Arthur Attwell

Publishing, technology, and related opinion

Applying publishing tech in southern Africa

Tuesday, 13 October 2009 16:28

Here is my presentation from this morning at TOC Frankfurt. You can also see and download the slides (in slightly better quality) on Slideshare here. I've put all the URLs of projects, companies and issues I raise first. Incidentally, I've changed the title of the talk to refer to southern Africa, since that's what I can be confident about – though I hope much of what I say can be generalised, with care, to other developing countries.

URLs

Erik Hersman/whiteafrican

eSplash

Espresso Book Machine

One Laptop Per Child Project (OLPC)

Paperight

Random House Struik

More Phones in Developing World than Developed: The Guardian

SA Parliament Paypal discussion (text search for ‘Paypal’)

Sakai

Sam Wilson

Scribd

Shuttleworth Foundation

Silicon Cape

South African Social Investment Exchange

Steve Vosloo

Tactical Technology Collective

Thutong

Wizzit (mobile payments)

Zap (mobile payments)

Introduction

Good morning. I’m Arthur Attwell, and I run a small R&D and publishing company in Cape Town, South Africa, called Electric Book Works. I live and work in this ridiculously beautiful place, so when we get any work done we’re especially proud of ourselves.

I started EBW in 2006 because I couldn’t quite believe that so many local publishers – which is not that many in South Africa – were so blasè about the importance of technology to their work. I’d worked with big and small companies for ten years, and still I couldn’t understand their reticence to engage with technology in any depth, not even with the digital tools they were already using, but in a very limited way.

For example, In 2004, a dear colleague of mine insisted on playing out a 2000 page, three-volume book to film for manual imposition, despite the fact that local book printers had been producing plates from PDFs for years. And in 2005, I had to explain to my bosses in a written report why, as schoolbook publishers, we really needed internet access.

Why were these otherwise sensible people so reluctant to engage with technology? Well, I know some South Africans think this is a developing-country problem. They say that living in a backwater country predisposes you to ignorance, and that I shouldn’t be surprised. But I don’t buy that.

It’s comfort, not struggle, that predisposes you to neglect technology. The more comfortable you are, I think, the less you need to look for better ways to get your job done. And living in a developing country is not comfortable: whether you are poor or you live with the poor on your doorstep and on your conscience, your discomfort should surely predispose you to a constant, almost desperate need to improvise, and to discover new technology in doing so. And the more you learn about technology, the better you can improvise.

So technology is not a luxury in the developing world, it’s a window on opportunity, and it accelerates social upliftment in clear, tangible ways. It means textbooks being remixed by teachers, schoolchildren getting homework help by SMS, libraries previewing books before purchasing them, books reaching schools more quickly, authors working collaboratively with editors and each other – and I’m just talking about schools publishing here. There are other opportunities in adult education, healthcare, community development – all of which could be positively affected directly by publishers digitising their content and experimenting with even small parts of it.

In industries, technological innovation and experimentation are booming. Here are a few examples I like:

DSTV, South Africa’s biggest pay-TV provider, in trialling streaming TV to mobile phones, for example.
MyMag has dozens of local magazines available for digital subscription
Matchboxology works with big corporates, like Levis, to combine brand promotion and positive social contributions, they use a lot of web video and music.
The Shuttleworth Foundation runs projects that ‘drive social and policy innovation in education and technology’, like their open-source SchoolTool school admin software suite. (Incidentally, Mark Shuttleworth from Cape Town is also behind Canonical, the company that maintains the Ubuntu Linux distribution)
SASIX is the South African Social Investment Exchange runs a stock-exchange-type system that provides ‘investment opportunities with a social return’.
Thutong, a government initiative, provides a portal to educational resources for educators -- though it is a little patchy.
Universities in South Africa are increasingly using the open-source Sakai for course content, communication and collaboration.
The Silicon Cape initiative was launched last week by some of South Africa’s brightest tech entrepreneurs, and it aims to drive tech innovation in our part of Africa.
And the Tactical Technology Collective, an international NPO with a southern African project lead, develops tech to promote freedom of expression, participation and accountability.

