Yahoo! Pipes

So just when you thought you were getting sick of the whole Web 2.0 thing and that you'd seen every possible mashup possible, Yahoo! comes up with a new tool--Pipes--to help you mash all by yourself (no servers or complex coding required). It's basically a visual tool for taking various information (RSS feeds, CSV documents, geolocators and more) and joining, filtering, and/or combining them to produce a specific output that you want.

So what can you mash together? Well, the first thing I did was to take news feeds from CNN and Google, combined them together and put them through a filter to pull up everything on the search terms "Tibet," "Tibetan," and "Dalai." (You can tell what's going on at Emory at the moment.) You can see the result here. You could then use this page for a continual update on this subject--or you could use the page's RSS feed in your reader of choice and get the particular news items delivered to you. No more searching through Google or CNN for the particular stories you're interested in (as long, of course, as you know the search terms that will likely appear in the story). You can see also the representation of the pipe in the lower right corner of the page, and if you have a Yahoo! log-in, you can edit the pipe itself to adapt my work and make it more suitable for your own.

But wait! There's so much more. I then decided to take the Emory news feed, filter it to only bring me news related to Tibet and the Dalai Lama and rolled this together with photos on Flickr that are of Emory University and have tags that relate to the same search terms. The results are here. What's new with this pipe is that it also returns a map. One of the photos has been mapped to a particular location within Flickr by the photographer, and Pipes automatically incorporates this information. If you click "Links" about the map, you will see the results: as of today, there are two hits from Emory News and two photos from Flickr. You can again edit the pipe yourself.

I then wanted to create a crazier looking pipe with more variables. The result was taking the news feed from Emory and using all the photos from Flickr that relate to Emory (not just ones with Tibet related tags). You can see the results and edit the pipe, as always.

But there are a lot more tools within Pipes. Say you want to read the sports from Le Monde, but your French is just not where it was back when you had to, say, pass a Ph.D. translation exam. You can get the feed from their sports page, run it through a translation loop, and get the results. It might not be perfect, but it's a sight better than my own translation.

Other functions within Pipes allow you to map things. I've been trying to figure out if I could use Pipes to solve my Timeline XML problem, but that has yet to happen. To get more ideas about how to build interesting pipes, you can check out the public gallery and copy their source and fiddle with it.

In any case it is a fun service to use and a lot easier than similar things I've tried in the past. It appears that the other big names in the online world (Google, Microsoft, etc.) are developing their own similar tools. But now you can be the first on your block to have piped.