Facebook Platform API

So I finally got around to playing with the Facebook Platform API yesterday and my initial take is that there’s a lot of potential to leverage the API and extract and transform the data the resides on Facebook. Of all the API functions, I think ‘fql.query’ alongside understanding the Facebook Query Language gives you the most power to pull and analyze the social and marketing data that resides on their servers around millions of users.

Simply having the ‘uid’ of a user, along with the ‘Users.getInfo‘ function, gives you a ton of information about that user and the only limitation is the user themselves. Unless a Facebook user realizes the security concerns of the data that they’re sharing, their information is free for anyone to see and extract.

PHP is the language of choice for most Facebook developers I think (myself included) but I decided to take the ASP.NET / C# route to see how I could integrate it with Consona CRM. The Facebook Developer Toolkit, which just released a new version (lucky for me) is fairly comprehensive but because it’s so new, the documentation needs some polishing up and there’s not many code samples out there yet around it. I created a quick and dirty solution that takes the customer information and passes it to Facebook to either pull up the customer Facebook profile information or run a search and see if that customer is a Facebook user. One issue that I’ve run into so far is the around the session data. If a parent frame has a different domain than the child page in IE, the session data (stored in the Session object) is not preserved as a security precaution. So unless I preserve the initial session that is created, the user currently needs to login each time a new customer record is pulled up.

Ribbit API

I’ve been fiddling around with Ribbit the past couple days and trying to figure out how to integrate it with the Consona CRM platform. Ribbit has an incredibly full featured API and is developed primarily with Adobe Flex so I’ve had to do a crash course on the side but it hasn’t been all that bad. The Adobe Flex Builder leverages Eclipse so if you’ve used Eclipse IDE for development in the past (Java, Aptana, etc…) it’s really a piece of cake. The SDK that Ribbit provides also has some great examples.

Functionally, Ribbit easily replaces the Consona CRM CTI toolbar. What I need to figure out is how to integrate a user logging into to Ribbit as a user of the Consona system and then be able to track incoming and outcoming phone calls to the customer records in CEDB. Unfortunately I’ve been sidetracked with some other issues the past couple days so I haven’t been able to give my full attention to this side project but I’ll post updates in the future.

What has suprised me is the lack of attention that Ribbit has received in the marketplace. It was acquired by BT a little while ago and they’ve been quietly been release newer versions but the potential of this API is great.

AuthSub token has wrong scope

I was playing around with the Google Data API last night and I kept running into this error. The app I’m creating allows me to pull down a list of Google Calendars and then insert a new entry into the calendar that I select. Pulling the list was the easy part, adding a new entry was the tough part.

My problem was that I was declaring a ‘scope’ that was too limiting for any additional calls to the API. Instead of declaring the scope for ‘allcalendars’, I reduced the scope to the largest URI possible which is ‘http://www.google.com/calendar/feeds’

Be sure to read this this doc about using AuthSub with JavaScript before diving into the Google Data API http://code.google.com/apis/gdata/authsub-js.html

Google App Engine

I see Google’s new App Engine as a way for Google to employ thousands of new developers. Also instead of paying them, the developers pay Google to use their servers and manage their scalability. Unlike Amazon’s Web Services which breaks it into separate components with EC2 and S3, you are kinda restricted to using Google’s GFS and BigTable. Currently Python is the only available scripting tool but when they start offering PHP and Java, I think development will really take off. Once a company creates an amazing application with this new tool, Google can simply acquire that company and not have to worry about retooling the new application to work in their environment.

EveryBlock Chicago

Content aggregation is pretty popular these days. Just look at the last post that I had regarding FriendFeed. While FriendFeed is for social media, EveryBlock is for news for a particular city. Right now it only gathers news for Chicago, New York, and San Francisco so the audience isn’t too large (large in the sense that it covers the entire world considering Chicago, San Francisco, and New York are in the top 10 largest cities in the US). The site was created by a local Chicagoan that created the Chicago crime Google Mashup awhile back. He took that tool and rolled it into the site and now includes content like photos, street closures, missed connections, etc… Anyways, check it out to see how your neighborhood stacks up against the rest of the city.

FriendFeed

I’ve been getting into FriendFeed lately as has most of the Web 2.0 industry. Most are either really excited about it (e.g. Robert Scoble) or just see it as another social aggregator to add to the mix. I’m unsure at this point what camp I’m in. Similar to the blogging world where there are dozens of tools like Vox, Blogger, MSN Spaces, etc… it’s hard to decide what to use as well as which has a broader audience.