And if you want to hear about more initiatives, I don’t know a better resource than Erik Hersman on Twitter at whiteafrican.

Thing is, I have a really hard time finding book publishers in all this activity. So, again, why are otherwise sensible publishing people so reluctant to engage with technology?

The honest truth is that 90 per cent of publishing decision makers in South Africa, including me, don’t really live in a developing country: given our wealth relative to most South Africans, we live in a first-world bubble inside a developing country. This is true in many developing countries: mainstream publishing is mostly run by comfortable people, and so the benefits of technology take longer to become clear and urgent.

Of course, they would say there are resource, financial, reasons for not engaging with tech. Perhaps -- margins in local book publishing are so small it’s sometimes run more like a semi-professional pastime than an industry. And this doesn’t improve as long as publishers don’t find economies of scale on the production end, which only an engagement with technology can provide. In isolated cases, some publishers have found economies of scale in tech, specifically in dictionary publishing where XML-based systems are incredibly valuable. And these isolated cases only highlight the fact that it can be done without the financial sky falling in.

So it seems to me that neither financial reward nor moral obligation has tempted book publishers to engage with technology in southern Africa.

If I was less polite, I’d say that – given the benefits technology can bring to low-income economies – the neglect of technology by so many South African publishers – of even the simplest digitisation and repurposing experiments – is even scandalous. But I am very polite, so I won’t say that.

And there are, after all, some wonderful, imaginative publishing people using technology in simple, sensible ways. And I’ll mention a few of those I know about in a moment.

Improvising

Anyway, what technology am I talking about? I must point out that I don't expect publishers to develop or buy complex software or data-asset-management systems. Let’s assume there’s no good financial reason for South African publishers to invest in those.

I’m more curious in the ways that we can improvise to make and distribute book content in ways that cost almost nothing.

So, what do we have to work with as we improvise?

In this slide, I’ve created a rough and, I’m sure, incomplete mind map of the digital-publishing industry buffet. All these things play a part in making the industry what it is internationally:

reading mechanisms
distribution and retail providers and services
production factors
a choice of payment systems, this is really important
small and large innovators
existing markets for books
an active rights industry
and a range of industry bodies that can coordinate efforts.

Now, let’s fade out the things most developing countries don’t have, things that won't feature in developing countries for a few years at least.

Beyond our little first-world bubble, we can’t realistically expect e-ink ereaders to catch on beyond wealthy early adopters for a few years at least, by which time ereaders are likely to have evolved into something completely different anyway. The same goes for meaningful smartphone penetration. iPhones and Android phones currently cost almost half my monthly salary in cash.
PCs and netbooks will have moderate penetration. In South Africa government and business initiatives are putting laptops in the hands of teachers en masse. The One Laptop Per Child project may also traction in several countries.
In distribution and retail, many services provided internationally are available in sub-Saharan Africa, certainly major ebook retail aggregators like Ingram and Overdrive, and the Internet Archive and Google Books. Where we lag currently are in the widespread growth and integration of repositories, publisher websites that carry and sell content, investment in consumer education, and of course bandwidth. I’m leaving bandwidth only a little faded, since southern and East Africa are seeing real improvements in infrastructure here right now.
In production, we have serious skills problems. It’s astonishing that my company is the only one I know in South Africa offering any in-depth ebook expertise. We have no true-POD printer–distributors, and very little adoption of DAM systems. Conversion services are of course available just across the way in India.
Payment systems are limited. Users in sub-Saharan Africa cannot withdraw money from Paypal or Google despite the fact that they can pay with and into them. This is a huge issue, and has even been raised in the South African Parliament as a major limiting factor to local entrepreneurship, but we’ve seen no progress on this so far.
Credit cards are just not widespread enough for direct card payments to be a truly scalable option.
As for innovators, venture capital for startups is very limited in sub-Saharan Africa. But innovation can and does come from other large and small organisations.
Existing markets for books are healthy in the first-world bubbles I’ve mentioned, but – except for school books – there are no substantial established markets for print books beyond that.
Our rights industries have a long way to go, though we’re seeing some nice developments in open licensing.
And the presence and effectiveness of industry bodies is very patchy, at least in their engagement with technology.

