the project members had a steep learning curve for everybody involved but it is both too far ahead of the curve and too far behind.

It’s behind the curve in that it created a very skeuomorphic book app that is in most ways that count (UI and features) a fairly standard member of its species. This is the bit that every major publisher has done at one point or another, which means that it isn’t much of an innovation.

It’s an interesting and useful business development project for the software developers, it’d be churlish to knock it for that, but in my (admittedly arrogant) opinion a little bit out of place in a sandbox like this.

The bit where they’re ahead of the curve doesn’t really work either, but its interesting, which in this context makes it a success. Exactly something that nobody would try under normal circumstances. They tried something. It didn’t quite work, but they know where to go from here.

That’s exactly what a sandbox like this is for.

So, what was it that they tried?

NFC bookmarks. You use the bookmarks to unlock specific features and sections in the app.

My niggle-y comments:

  • The UI didn’t quite make it clear where or how the bookmarks worked. Most of their target market (elderly people, given the subject matter) are going to need someone to explain it to them.
  • When given a choice between stocking NFC bookmarks for a single title and generic gift cards for an entire app or ebook store, most booksellers will choose the gift cards every time.
  • Most tablets don’t have NFC support. Neither the iPads nor Kindle Fires support it and there is no guarantee that the next versions will.

I’d still like to see them develop the idea further, specifically to try it in different contexts and with different subject matters. It’s a high risk research project with the odds stacked against it, but those are exactly the ones that have big upsides.

And y’know… Do it for science :-)

George Walkley, of Hachette UK, said this here thing:

It is no exaggeration to say that in five months the Sandbox has delivered as much as some mainstream publishers have done in five years.

The quote highlights a couple of things that have become clear over the past few years. The first is that publishers aren’t that good at experimenting in the digital space. The second, which is a corollary to the first, is that most innovation in digital publishing comes from outside the publishing industry.

I don’t think there is any way around that. Any corporation that is older than ten years old is either going to be fundamentally innovative by nature or conservative to the core. Changing the nature of a corporation involves changing the personality of most of your hires over the past decades, changing every process, modifying every rule and procedure. That’s impossible. Any major publisher that has been around for a while is either capable of deliberate innovation or they are not.

The question is whether they should worry about it or not. And the answer to that depends on whether you think that digital publishing—content apps, websites, and ebooks—is disrupting print publishing or not.

And that’s an issue that deserves a series of blog posts, not a footnote in this one.

You can also find me on Mastodon and Bluesky