Wow Tim, you really touched a nerve

From the "Been there, done that, how many times?" department: Tim just posted this to Oreilly Radar:

Yahoo!'s new Pipes service is a milestone in the history of the internet. It's a service that generalizes the idea of the mashup, providing a drag and drop editor that allows you to connect internet data sources, process them, and redirect the output. Yahoo! describes it as "an interactive feed aggregator and manipulator" that allows you to "create feeds that are more powerful, useful and relevant." While it's still a bit rough around the edges, it has enormous promise in turning the web into a programmable environment for everyone.

Wow, Tim, I would say "don't get me started," but its too late.

I almost don't want to mention contemporary examples of this idea, because it doesn't illustrate just how old an idea is, plus these examples are recent enough where it is plausible that Tim may not have heard.

I will start with Plagger. Plagger, invented by Tatsuhiko Miyagawa of Six Apart, is a configuration file driven Feed scripting language written in Perl. Plagger is easily extensible and pluggable and allows people to chain feeds through a variety of services and filters. Everyone at Six Apart uses it everyday and probably doesn't even know it. It is open source and freely available. Is it visual? No, but in some respects it is probably easier to use. As scary as that might sound.

On the visual side of things I might point out Coghead. Coghead is not focused purely on feeds, but the concept is similar. Take all these XML-based services that are out there and stitch them together.

But the nerve Tim touched was the fact that I spent four years of my life working at a company building exactly this: a visual, drag-and-drop XML scripting framework. I remember being pitched on the idea by Ken Norton (who later became the Product Manager behind Jot in a small cafe in the financial district of San Francisco. He described a system not unlike the Unix command line - a simple framework through which one command can "pipe" their output into the input of another command, and so on. He described it as an "operating system for the Internet." Tim describes it exactly the same way:

I want to talk about the implications for that marvelous aspect of the fundamental UNIX design: the pipe, and its ability to connect small independent programs so that they could collectively perform functions beyond the capability of any of them alone.

gc_logo.gifI thought it was the coolest thing I had heard in years. It was 2000. The company I joined was Grand Central Communications, where over the course of four years we built a product similar to "Pipes." It was not structured around RSS exclusively, but rather SOAP and any other flavor of XML you could think of. The company failed - perhaps because it was way ahead of its time, but perhaps also because while interesting, no one really wanted to pay for a product that ultimately is more work for the user then its worth - a fact that even Tim cops to in a way. That despite the visual interface, a lot of technical knowledge is still required. In the end it is far easier to stitch or "mash" services together in code then trying to map inputs and outputs via XSLT or some other mechanism mapping engine do in code.

But Grand Central took it one step farther by making these composite services callable as well. They were meta web services in that you could send a message to a endpoint that actually did nothing with the data directly, instead it just cycled your message through a series of services, transforming it as it went, and then returning the resulting value.

A powerful concept, even more so in 2000 when it was invented. It took three years for the technology to come around for us to build a drag-and-drop tool to compose these processes. But we did.

And that was Grand Central. There was also General Interface which produced the first Ajax IDE ever. Period. It was an awesome product. It too was ahead of its time. It was doing Ajax long before people knew what Ajax, Ruby on Rails, Web 2.0, JSON etc were. This IDE worked and looked just like .NET, except that it was hosted online. It did not give users a drag-and-drop RSS scripting language, but it allowed people to mash products together manually. And the applications they and their customers built using the tool are nothing short of amazing. Of course most of the apps GI built exist behind the firewall so they can't be demoed easily, but they exist. But take my word for it. The apps took everyone's breath away when they debuted four years ago.

Tim completely overlooks these tools and gives far too much credit to Yahoo for innovating here. And this is especially surprising given that he sat on Grand Central's Board of Directors, so I don't think he can fairly assert that Yahoo has reached a milestone here.

Recommended Entries

4 Comments

That's right, we first pitched Tim on the Grand Central concept very early on, maybe July or August of 2000.

It's true. Tim is kinda a putz. He's been disconnected from the underground devs since his book empire reached it's high. Sucks how the very medium he published about is slowly and surely destroying his core business, huh?

Anyways.

While I totally agree with you that Pipes is definitely and old school concept, you do have to give Yahoo! some credit here. They are helping to bring that old-school concept to the masses. SixApart & Grand Central may have started the idea of "mashing" the data with tools, but obviously, the marketing wasn't there to get the word out. I'm sure you know - even though it's a very, very painful, dirty, nasty truth - that no business can succeed - no matter the idea - without a decent (or, innovative) marketing thing behind it; be it ads, viral, or whatever. Yahoo is doing that, and bringing the idea to life.

Give them a little credit here. They deserve some.

And, of course, none of this is really going to be fun until Yahoo opens it up just a wee bit more, so that you can write these "manipulator" modules as services on your own server anyways. I'm sure that's coming up soon - it's too obvious not to - and that's when the innovation will come in droves - by the people and individuals who use that open system.

But I preach to the choir.

:)

Michael, old buddy, you are absolutely right. Yahoo does deserve credit for build a really cool tool. It looks awesome, and whether it is the product, the marketing, the timing or all of the above, they get all the credit.

The nerve that got touched was giving credit to Yahoo for inventing something they didn't. If the milestone is anything, then it is finally getting people to notice a cool concept that others have been trying to share with the market for years.

Hey Byrne, long time no see.

While what we did at GC was visual, i don't think it reached the level of eye candy that Yahoo Pipes is at with its web-based visual editor. That is what's driving the "coolness" aspect that everyone is raving about. That and I think Pipes is in a bit of a sweet spot where it has appeal to technical folks as well as non-technical folks.

Also, I think you are spot on in that the audience for this type of tool may not have existed just a few years ago, or at the very least -- that audience wasn't ready for it. Fast-forward to today, where blogs, RSS, and other formerly power-user type activities have become known to many more people, and those are the same people who are falling all over themselves about Pipes.

Leave a comment

what will you say?


Recent Comments

  • Hey Byrne, long time no see. While what we did at GC was visual, i don't think it reached the level of eye candy that Yahoo Pipes is at with its web-based visual editor. That is what's driving the "coolness" aspect tha...

  • Michael, old buddy, you are absolutely right. Yahoo does deserve credit for build a really cool tool. It looks awesome, and whether it is the product, the marketing, the timing or all of the above, they get all the credi...

  • It's true. Tim is kinda a putz. He's been disconnected from the underground devs since his book empire reached it's high. Sucks how the very medium he published about is slowly and surely destroying his core business,...

  • That's right, we first pitched Tim on the Grand Central concept very early on, maybe July or August of 2000. ...

Close