It’s important to point out that these are not gripes and limitations. Every one of the missing pieces can be an opportunity, one way or another – a gap in which to innovate and improvise and establish new business.

Out of what’s currently available in the buffet, I think shared computer labs in universities, schools, libraries and internet cafés are going to be a critical part of digital distribution in Africa, and it's really, really important to build business models that make the most of this, both for the publisher and the users.

And later I’ll come back to print-on-demand, which I think could be even more important.

One exciting area of development is in mobile payment systems, especially M-PESA in Kenya, which has over 5 million users. There's also the smaller Wizzit in South Africa and Zap in Kenya, Tanzania and Uganda. Using M-PESA, anyone with a mobile phone in Kenya can send and receive money without a bank account, sending money to each other and depositing and withdrawing money at thousands of airtime resellers. It’s used widely for buying and selling, from rural markets to city-centre taxi rides. I’m really looking forward to seeing how publishers use these systems to let people pay for digital and print-on-demand content.

So, let me describe some examples of our and others’ improvisations with what’s available to us.

One of my favourite improvisations is a two-month-old retail setup called Little White Bakkie, named after the indestructible little Nissan pick-up that’s such a feature of South African industry.

Little White Bakkie is a project of BOOKSA, South Africa’s leading literary website, run by Ben Williams. In only six weeks, Ben’s managed to get nine South African publishers, including three of the country’s biggest trade publishers, to give him 40 frontlist titles to sell on Scribd, racking up over 11 000 views – he's had over 10 sales, which by South African ebook standards is about 10 more than anyone realistically expected.

In getting these publishers on board at all Ben’s managed something I didn’t think possible, and it’s really because these publishers trust him personally, and because he’s made the process absolutely zero-cost and non-exclusive for publishers. Trust and zero-cost.

Of course, instead of trying to attract customers by providing a big catalogue, he’s just placing the books on relevant pages on his literary website. So Ben is primarily a social network developer, integrating book sales into his social platform, rather than trying to build a social network onto a sales platform. It's like Goodreads for South Africa with ebook sales built in.

Incidentally, Ben’s American, and this lets him sell in the US-only Scribd Store. To us South Africans, he offers himself as a human improvisation.

He’s helped my company use Paypal so we can publish to the Kindle, and to sell ebooks directly from our Mousehand website.

Mousehand is an Electric Book Works improvisation. We developed it to provide design and print-on-demand and ebook distribution to self-publishers in South Africa.

For international POD and ebook distribution, we work through Lightning Source to get our clients’ books on most book and ebook retailers, which is the really easy part.

But local distribution is provided by a hacked-together string of supplier agreements with digital printers and distributors that form a pseudo-POD system. And our ebooks are all made by hand, though we've developed a strict set of rules for design and typesetting that gets us halfway to epub in CS3 during the production process.

Our flagship improvisation, from a tech point of view, is EBW Healthcare. EBW Healthcare is a series of coursebooks that professional nurses can use to run their own courses. We publish the informational content of our books online for free, as web content that users can comment on almost at paragraph level, and as downloadable, DRM-free PDFs. We also distribute this free material on Scribd.

We’ve had tens of thousands of downloads and views.

What we sell is the printed books and ebooks that include the multiple-choice questions a user needs to run a course. So, course content for free, learning component, the questions, are paid.

The key to the system is openness: nurses are responsible for their own learning; anyone can register and submit comments on our content at sub-chapter level; and the content is almost all free.

The fact that our partner NPOs part-fund the initial production costs of each book reduces risk for us. But even without that funding we’ve found that our print sales would cover those costs if they had to. Don’t tell our funders that.

