« August 2009 | Main | October 2009 »

September 30, 2009

Anyone interested in a Google Wave Scientific Hackfest in London?

Well, today is the rollout of Google Wave proper for a select 100, 000 accounts. If you have an invitation, are up for a bit of hacking and have an interest in creating scientifically relevant applicaitions then Cameron Neylon wants you!

He is calling for interested people to sign up for a science google wave hack day either in London or somewhere near Didcot. I think some people from NPG might pop along. If you are interested go let Cameron know over on the doodle online poll.

September 21, 2009

Joi Ito visits Nature

Last Friday Joi Ito, CEO of Creative Commons (and much else besides) visited our office in London and delivered a talk entitled 'Innovation in Open Networks'. Here are my (incomplete and impressionistic) notes from the session:

Innovation in Open Networks by Joi Ito

Joi used to work in media before the web. Then he jumped ship to the Internet and set up first commercial ISP in Japan.

In the past you had to be in a big company or a research lab to work in information technology. The traditional approach to developing a service that required all possible use cases and errors to be documented in advance. The internet, in contrast, allowed anyone to participate. The open standards and the ability to participate without asking permission were key to its success – "small pieces loosely joined" in David Weinberger's phrase.

The stack: Ethernet (computers) -> TCP/IP (network) -> HTTP (documents) -> Creative Commons (ideas).

Each level in the stack has reduced friction and enabled innovation. Creative Commons (CC) does this at the legal layer.

Joi explained Creative Commons licences. In the analog world for which copyright was originally created, most people don't want or need to make copies. In the digital world, the act of consumption results in copies, so strictly speaking copyright restricts most uses. CC overcomes this by providing licences that don't require lawyers, or even further permission from the copyright owner, to enable reuse.

The W3C manages web standards. RDFa has now reached the highest status of a W3C standard (i.e., 'recommendation'). Among other things, it allows machine-readable licence information to be embedded in a file. This information was previously put into the HTML comments, which was crude and limited to HTML files. Google, Yahoo! and Microsoft have started to parse and use RDFa-encoded licence information. This approach potentially allows fully automated rights clearing. CC is pushing to get this functionality included in HTML 5 too.

Examples of CC users: Jamendo allows artists to share their work. Al Jazeera CC-licenses their Gaza footage. (This has resulted in other sales leads, so the net commercial effect is positive.) Bloomsbury Academic. Cory Doctorow.

CC licensing increases demand but cannibalises sales. So there's a trade-off depending on the demand and sales of any given piece of work at any given time. The questions about CC licensing are therefore practical ones about what, who and when – not a religious decision. (Though for scientific and certain other types of content, CC believes in free access.)

In these situations, the communities often police themselves, enforcing social norms such as attribution and non-commercial restrictions. [Joi gave a Star Wars fan fiction example.] In some cases, fans create derivative versions (e.g., subtitled films) to encourage the publisher to release an official (paid) version, removing the amateur version when they succeed.

CC+ ("CC Plus") is a way of adding CC to other licences. For example, you might use a CC non-commercial licence with another licence that covers commercial use.

Changing business models is hard. In places where there are no business models we often see the most innovation. 'Tecno brega' music parties in Brazil have equipment sponsors, etc. and make a lot of money even though the music itself isn't charged for. These artists treat street vendors of CDs as a marketing channel, not a revenue stream. This approach is bad for traditional record companies.

In his venture business, Joi thinks about where the user perceives value, and tries to be part of that. In the early days of Infoseek, they planned to charge users for each search(!).

Wikipedia is now using CC, not GFDL anymore. This makes mixing of content with other CC content (like OpenCourseWare). This is also reason they don't like 'vanity licences' such as the BBC's, which effectively create multiple incompatible internets for information.

Copyright extension in the US is preventing almost all works from going into the public domain. Lots of orphan works. The GBS settlement is a kludge.

CC Zero: Not a licence, but allows people to give up all rights as far as possible under the law. Good for, e.g., scientific data. (Attribution data can be clumsy – and in some databases exceeds the size of the original data.)

