(What follows are the text contents which I submitted to the on-line application form at http://www.uutisraivaaja.fi, the innovation contest by the Helsingin Sanomat Foundation, on March 1st, 2011:)
The name of your organisation, company, community, or project
First and last name
Phone number (International format)
+358 50 59 33 006
Kaislapolku 9, 52700 Mäntyharju, Finland
Name of your project
The annual budget for your project
The amount of money you are applying for in the contest
Duration of the project (1 to 5 years)
The entire budget for your project, including other funding sources
If you have any other funding sources for your project, please list them here
Describe your innovation: (max. 1800 characters)
Creating a hyper-personalized recommendation engine for news and information.
This approach is new and unique in that it emphasizes the aggregation and analysis of “gestures”, leveraging the Social Web.
Gestures are ways in which web users respond to the information which they encounter, thereby indicating relevance. Gestures may include: subscribe, read, store, share, tag, rate, copy-share, send, comment, blog, micro-blog, pipe-through, link, approve/reject
(see also: http://josschuurmans.com/a-hierarchy-of-gestures-for-the-holy-grail/).
Every gesture contributes to a collective human news filter. The aggregate data can be used to inform a personal news offering to individual users.
When users sign up, they will be asked about Internet destinations which capture their gestures.
A user’s blog is a good example as it contains a lot of strong gestures in the form of embedded links. Other examples include Twitter streams, shared reading on Google Reader, comments on WordPress and Disqus, Facebook’s “likes” and “shares”.
For every pair of users in the system, the engine will calculate the proximity of their past gestures.
If user A and user B have a high proximity of past gestures, the next gesture user A will express will be highly relevant to the engine’s recommendation of news and information to user B; and vice versa.
People also want to find and receive news and information which they were not aware they would be interested in. One envisioned feature on the system is a “serendipity pot meter” or slider, which will allow the user to fine-tune the serendipity of the news and information recommended to them.
How does your innovation improve the distribution of information? (max. 750 characters)
“Give me the information which is most relevant to me; nothing more, nothing less.”
This has been the unattainable Holy Grail of the Web for as long as it has existed. Only now, thanks to a new abundance of on-line social news gestures, is it becoming possible to deliver on that vision.
This innovation of a hyper-personal news recommendation engine improves the distribution of information by giving Web users exactly what they want (with a grain of serendipity – but only as much as the user allows).
Tell us why your idea is groundbreaking. (max. 750 characters)
1. From a socio-technological point of view, this idea is groundbreaking in that it leverages social news gestures. They are the so-far un-leveraged traces of our collective news consumption and appreciation.
2. For a user, this idea is groundbreaking as it offers a new and superior news consumption experience. Only signal, no noise. As New York University Professor and Berkman Fellow Clay Shirky has famously remarked, the problem is not information overload; the problem is filter failure. This engine addresses the problem of filter failure.
3. From a business standpoint, this idea is groundbreaking as it offers news organizations a sophisticated means to give users exactly what they want. The system will offer data to inform publishers not only as to what to offer on their web sites now and to whom, but also as to which news stories their users are likely to appreciate in the immediate future.
Tell us how your idea differs from existing services. (max. 750 characters)
Existing concepts of personalization on news publications are usually based on categorization and editorial selection. Those concepts are not sophisticated enough – they lack flexibility and scalability – to really offer a hyper-personalized experience.
Somewhat more sophisticated systems may analyze readers’ clicking behavior within a web site and compare page content in order to inform them about a more personalized news offering. These systems however usually work only within one (or a limited number of) publishing web site(s). They have a great challenge striking the right balance for serendipity, and since they don’t take social gestures into account, they can’t offer the level of hyper-personalization which we propose.
What kinds of experience do you and/or your organisation have that will enhance the implementation of your idea? (max. 1600 characters)
After six years at Nokia’s corporate communications team, Jos Schuurmans started Cluetail Ltd. in 2009 to help organizations and individuals extract more value from the conversations in which they engage on-line.
Trained in journalism and social sciences, he has been professionally engaged in developing and managing Internet- and intranet-based media concepts since the early 1990s.
At Nokia, Jos was global concept owner of ‘Nokia-internal news’ and the ‘Central News Desk’, a company-wide platform for internal news creation and distribution. The ‘News Hub’, Nokia’s news and conversation intranet site, where Jos was a concept developer and editor, was granted an Award of Excellence at the CiB Awards 2007, by the British Association of Communicators in Business.
