eBooks Nasty Little Secret

In PastLife™, when I was a Program Manager (aka PM) at Microsoft, one of the interview scenarios I used with potential college hires was around eBooks. It was a great way to understand how candidates thought about, and broke down, unfamiliar problems. It also provided lots of opportunities for discussion across the whole gamut of PM skills – from UX design to diving down into technology to the business of software. It was also an excuse for me to keep a stack of CDs next to a 1940 edition of The Ilford Manual of Photography on the shelf as a visual cue: “CDs have gone digital so why not books?”

At that time eBooks were a niche. The Palm Pilot had had them from the late 90s and Microsoft had the floundering Microsoft Reader application but there wasn’t the ubiquitous eBook experiences that we have today. Which is why it was such a fertile area to explore in an interview.

Then along came Amazon with the Kindle.

The Kindle is a great product (as are it’s imitators). It has done much to spur a revolution in how readers consume the written word. Critically, a key part of the Kindle’s success is the integrated end-to-end model that Amazon provides: the device, the software, the content and the services that make that content easy to access.

But there’s a snag.

The problem isn’t the reading experience –  the eInk displays are just fantastic – although many still prefer paper. The problem is the formats often used to store eBooks use a technology called DRM to make sure that only the people that are licensed to can copy them. Actually, I should say that DRM tries to ensure that only the people that are licensed to can copy them. DRM systems aren’t perfect and sometimes you can’t access something you have rights to, and that’s the rub of the problem.

DRM (an abbreviation for Digital Rights Management) provides a way for someone to protect digital content by encrypting it. Before that content can be used, it needs to be decrypted. DRM systems include mechanisms to check that the person, or device, using the content is allowed (i.e. has a license) to use it. No license? No decryption, and the content stays locked away and unusable.

DRM has been used in a range of places. It protects music, DVDs and BlueRay disks, sensitive documents and, more recently, lots of those eBooks that more and more people are stuffing onto their Kindles.

So, what’s the problem? We’ll here’s a couple:

  1. DRM systems are, by their nature, closed systems. This means that the DRM used to protect content is tied to the devices that consumers use. This means that, for example the FairPlay DRM system used by Apple for iTunes is tied to Apple devices. If I have some content protected by FairPlay then I can only access it on an Apple device.

    This also creates interesting problems for people using Open Source software like Linux since DRM systems rely on obscurity but Open Source is open to everyone.

  2. The DRM system often needs to check that you are still allowed to access the content. To do this it needs to ‘phone home’ to check that the license you have is still valid. In some cases this can result in the content being inaccessible because of some delay in performing those checks.
  3. A company that sold you some content could have second thoughts and prevent access to something you’d purchased at a later date.
  4. A company could go out of business or close down the product leaving you unable to access the content you’d purchased.
  5. You will die. When you do who inherits your collection of DRM-locked books and how do they read them?
  6. Technology changes rapidly. I can still read my 70 year old copy of The Ilford Manual of Photography but given the rate at which we currently make technology redundant, it’s unlikely that will apply to a DRM-protected digital equivalent.

Not convinced? Here are some examples…

DRMfail

What did music do?

In many ways the book publishing industry is where the music publishing industry was several years ago.

The first CD I ripped to MP3 was, ironically, one of the Sony protected disks that didn’t work in our car audio system (although sold as a “CD” the disk didn’t comply with the relevant standard, known as the ‘Red Book’). For years, the music industry persisted with selling DRM-protected music as ‘digital downloads’. The justification given for this was to protect the rights of producers and to prevent piracy. However, it quickly became apparent that the impact of DRM was on those that had purchased the music. Those that didn’t want to pay had many ways of avoiding the protection, leaving the inconvenience of imperfect DRM systems to be felt by paying customers. Obviously, the wrong way round to have things if you want to keep valuable customers for your product.

By early 2007 Apple, estimating that some 97% of music on iPod devices was *not* protected by DRM, put pressure on the major labels to allow the iTunes store to sell non-DRM music. As a result, market pressure quickly led to the removal of DRM, resulting in the situation we have today where most music sold is not DRM-protected.

The music experience is quite clear. DRM doesn’t protect business models that the market no longer wants to support, but DRM does inconvenience those that legitimately pay for the content.

Given that, are we seeing the publishing industry making the same mistakes as the music industry did? Certainly at the moment it looks that way, but there’s a couple of reasons why publishing may get away with DRM in the short term:

  • Where the content comes from: The availability of source content in a digital form as the CD disk, a format that could easily be converted to MP3 and shared quickly, led to a breakdown in the traditional business model for the music industry. Books aren’t the same. Converting a paper book to digital by scanning is a costly and time consuming process. This barrier to conversion means that the majority of eBooks consumed on eReading devices come from publishers (like Amazon or Barnes & Noble) rather than being converted by the end-user from an existing library of content.
  • Devices that people are using: because end-users buy content through eBook suppliers (like Amazon) rather than converting their own existing content (as with CDs) there’s a tight link between the device people use to read and the services that provide the content. For example between a Kindle device and Amazon’s Kindle store. This means that relatively few people switch between devices from different providers, which in turn means few people have encountered the problems switching between different DRM systems.

Solutions?

Here’s some things you do (if you can think of others, I’d love to hear about them).

  • Stick: stay with paper books. Most people can figure out how they work, and they tend to keep on working.
  • DRM free: Buy DRM-free eBooks – for example from the some publishers that are considering DRM-free or providers like Project Gutenberg.
  • Put up: Go ahead with DRM eBooks and the potential consequences from all that brings.
  • Techie: There’s also a technical solution: it’s possible to remove DRM. Whether that’s legal depends on where you are. Whether that’s within the terms of the license for the content depends on where you purchased it.

Assuming you’ve paid for the content, all this is rather depressing when you think about.

More reading

Advertisements

JCQ Not Open

This post was going to point to the JCQ data on 2012 GCSE exam results. Instead I’ll link to the Guardian Datablog post (which has a link to a Google spread sheet version of the data).

Why so?

The JCQ is the Joint Council for Qualifications (see Wikipedia for a description). It acts as ‘a single voice for the seven largest qualification providers in the UK’.

The Terms and Conditions for the JCQ website include this clause:

2.6 You undertake not to:

[…]

establish a link to this website from any other website, intranet or extranet site without our prior written consent;

Complying with this clause means not *establishing* a link with the JCQ site [1]. A rather strange clause to include in the T&Cs for *using* a web site. The result? I guess it makes JCQ a single voice that’s not part of the conversation.

To quote Nate Anderson in Ars on a similar issue:

But those wanting to link to a normal Web page on the site certainly don’t, as a general rule, need permission to do so; indeed, the Web would be a hugely different place if linking were permission- and form-based. One can see why Lowe’s likes such agreements, but it’s harder to see why anyone would sign one.

I wonder if they really mean this. I’ll ask.

In the meantime…

Footnotes

[1] I suppose there’s an interesting legal discussion somewhere on whether including an HTML anchor tag is *establishing* a link (or if that’s what happens when the browser makes an HTTP request to the webserver hosting JCQs site). Irrespective