For three years to last July, all the book production and web development was done my me in about 20 per cent of my time, and it’s important to note that I’m not a developer. Apart from CS3, we've not spent a cent on software or development -- I was just using free tools that required only a fundamental grasp of HTML. This is not expensive stuff if you can find and afford just one person interested enough in it to try something new, somewhat obsessively, almost every day.

The EBW Healthcare website also has a simple template switcher that allows users to browse our site and read our books on a mobile phone.

The obsession with mobile

There is a justified obsession with mobile in developing countries. There are far more mobile phones in the developing world than the developed, and therefore a delivery device in nearly every pocket.

In South Africa, a small company called CellBook has been putting books on phones for over a year now, working mostly with Random House Struik and NB Publishers, which are two major publishing houses. They use a Java app that can carry a variety of interactive book content, so they can distribute to a wide range of very basic phones.

They store the content in XML. Incidentally, they had endless headaches getting XML out of publishers. But the real obstacle as it always is, was marketing the book apps to consumers who aren’t familiar with the technology. In fact, after establishing an agreement with one particular publisher, and having already produced the mobile ebooks, both parties thought the other would be doing the marketing.

When the marketing did work well, it worked very well: they tell me an Afrikaans–English dictionary they produced last year sold tens of thousands of copies in its first few weeks, selling through a premium SMS payment system that cost users about $6. And they’ve just produced an interactive mobile version of Random House Struik’s bestselling K53 drivers-licence guidebook, which costs about $5. A RHS insider reckons it’ll pay for itself in about six months.

MXit is one of the biggest players in mobile in South Africa. It’s essentially a mobile instant messaging app, and it’s grown into a massive social network with over 12 million users in South Africa and abroad, mostly teens, and slightly more male users than female.

In South Africa, users can also buy a MXit currency called moola, which they can use to pay for add-ons inside MXit, which are usually so low they’re effectively micropayments.

Earlier this year, a self-published author called Karen Michelle Brooks arranged with MXit to sell her fantasy adventure novel on MXit by the chapter – users paid for each chapter with moola as they opened it. There were 27 chapters, and each one cost about six US cents. MXit highlighted the book on their mobile landing pages, and within a month had sold over 5000 chapters.

Of course, while that’s great in South African book publishing terms, after the revenue split with network operators and middlemen, relative to MXit’s millions of users this wasn't really worthwhile financially to MXit. Even the author tends to say it was a great marketing venture, driving bookshops to stock the print edition of her book.

However, it’s fairly obvious that the model is not meant to work with single books, but with the scale of hundreds or thousands, generated automatically from a DAM system. Hopefully Naspers, South Africa’s biggest media company which now owns 30 per cent of MXit, will move actively to put their lists on MXit.

MXit is already working with the CSIR to put an interactive maths tutor on the system.

I should also mention the m4Lit project being run by Steve Vosloo of the Shuttleworth Foundation. He’s distributing a specially commissioned story called Kontax, aiming particularly at teens in poor areas of Cape Town. They’re releasing the story over 21 days this month, and by day eight they’d had over 15 000 unique visitors to the mobisite where the story’s chapters are posted each day, which is a much bigger success than they expected. There’s also an active social network around the novel where readers can friend and follow the main characters, whose profiles change from day to day based on what happened in that day’s chapter. There’ll be a research report about the project available in November.

One last mobile phone product I really like is airPAC, developed by the University of South Africa, or UNISA. UNISA is one of the biggest correspondence universities in the world, with over 250 000 students, and airPAC is their library catalogue, which you can search on a mobile phone. And if you’re a student, you can read their ebook content from NetLibrary on your phone too.

There are some really exciting things happening at UNISA, which really pleases me. I used to walk their halls as a higher-education publisher, feeling really worn down by the conservative thinking often prevalent there. Now I also think that UNISA, who are genuinely working in the developing-country part of South Africa, may just recognise and move to embrace technology quickly for such a big organisation, and may drag a few local publishers along with them.

