Image by luc legay via Flickr

Two very simple and powerful APIs keep cropping up in discussions at Gist: the Twitter API and the Google Social Graph API. These are very significant for identifying people in the arriving “mash and pivot” era of Web services. Humans are uniquely identified on the internet by their email address or URL (OK and maybe their Facebook user ID). From these “keys”, other information like photos, addresses, and phone numbers flow surprisingly freely. This is a glimpse into what the semantic Web will be like. Humans on the internet currently enjoy “privacy through obscurity”. In the near future, the internet is going to become a very public phone (and photo) book.

Lets say for example, all you have is an email address (as is often the case when Gist first discovers someone). The Twitter API lets you look someone up by email address. If Twitter finds a match, you will often immediately have not only a link to the person’s blog and their most recent status update, but oftentimes also recent GPS co-ordinates.

Here’s me, in XML looked up by email. You don’t even have to log in to access this simple machine readable representation of my recent location.

That’s great, but not everybody is on Twitter yet. Email lookups can also be done against much more mainstream sites like Amazon (for wish lists), and Flickr (for photos). How about if all you have is a URL (like to the person’s blog or their public LinkedIn profile). The Google Social Graph API lets you look up related sites by URL. Here are other sites about me that Google found from my public LinkedIn Profile. Here’s the JSON for the the machines reading along.

Now, I’m a very sharing person. But this resolved my public LinkedIn profile into my favorite FriendFeed posts, as well as my music listening habits. All this personal data is made available with very simple APIs built on Web standards, and microformats based on existing HTML markup. The point is that none of the information is centralized, and all of it can be assembled very easily with great confidence about its accuracy.

Some of the Gist developers shared Pizza with with the TalentSpring team the other day. Interestingly, the conversation focused on data partners and Web crawl rather than Web frameworks and software. There was some talk about Amazon’s EC2, which we’re all excited about, but much more talk about data providers like ZoomInfo, Jigsaw, LinkedIn, and Facebook. These are the sources of the the company and personal data that Gist and TalentSpring build upon.

I think this is a challenge a lot of software developers are facing as we move into the “mash and pivot” Web. All new Web apps are built on top of other Web services and data sources these days. The code (Ruby, Python, C#, Java) that glues services together is important, but starting to play a proportionately smaller roll to the interfaces and data that make up modern Web applications. Depending on an application’s problem domain, developers will usually be faced with choosing the best API (usually on the basis of data provided and coverage of their anticipated market). New applications add value by wedging into a particular niche and adding value by re-mixing. Gist gives its users a competitive edge by giving them the first look at the phone book of the future.