pipes creator here: sad to see pipes shutdown. i am surprised it lasted this long -- it was abandoned years ago. on the positive side, it is nice to see it mentioned in the same sentence as y! maps which used to be a big deal back in the day.
Pipes has long been a sort of monument to the faith of a bygone age -- the first flush of Web 2.0 optimism, when standard formats and open APIs were going to let us mix and match and mash up services of all kinds at will.
That faith is dead now, of course, scoured from the earth by walled gardens and VC money. But like Catholics in Elizabethan England, some of us quietly tend our secret shrines and pray for its return.
(Hopefully that will work out better for us than it did for them.)
Thank you for your comment, for a (long) while I thought that I was the only one remembering those times and especially the ideals people were bringing up then. Sometime in 2005 I even copy-pasted this quote on my blog, high in believing that the future will belong to open data and open information:
> Bosworth advocated an open model for data.(...)
Imagine if you can query any data that is available anywhere in the world. Bosworth said that what this requires is a single, simple, open wire format for items. The format needs to be simple for any P programmer to deliver and any JavaScript programmer to consume. He also pointed out that "complex things tend to break and simple things tend to work." Google has the simplest query language in the world. There is no structure and no syntax. (my note: at that time Bosworth was VP of Product Management at Google)
Love your analogy, sure made me laugh (visualising a geeky dev like myself tending a weathered gnarly monument in a secluded English glade... in real life, that is, not in an RPG).
However, why such pessimism? I hardly think that standard formats and open APIs are dead. Sure, there are some nasty blights upon cyberspace in this regard (ahem, Facebook, Apple), but other players are still keeping the dream alive to some extent (e.g. Google, Yahoo).
I'd count google mostly as an example of deprecating or limiting useful interfaces. Yes, you can feed them stuff in "open" formats, but getting stuff out is harder. First example that comes to mind is youtube APIs, which offered easier access to stuff. Now try to get a feed of someones videos or similar...
But it is possible that it's confirmation bias + uncommon sample on my part.
Yeah... for Google / YouTube, I blame OAuth for most of the difficulty in accessing data (while retaining one's sanity) these days. Once you've waded through the quagmire of access tokens, auth libraries (in different languages), account permissions, etc... making the actual API calls isn't so hard.
Let me just tell you and anyone that worked on the team, thanks so much for it. You saved me hours of work and let me do it really damn well. This is a sad, sad day. I used Pipes extensively and ironically enough at Yahoo for some features/prototypes on News and Search. It was such an elegant and powerful app, I really wish Yahoo had given it more love.
It really would be oh so nice to have something that can ingest RSS and Tweets and then apply filters to them while visualizing the logic through a graphical interface. Maybe even some fuzzy machine learning to use in building filters. Some more advanced dupe technology would go a long way. People would be building their own techmemes in no time. You could build ranking systems that award bonus points if something like longform or techmeme mention something.
Maybe im dreaming about a free version of Percolate, which I have heard is really cool.
This is exactly what we are doing at http://www.feedsapi.com , we don't use a graphical UI for the logic but keywords and RegEx to create filters... Many folks are already building their own techmeme with our API, it's however not a free service. I'm not sure if this fully fits your requirements but you should take a look.
One of our open source projects[1], lets you consume RSS, Email, and Tweets, with more "stuff" coming, and apply filters and set triggers, etc. It's still pretty raw, and we don't have the really awesome visual interface for laying things out, but that sort of thing is something we hope to get to eventually. One other neat thing we do is pipe content through Apache Stanbol to do semantic entity extraction, and build up a knowledgebase as content is consumed.
Founder of Enginuity Analytics here. You can ingest RSS and Tweets into our platform where we apply machine learning to provide social data, sentiment analysis, rankings, filters and demographic analysis of all the content for 27 languages. And easily export the data to Excel and more. http://theenginuity.com
Instead of closing it down, releasing the source and giving it to the community could breath live into it.
I remember being blown away by it, the fact that it was neglected and just kept on says a lot about it's good design.
We (Webflow) are actually considering building a Pipes alternative in the near future, and I'd love to sit down with you and brainstorm some ideas. If you'd be willing to make some time to see a very rough demo, I'd love to get in touch - my email is in my profile.
Several reasons I can think of. For one, no one at Yahoo seemed to understand what Pipes was for (part of a wider problem where Yahoo didn't seem to know what the company was for).
For another, Yahoo legal prevented us from doing any kind of long-term storage of RSS feeds. That made the service far less useful than it could have been.
Finally, they didn't give it any resources at all. Management was too preoccupied with the failed Brickhouse project (internal startup incubator) and the host of failed ideas that it spawned.
Filtering and triggering without the annoyance of having to write code in some syntax-of-the-month or dredge through poorly written documentation for half-ass APIs.
If you are redistributing, hosting or using it commercially you have to have the agreement of the author of the content. RSS just gives you access in another way than HTML, but it does not circumvent copyright.
This is an amazing product. Have you tried talking to VCs or anyone about someonr adopting it somehow?
I wonder if it would be possible to build something similar using IPV6/some kind of P2P/STUN/TURN/UDP hole punching/swarm/ethereum / any way to make a distributed system that does the same thing. Or maybe just volunteer running Node on their own VPSs.
