Let’s be honest, listening to people talk about tweeting for 8 hours gets kind of annoying. Given that I had never previously attended a Twitter conference I thought I would at least give one a shot. Thankfully the people over at Modern Media brought a Twitter conference to D.C. and it wasn’t as hype-filled as one would expect. At the end of the day though what we’re always listening for is valuable anecdotes and Twitter is the perfect platform for sharing those.
This afternoon Rohit Bhargava wrote, “Tweeting can be like putting $$ in a tip jar when no one is looking. You won’t always get credit for your own brilliance.” It’s a true statement and hundreds of these anecdotes streamed through Twitter throughout the day. Unlike a large portion of the attendees who were glued to their favorite Twitter desktop client throughout the day, I opted out of the background chatter for the most part.
I also happened to miss the keynotes from Laura Fitton and Craig Newmark and all the morning panels. However spending at least a few hours in a Twitter conference is plenty of sampling time. So what was my biggest takeaway? It wasn’t about Twitter. It was that rather than printing out entire conference program guides which can cost you thousands of dollars, you can use bookmarks and save a bundle! The other lesson is related to how Twitter data is being archived.
There are countless products that are each storing all of the tweets coming through the public Twitter stream. How that information is accessed depends on each product’s implementation. The most obvious system for storing and accessing a large data set is Hadoop, which is how most of the companies are now storing their data. However one company I spoke to built their own proprietary system which essentially organizes data based on the expected queries from users. A third company which had an iPhone application decided to cache information locally on the iPhone.
While non-developers may not understand the challenge of storing extremely large data sets, this new ocean of data presents some exciting new issues. While large companies like Facebook, Twitter, and Google are attempting to tackle these issues the applications built on top of them are also facing similar challenges. Anyways, enough about the technical issues of building scalable Twitter applications.
The real question to ask is: has Twitter finally jumped the shark? In the last panel of the day, Steve Rubel asked “Where is Twitter going to get its next million users?” Clay Johnson said it was the exact opposite and that “Twitter could be the next email”. Yes, the traffic charts of Twitter may suggest that Twitter could have already jumped the shark but a product that has brought together a large number of people together at events around the world, couldn’t already have jumped the shark, could it?