Web 2.0 guru Tim O’Reilly can barely contain his enthusiasm for Yahoo Pipes, calling it “a milestone in the history of the internet”. It is certainly fun to play with: using a drag-and-drop visual interface, you can combine and process data from multiple internet sources. Within minutes, I was viewing my blog in French, by piping the output from its RSS feed into the Yahoo Babel Fish translation module.

The Pipes editor makes impressive use of JavaScript to create a designer reminiscent of Apple’s Quartz Composer where inputs and outputs are connected by dragging lines between connection points. It is an interesting prototype, though limited in its first release, but is it really an internet milestone?

O’Reilly says Pipes is “turning the web into a programmable environment for everyone”. I think his spectacles are rose-coloured. In particular, Yahoo Pipes does not address the core problem of the service-oriented web: the business model. Participating in mashups works well for e-commerce sites like eBay or Amazon, because it drives sales, but that model fails for other kinds of services.

Giving developers programmatic access to data means they can strip out advertisements or other unwanted elements, in the same way viewers skip the commercial break when watching a video recording. Some will see Yahoo Pipes as just another way to steal content, which explains why Yahoo’s troubleshooting notes describe how to block your data from Pipes users.

In December last year, Google withdrew its Soap search API in favour of a JavaScript widget that users can embed on their sites. The likely reason is that Google gets no benefit from programmatic search access, whereas the widget gives Google full control of a little corner of the host web site. Yahoo’s Babel Fish service is another example. This is a free service on Yahoo’s site, and there is a widget available, but no API. When asked why, Yahoo’s Kent Brewster said, “It’s complicated; machine translation is expensive and tough to monetise.” Currently, it is not even clear how Yahoo intends to monetise Pipes itself. No business model means no long-term future.

Yahoo Pipes is undoubtedly worth a look. It is easy to get excited about the possibilities of snapping together new and insightful applications by combining data using the lingua franca of RSS. Some of the concepts can usefully be applied internally, as businesses discover new ways of publishing data internally in enterprise portals and mashups. Just don’t call it a milestone.