Today is the day when MetaCert comes out from behind the curtain of 'stealth mode' and goes public with its first offering.
What is MetaCert?
It's a source of data about Web sites — descriptions of online content in terms that matter to people:
- is the site free from malware?
- is the site owner correctly identified?
- does the site contain sexually explicit material?
- does it meet accessibility standards?
The vision is large scale. As everyone knows, there are billions of Web pages so providing accurate descriptions of them, even in just these terms, is a huge undertaking. But that's the task that MetaCert is beginning today.
Who is Behind MetaCert?
In short, Paul Walsh.
I first met Paul in June 2005 when he and I both joined the W3C Mobile Web Best Practices Working Group. His company, Segala, provides accessibility testing for Web site owners (notably O2) and the close link between accessibility and mobile means that he's always seen certification of sites as being mobile-friendly as a future business opportunity. I was there because I wanted to promote the publication and consumption of accurate metadata about Web sites. Paul and I came from different perspectives but clearly had a lot in common.
Roll on 6 years (has it really been 6 years!) and here we are. The underlying technology for MetaCert is, of course, the Protocol for Web Description resources, POWDER. Segala, helped to instigate the work that lead to the definition of POWDER and I'm obviously delighted to see the standard being used in this way.
This comes on top of a couple of other recent activities to use or recognise POWDER as a useful tool: the Ontology for Media Resources 1.0 recently became a W3C Candidate Recommendation and the Wholesale Applications Community, WAC, is using it too (see my notes on this).
Will It Work?
I fervently hope so. That's why I'm delighted to be part of Paul's team. He has a sound business model that will underpin a service that can be of genuine benefit to end users. It's not about labels and filters — that's what ICRA tried and failed to achieve. It's about providing trusted data about online content and doing it in a way that makes sense. The data MetaCert creates will be enormously valuable.
MetaCert launches today with its first products: the adult content labelling service and the Firefox browser add-on that recognises those labels. The add-on comes into action when the user does a Google search. Links to sites with adult content labels are annotated with a simple icon. Hovering over that icon provides more information.
As well as annotating search results, the add-on can block sites labelled as containing adult content and there are links to MetaCert through which users can classify a site as XXX (this is verified before being added to the database so it can't be used maliciously).
The range of icons and information provided by the add-on will increase quickly over the coming months and of course there will be add-ons for other browsers, but for today, we launch with the XXX icon and the Firefox add-on.
The reason for beginning with this product is simple enough: MetaCert has a multi-million dollar contract with ICM Registry, the company behind the soon-to-launch .xxx top level domain. Every sexually explicit site on the .xxx domain will be labelled by MetaCert and we have spidering technology that constantly crawls adult sites and the similar sites they link to. That's going to be a big number very quickly.
How we're going to do it, well, that's commercially confidential of course, but the whole point of MetaCert is that it will very quickly become a source of accurate, trusted metadata about the Web. It's yet another commercial venture based on Semantic Web technology.