A project lead by some of the most-respected leaders of the Internet has secured $240,000 in funding to build a prototype system for both expert and peer review of all the content on the web, sentence by sentence. Called Hypothes.is, the project is lead by early search engine innovator and climate change activist Dan Whaley and backed by advisors from John Perry Barlow of the EFF to Garret Camp of StumbleUpon to Kaliya Hamlin of the Internet Identity Workshop to Nate Oostendorp of Slashdot, and many more.
We wrote about the Hypothes.is fundraising effort on Kickstarter last month. That effort succeeded, including with a large matching pledge by cleantech and web venture capitalist Sunil Paul. And so Hypothes.is will be built. What do people want out of it and does it stand a chance to change the web? Opinions differ.
Hypothes.is is planned as a peer review system to check, verify and critique content all over the Web – and beyond. “Improving the credibility of the information we consume is humanity’s grandest challenge,” Whaley says. I can’t help but suspect that Whaley’s work on climate issues, which are challenging to understand, deeply contentious and of the utmost importance to humanity’s future, played a role in his coming to feel this way.
Topic experts will be enlisted in addition to crowdsourcing, a reputation system, browser plug-ins and APIs are on the roadmap and all the data will be stored at the Internet Archive.
It sounds like Google’s ill-fated Sidewiki project and countless other startups that have sought to enable annotation of distributed web content, but it also sounds a lot more sophisticated.
Mark Surman, Executive Director of the Mozilla Foundation, is one of the project’s advisors and it’s not hard to imagine something like this shipping as an optional plug-in for every browser. Google acquired contextual-search plug-in Apture this month and said that functionality like that, highlight a word anywhere to learn a lot more about it, would be baked into the Chrome browser as soon as possible. Why not bake in something like Q&A site Quora, where answers to questions are voted up and down by users and topics are easy to track over time, into every browser regarding every sentence on every web page?
Topic experts could be invited to participate the way that Google did with that Wikipedia-killer project no one even remembers, Knol, but they could also be discovered Quora-like by tracking which users of Hypothes.is got answers voted up the most in reviewing content on a topic in multiple instances.
Expertise may be a subjective matter, though, and is often deeply swayed by persistent participation. Just showing up. Wikipedia, for example, is alleged by some not to be the world’s finest, neutrally written encyclopedia, but rather a domain dominated by pushy know-it-all youngsters who bully their way into describing the world in a particular way.
Hypothes.is offers a list of twelve principles that indicate it’s already looking out for concerns like this. Those principles include the following:
- Open source, open standards.To the extent practical.
- Work everywhere. Without consent.
- Non profit. Sustained by social enterprise.
- Neutral. Favor no ideological or political positions.
- 100% community moderated. Bottoms up, not top down.
- Merit based. Influence based on track record.
- Pseudonymous. Credibility without public identity.
- International. By design.
- Transparent and audit-able. In systems. In governance.
- Think long-term. Infrastructure for 100 years? Or longer?
- Many formats, many contexts. HTML, PDF, video, books. News, blogs scientific articles, legislation, regulations, Terms of Service, etc.
- Work with the best. Remain humble.
If this works, it sounds great. Imagine having a channel in StumbleUpon that just serves up pages from around the web that have been validated by experts in marine biology, for example.
I can’t help but think there will be an enormous user experience challenges, too, though. Most people in the world don’t know what a browser is, they Google for Facebook and they fall for phishing and spam often enough to make such tactics extremely profitable. Maybe something like Hypothes.is could help with all that, but it will probably be a challenge to get a critical mass of engagement, too, across the unbounded expanse of content on the web.
How will unpopular opinions fair? That’s another question I’ve got.
None the less, I’m really excited to see how this project unfolds. I’ve reserved my Hypothes.is username already, so that my reviews of content can appear under my usual handle. Hypothes.is says it aims to have a prototype available early next year.
