Pipes: Programming for Prosumers

Yahoo! launches the Pipes service for everyone to reprogram resources on the web. It is a visual programming environment for aggregating and manipulating feeds, a “Mashup 2.0” application in a way. Pipes derives its name from UNIX pipeline where the output of each process feeds directly as input of the next one. With Yahoo! Pipes output of web services can easily be hooked up to filters and manipulators, so you can chain a set of web services and processes.

Yahoo! Pipes Editing

A Pipe connecting two feeds from different news alerts based on the search query (simplified version of Aggregated News Alert).

There are already comprehensive Pipe reviews here, here, and here. Also check out the Aggregated News Alert pipe as an example to see how a pipe can be used. I will just point to details I found interesting. Each pipe has a permalink. You can run, clone, view the source of each pipe from the web interface. Any url can be used as an input, so RESTful urls become more important. A pipe can also be hooked up to another pipe as input. Pipes aggregates metadata similar to a usual web application. Run counts and Clone counts make a pipe hot or not. Yahoo! Pipes probably tracks the programming activity such as the counts for most used filters, resources, operators and their combinations. So Pipes aggregates metadata about programming as well.

Yahoo! Pipes will sure open various directions in the informational space. It will also accumulate capital through unpaid surplus labour performed by the Yahoo! users. Tim O’Reilly says: “What’s really lovely about this is that, like the Unix shell, Pipes provides a gradual introduction to web programming.”

Yeah that’s true, prosumers (producer + consumer) will learn how to program, so in a way Pipes enables programming for prosumers. I think Pipes is a powerful step for making the programming as part of the immaterial labor.

Leave a Reply

Your email address will not be published. Required fields are marked *