UNISA is also using Freedom Toasters. Freedom Toasters are like digital content vending machines, and UNISA, along with other institutions are using them to distribute course content. The content on the Freedom Toaster is all free, so it’s not a commercial distribution model, but that doesn’t mean it can’t be really useful. We’re in the process of adding our healthcare books to the Toaster.

Decentralising print-on-demand: Paperight

I want to mention one last project, which is the one I'm working almost exclusively on now.

As I mentioned earlier, I think print-on-demand is going to remain a critical part of content delivery in developing countries for several years. My newest project, called Paperight, aims to explore and make the most of that.

In the slide I’ve outlined the part of the digital publishing industry we’re building Paperight on: aggregators, digital printers (I mean the physical machines as much as the businesses), and rights.

In a nutshell: Paperight aims to turn any copy shop into a print-on-demand bookshop. To its users, it’s a website for finding book content and, where content isn’t free, for buying the right to download and print it out.

Behind the scenes it’s a rights, metadata, and ecommerce platform that any content creator will be able to use for distribution. Importantly, it will get book content to places where no bookshops exist.

There are copy shops, often combined with internet cafés, throughout the developing world. Far more than there are bookshops. It’s a simple fact that soon enough, if not already, those copy shops are going to be printing ebooks and selling the copies. I want to provide a legal, convenient, low-bandwidth, affordable way to do that, so that decentralised, on-demand copy-shop printing becomes an integral part of our book supply chain, rather than a revenue leak.

There is no good reason most book content should travel as paper when it can travel as data and be printed in front of the customer. And any environmentally minded point of view will see the absurdity of shipping books and trucking them across the vast distances of most developing countries.

So, what content? In addition to the usual load of public domain, open-licensed and free content, Paperight will have the ability to sell the right to print commercial content, a little like the Espresso Book Machine. The difference is that our model is highly decentralised, and requires no special hardware or paid software.

In return for using lighter DRM (or no DRM at all) than the EBM, and a simple, clear legal agreement with registered copy shops, we get massive benefits of scale by decentralising distribution. We can let any copy shop register for free, and let end-users decide what production values they can afford.

And we can get a wide range of institutions to use the system, because of course there are many kinds of copy shop.

For instance, as part of a small pilot we’re running in South Africa, with the support of Creative R&D and the IDRC, we’re already getting interest from departments of education and universities about how they could use Paperight to distribute course material and produce textbooks on site. And we’re looking into ways creators of open-access materials can use Paperight to cut project costs and distribute more effectively.

If we have time, I can answer questions you might have – I sometimes get questions about security and legal issues, but mostly people ask me how we're going to make any money from the system. Well, there’ll be a few ways, but for now I’ll say we’ll take a small cut of commercial sales, and we may develop customised add-ons to the system for institutions for a fee.

Beyond long-form reading

I want to wrap up with a last observation about digital publishing in developing countries: developing countries have never had an established market for print books, so the digital publishing trajectory there is going to be different to the developed world.

I don’t envy first-world publishers who have to wrestle with their existing print-centric market’s inertia and uncertainty. We won’t see a similar print-to-ebook transition in Africa, because we’re just skipping the whole print-book-market thing. Sure, conventional print will be there in our first-world bubbles, and textbooks for schools will be in print for a long time yet. But beyond that, expect the unexpected.

Patrick and Barry Kayton are entrepreneurs in Cape Town developing a fabulous platform called Cognician. Cognician is hard to explain in the time I have, so I’ll just say it’s a whole new interactive way to explore content that we normally consume in book form. The Kayton brothers mentioned to me recently that Gutenberg didn’t just give us the printing press. He inadvertently replaced much of the interactive, one-on-one teaching and learning that had gone before with the solitary act of reading alone.

It is very likely that Africa will skip the Gutenberg effect completely, and rediscover what digitisation really offers: which is a more social, interactive way to share complex information. If that does take off, whether on mobile phones or otherwise, then we'll see some really exciting models emerging in what is right now a market waiting to be tapped in really creative ways.