Distinction between the law and social norms (e.g., plagiarism is not illegal but is discouraged by implicit social contracts.

[There was then a Q&A session, including discussion of further scientific examples, but I'm afraid I didn't make notes.]

September 08, 2009

Andrew Savikas visits Nature

Last Thursday, Andrew Savikas, VP of Digital Initiatives at O'Reilly Media and ebook expert, paid us a visit. We had some very interesting conversations about the future direction of publishing, and Andrew delivered a great talk on the topic. He kindly provided a copy of his slides (16MB PDF). My (partial and impressionistic) notes are below.

The Digital Future of Publishing
Andrew Savikas, O'Reilly Media

Printed book sales through retail outlets falling, at least for the computing segment (served by O'Reilly). But ebook sales are really taking off – doubling every 18 months for the last 5 years.

Oreilly.com sales are higher for ebooks than print books by 2:1. These are not cannibalising print, but often reaching new (frequently overseas) customers. Based on a back-of-the-envelope calculation that Andrew outlined but I didn't capture in these notes, the market for ebooks in the US is arguably as big as for all computing books – and it's growing rapidly.

O'Reilly offers its books in EPUB, PDF and Mobipocket (for Kindle). PDF is the most popular, followed by EPUB. They don't use DRM. Bookworm.oreilly.com enables online ebook reading across desktop and mobile, and remembers where you've read to so you can pick up where you left off.

Google now indexes >1 tr web pages. This compares to ~13 bn pages in the Library of Congress. People are reading and writing more text than ever. Connecting your book to the web makes it part of this system. But it also fundamentally alters the dynamics of choice and consumption.

The Long Tail: Books on Safari that sell no print copies still get significant online traffic. This 'tail' accounts for about 23% of all demand. 2/3 of O'Reilly iPhone apps are sold outside the US – another example of the Long Tail.

Do online books have to be free? People already pay for the internet (e.g., $25.8bn in access fees).

Scott Adams' article on free: "Free is more complicated than you think". Nine Inch Nails example: $300 deluxe edition sold out in less than a day despite the same audio being available for free download.

Some O'Reilly books sell very well even though they're free on the web. Real World Haskell was their best-selling technical book despite being free on the web. Content was posted online and received comments. 21 people left at least 75 comments, the level expected by a paid technical reviewer. People pay for packaging and convenience, not just content (analogy with bottled water).

In the developing world most people don't have laptops or even bookshelves, but they have mobile phones. According to IBM Research there will be 1bn web-enabled phones by 2011. Mobile reading is taking off. Stanza downloaded >2m times; Kindle and Sony Reader selling well. For iPhone: The Missing Manual O'Reilly sells more iPhone apps than print books (and it would be the bestselling computing book of all if the app version was included in industry stats). This is true of several O'Reilly titles; as of yesterday (2 Sept 2009), they had 100 iPhone apps. But the iPhone is only 2% of the phone market.

Price-sensitivity is higher online, but these are additive sales, and different offerings allows price discrimination. Example of Twilight app from Stephenie Meyer.

O'Reilly market share in print has grown despite the availability of online books via Safari. The same will be true of mobile, and more people will read books in digital form than in print.

Now all publishing is digital, and all writing is for the web. Ebook content without links looks 'naked' next to a blog entry. Examples of online communities: BookGlutton.

Joe Wikert of O'Reilly Media: "The first TV shows were basically just radio programs on the television, until someone realized it was a whole new medium." We're just at the beginning of a transformation as important as Gutenberg.

"Nascent Web publishing efforts have their genesis in a burning need to say something, but their ultimate success comes from people wanting to listen, needing to hear each other’s voices, and answering in kind."
Rick Levine
The Cluetrain Manifesto

Subscribe

Subscribe to this blog's feeds:

[What is this?]

The Life Scientists on FriendFeed

Recent Comments

Out of 405 total comments.
The most recent three were on:
Powered by
Movable Type 3.2