Tabblo’s approach to rich Internet apps

Community-developed RIAs are harbingers of things to come

If you want a peek into the future of RIAs (rich Internet applications), take a look at Tabblo ( The model that Tabblo has set into motion for photographers -- both amateur and professional -- will soon be adopted by enterprise IT to empower its user base.

Tabblo allows its members to store photos and text online and use its online photo-editing application -- either on their own or in collaboration with other members -- to create a finished album, book, poster, postcard, presentation, or portfolio.

The technology behind Tabblo’s RIA is not unique. The company made a big bet on commodity hardware, a big bank of image servers, and open source software. And when a request is sent in from any browser, it gets farmed out in a stateless way.

"We are borrowing architecture from big scientific compute clusters, parallel processing," said Antonio Rodriguez, founder and CEO of Tabblo.

What is unique is how Tabblo wraps it all together, creating a specific context for photos, text, and templates and enabling its community of users and developers to help build the finished product.

Most of the features Tabblo develops -- from tags to the ability to create photo variations -- starts from a single designer. The richness of these features, however, comes from Tabblo’s community approach. A developer builds just enough of a tool to see how it is used, and others can examine how the tool was built and then add to it.

"This community approach will sweep software development," Rodriguez says.

Of course, this is the same idea behind open source. But open source focuses more on infrastructure. With RIAs, the technology is more immediate. Developers instrument the application, publish it, and see how users interact with what they’ve built, creating a quick feedback loop.

"We launched our solution seven months ago, and we learned that what a soccer mom in Iowa wants is not what a graphics designer in New York needs,” Rodriguez says.

If it is true that the personal computer put about an 8 percent increase in worker productivity into the economy, Rodriguez predicts that building applications interactively using an open approach will put in another 8 percent.

The result is better software that is cheaper to build -- no packaging to pay for; no warehouses to store it; no truck to deliver it; no stores, brick-and-mortar, or even Web-based, to sell it.

The line between big desktop applications and online applications is becoming blurred. Soon, we will move everything to the cloud, including compute cycles and data storage. Excess compute cycles can be used for more than just searching for extra-terrestrial life. Not to mention the fact that, if everything finds its home in the cloud, the device you use to access it becomes totally irrelevant.

The downside of a Web-based application is that if you lose your Internet connection you are hosed. But the same could be said for the telecommunications network. You just have to trust that 99 percent of the time you will have a dial tone.

The more I look at companies such as Tabblo and at SaaS (software as a service) as a delivery model, the more I begin to see that we are being inexorably driven into the cloud. And if that means better applications that are easier to manage and use, I don't think it is such a bad idea.