News Credibility, Verification and the Madness of Crowds - A Junk News Roundup

As the Associated Press states in our News Values and Principles:

"We have a long-standing role setting the industry standard for ethics in journalism. It is our job — more than ever before — to report the news accurately and honestly."

It is easy to see how AP is taking concrete steps in this area by, for example, our fact checking work (online, on twitter). And the AP Verify project is building a "newsroom tool that will combine artificial intelligence with our editorial expertise to automatically source and verify user-generated content."

I thought it would be interesting to take a look at some efforts going on elsewhere in the areas of credibility, verification and identifying junk news.

Standards Efforts

The IEEE is working on P7011 "Standard for the Process of Identifying and Rating the Trustworthiness of News Sources". The IEEE is a formal standards body, responsible for many of the technical standards which underpin the internet.

The Credibility Coalition describes itself as "an interdisciplinary community committed to improving our information ecosystems and media literacy through transparent and collaborative exploration." It is not, in itself, a standards body. If you examine the CredCo "about" page, you will spot my photo - I attended early meetings.

The Credible Web W3C Community Group describes its mission as "to help shift the Web toward more trustworthy content without increasing censorship or social division." There is a significant overlap between members of the Credibility Coalition and the Credible Web Community Group.  Despite the W3C link, this is not a formal standards effort - Community Groups are open to anyone. There are weekly video conferences to define an informal standard.

The Trust Project describes itself as "a consortium of top news companies" and says it "is developing transparency standards that help you easily assess the quality and credibility of journalism." Again, the Trust Project is not a formal standards body (like IEEE, IPTC or W3C).

Verification Projects

At the recent IPTC meeting, we saw presentations about two European projects aimed at helped to identify the spread of misinformation.

Truly Media is a joint project between ATC and Deutsche Welle. It is a "a web-based collaboration platform developed to support primarily journalists and human rights workers in the verification of digital content," and was developed with funds from EU and the DNI.

InVid aims to develop "a knowledge verification platform to detect emerging stories and assess the reliability of newsworthy video files and content spread via social media." It is an EU-funded project. Their demo was quite sophisticated. They also have a browser plugin which lets you verify news video and images yourself.

Wisdom and Madness

Finally, via Fair Warning, I saw "The Wisdom and Madness of Crowds" - a fun explainer in the form of a game. It walks you through why some crowds turn to madness and some to wisdom, with a focus on the spread of misinformation but also good information. It helps give some insight into the different dynamics at play and even some suggestions for how to reduce the spread of junk news and amplify the spread of verified news.

Making Progress on Rights - W3C Permissions Obligations and Expressions First Public Working Drafts

I've been working within W3C's Permissions & Obligations Expression (POE) Working Group as an Invited Expert. We have just issued our First Public Working Drafts:

• The ODRL Information Model
• The ODRL Vocabulary & Expression.

"one" by Andre Chinn
https://flic.kr/p/5pGcyx

The W3C POE WG aims to create recommendations for permissions, obligations and licensing statements for digital content. The WG is using the W3C ODRL Community Group specifications as the starting point for its work. These are the same specifications which form the foundation of IPTC's RightsML work.

"poe" by 为民 王
https://flic.kr/p/gp2Bc

If you're interested in digital content, then I recommend looking at - and commenting on - the W3C POE drafts. The ODRL Information Model describes the foundational concepts, entities and relationships of ODRL. The ODRL Vocabulary & Expression describes how to encode the ODRL model in XML, JSON and RDF.

"Use in case of emergency" by Katia Sosnowiez
https://flic.kr/p/5MMhFz

The POE WG has also published the Use Case and Requirements Note. I have contributed one of the Use Cases: News Permissions and Restrictions. Again, the Working Group is looking for feedback on - and contributions to - the Use Cases, so that it can derive a detailed set of requirements for the POE work.

Mining for eBooks

In February 2013, the W3C, in partnership with IDPF and BISG, organized a workshop on eBooks, in conjunction with O'Reilly's TOC. I was invited to speak about AP's and IPTC's experience with implementing permissions and restrictions with machine readable rights. (ePub lets you include DRM statements; it seems that some publishers are using ODRL v1; IPTC have selected ODRL v2 for the foundation of RightsML). It was a great experience being on the panel and I got a lot of thoughtful and interesting questions.

eBook Readers Galore by libraryman
http://www.flickr.com/photos/libraryman/5052936803/

eBook Newbie
I'm a bit of a eBook neophyte. However, I learnt a lot from hearing the other publishers talking about their experience, hopes and frustrations with this digital publishing mechanism. And it struck me how similar the news industry is to the book industry. In his opening keynote, Bill McCoy talked about the three main ways that publisher deliver books these days: files, apps and websites. Of course, these are also the three main ways that news is delivered, today, too (not to mention dead trees in both cases for non digital publishing). As various other speakers presented at the workshop, they repeatedly used examples from newspapers and magazines (although, sometimes, as illustrations of what *not* to do). And, it has to be said, both book publishers and news publishers are in the same boat of trying to figure out their digital futures.

For more about both the eBook workshop and TOC, I recommend Ivan Herman's reflections.

Evolution of Readers by jblyberg
http://www.flickr.com/photos/jblyberg/4505413539/

Mining for eBooks

Given how easy it is to create and publish an eBook, it would seem that mining a news archive could yield some interesting books. Some news publishers are already conducting experiments with ebooks in this way. For example, the UK's Guardian have a series of Guardian Shorts. (Martin Belam wrote some quite interesting articles about how he worked with the Guardian archive to create ebooks on the Internet and the Olympics). Similarly, Vanity Fair have also started to play with ebooks.

Of course, ebooks aren't the only way to make use of a rich news archive. The New York Times recently launched their TimesMachine which lets you see browse back issues between 1851 and 1922 (“all the news which was fit to print”).

As software continues to eat the world, it will be interesting to see how formerly different kinds of publishers converge and diverge in their attempts to make their digital ways.