FluidDB: One Man’s Mission to Power the Web of Things

After 37 years as a programmer bouncing between gigs all over the globe, Fluidinfo founder Terry Jones, a self-described “hacker” and “flake” who would “never want to be trapped by some corporation” has decided to build a bona fide business. The 46-year-old Australian said in an interview today, “I’ve always described myself as having more potential and less ambition than anyone I knew. I realized in the last four years that I was completely wrong; I’d just never found anything I wanted to really go for.”

Fluidinfo founder Terry Jones

Now things are finally coming together for Jones, who closed an $800,000 round of funding led by Betaworks last week and launched his product, the FluidDB “platform for the web of things,” at TechCrunch Disrupt this week, after 13 years and three complete rewrites in the making.

Jones and a partner in Barcelona have been working full-force on FluidDB for the last three years, and they just hired a third. The going hasn’t been easy; Jones got ahead of himself with the grand scale of his idea, and was rejected by dozens of VCs in 2007, then was labeled as “the unfundable world-changing startup” in 2008. But with Amazon (S amzn) S3 now hosting 100 billion objects, 11 million active MySQL installations and 100,000 Twitter applications, perhaps the time is right.

Jones’ idea has been fairly consistent throughout. What FluidDB does is allow developers to share the information they store in their own databases, so it can be combined and organized in novel ways. Jones described this as “a shared writable object for everything.”

Just like you can put a bookmark in a book to mark your place rather than entering the page number in a spreadsheet, or you can write anything on a Post-it note and stick it to a location where it’s relevant, information stored online should be as “writable” online as it already is in the real world, Jones told me. As for the inspiration for FluidDB: “We were living in read-only or hand-holding world where the only things we could do were pre-approved by our parents in advance.”

The one thing I’d fault Jones for is that he seems to weigh on talking about how “revolutionary” his idea is — he just seems to love that word. And it’s true, FluidDB is a lofty idea, but it’s now something people can actually use. For instance, in a demo at Disrupt, Jones showed how you could query three different Twitter applications in tandem to find influential people who your connections have met and can introduce you to in person. Jones said that Fluidinfo differs from other similar startups like Freebase because he is interested in “small and idiosyncratic data” rather than big data organized in an ontology.

Jones plans to open-source FluidDB within the year — once he hires someone to be a liaison to the open source community. And after that, this is his business plan:

* Cross-licensing deals. FluidDB developers will be able to charge each other to access data with fine-grained permissions.

* Tagging objects with trusted domain names. Just like users are more likely to trust information because it’s on a website or from an email domain they recognize, if developer want to tag an object with their domain, FluidDB will extract a toll. If users know the reputation of your brand name, they’re more likely to incorporate your data.

* Instant API creation. Anything tagged within the system can be organized into an API, so users could jumpstart new APIs especially if they don’t have one already.

Will this latest bit of momentum be enough to push FluidDB through? By now Jones has had enough doses of humility delivered to know these revenue-generating ideas are “a fantasy” until he makes them work, he said. And it’s not like everything’s coming easily; Fluidinfo didn’t even make it into the second round of the Disrupt startup competition. The difference now is that the world is much more ready for what Jones is doing, and now people finally believe in him.

Further reading: Jones keeps a pretty excellent blog.

Related content from GigaOM Pro (sub req’d):

Big Data Marketplaces Put a Price on Finding Patterns