I played around with Pipes and aside from loosely matching iTunes songs with YouTube videos or loosely matching Flickr photos with NY Times articles, I’ve yet to ‘get it’.
The interface itself is pretty – and seemingly powerful and simple. Reminiscent of Microsoft’s Visual Basic or Apple’s Automator. But in the end, Pipes is chartjunk – or as Dave Winer said, “[a] visually appealing but information sparse IDE”.
If the goal of Pipes was to simplify the process of mashing up and filtering feeds – they left all the hard parts in; the geeky terms (“Base”, “Position”, “Query Parameters”, “Debug”, “Truncate”, “For Each”) and the URLs themselves.
Seems to me, the same functionality could be achieved with any number of the other feed aggregators out there (My.Yahoo.com comes to mind).
If you’ve been able to do something cool and useful with Pipes – let me know. I’d like to be proven wrong here (and move this post from ‘useless’ to ‘useful’).
Alex Iskold says Pipes has turned the web into a big database…I think we already had that with XML-RPC, RSS, SOAP, and REST.