As an Associate Producer at Talentum Group, Jos and his team pioneered the first-ever online news service delivered simultaneously as a Website, an email newsletter, a WAP service, via SMS, to Palm OS, Windows CE, and ‘thin HTML’ tailored to Nokia Communicator devices. The service won the NetMedia European Online Journalism Award 2000 for best technology reporting.
‘City OnLine’, an Amsterdam-based Internet portal, part of publicly listed Wegener NV, where Jos was a concept developer, reporter and editor at the time, won the Interactive Publishing Award 1998 for the best European Internet community website.
Languages: Dutch, English, German, Finnish.
(more on http://linkedin.com/in/josschuurmans)
The business plan for your project (Tell us how you will make your project sustainable)
I – Why does this matter?
Here is why this innovation is relevant to any news organization:
* People are increasingly less loyal to their newspaper subscription.
* The production tools for news have been democratized.
* The cost of publication and distribution is approaching zero.
* Paid subscription as well as advertizing revenue is under pressure.
* There is hyper-competition, not only from competing professional news organizations, but also because, to borrow from Clay Shirky’s book, “here comes everybody”.
II – Sustaining the innovation
The recommendation engine will be open-source. Its creation and further development will happen in an open-source project.
The engine will be deployed in Finland first.
A foundation may be established in order to safeguard the open-source nature of the project and the recommendation engine. The foundation would manage the project, including the acquisition, coordination and allocation of resources.
The foundation would be a not-for-profit. Sources of income could include consultancy fees relating to conceptualizing and implementing the engine at individual (news) organizations, as well as financial support from partners and sponsors.
III – So what’s the business model? (for news organizations / news brokers / news inter-media-ries)
1. The snarky answer to the news organization would be: Well, can you afford *not* to do this? What do you think will happen if you don’t? Only those news organizations who adapt to the new reality in which news users are no longer loyal to a single, or handful of subscriptions/channels, but instead want to taste all kinds of flavors when they “consume” news and information, and who want their information diet to be hyper-personalized (because that’s the only way to solve the problem of information overload or filter failure) – only those news organizations will have a chance of surviving and thriving. Those who won’t adapt will be obliterated …and end up in the Museum of Media History.
2. As this is really the Holy Grail, something that everyone wants and has been anticipating since the introduction of the graphical browser in the early 1990s, whoever deploys it first will enjoy significant positive brand exposure. Provided of course that the user experience is done right.
3. We live in an attention economy, says Edelman consultant Steve Rubel. One type of value that this system adds is the value of attention. Integrating it with existing news publications will add significant attention value to those publications. It will retain users.
As per Dan Gilmor, director of the Knight Center for Digital Media Entrepreneurship at Arizona State University’s: If you give people good links, they tend to come back for more. Well, this engine is about giving people good links.
4. One way to make money is to apply a “freemium” business model: provide free access to basic features and premium access to advanced features.
5. An iTunes-like pricing model: € 1,– per pop (e.g. for a browser session of max 24 hrs. of hyper-personalized news consumption)
6. Sell special-purpose mobile applications through the iPhone app store, Android store, Ovi store…
7. Advertizing revenue.
8. Use data mining to predict news demand. The system offers unique insights into information demand from news users. This intelligence can be used to predict which news will become a trending topic in the immediate future, guiding the next news offering and the allocation of editorial resources.
9. Use data mining to guide the placement of (banner) ads.
10. Sell data mining reports, focusing on specific topic areas for various industries.
Note: These examples of value models should not be seen as endorsements per se. Any of these implementations would have to be judged against the business objectives and the stated moral values of the news organization that considers to deploy them.
Anyone, including any commercial ventures, will be able to deploy the open-source software for free, and build commercial offerings on top of it.
News organizations would seem to be the most natural agents to deploy the personal news recommendation engine.
(What follows is the text content of the email message that I received in response:)
Olemme vastaanottaneet hakemuksenne Helsingin Sanomain Säätiön Uutisraivaaja-kilpailuun. Kiitos osallistumisestasi. Viimeistään maaliskuun aikana kuulet, kuinka ideasi menestyy.
Jos sinulla on kysyttävää kilpailusta, lähetä sähköpostia osoitteeseen email@example.com
Helsingin Sanomain Säätiö
We have received your application to the Uutisraivaaja innovation contest of the Helsingin Sanomat Foundation. Thank you for participating. We will inform you about the progress of your application by the end of March 2011. If you have any further questions about the contest or about your application please send an email to firstname.lastname@example.org
Helsingin Sanomat Foundation