Michelle’s Blog header image 2

Technology that Will Fuel the Real Time Web in 2010

December 20th, 2009 · View Comments · Social Media and Society

Being around so many talented professionals in the online space allows me to see what’s coming down the pike sooner than the average bear. I’ve been exposed to some projects that can only leave me optimistic for what the future has in store since working for Rackspace Hosting.

2009 was the year Iranians stood up to Ahmadinejad for all the world to see on YouTube and Twitter. 2009 also featured the first president to communicate via email lists and YouTube. We saw Sarah Palin leading the charge against him via Facebook. News like the Fort Hood shootings is regularly broken on Twitter far before it hits CNN.

Anyone who thinks social media will ever “slow down” or “go out of fashion” is simply ignoring empirical evidence to the contrary.

I often don’t talk about the technical side of this phenomenon. Since I don’t see many who do, I figured I would this time. Here are some trends that will fuel social media adoption and groundswell. If you think of others or have any additions/corrections to these, please comment.

1.) Dynamic frameworks like Tornado or Hadoop. Social networks are actually pretty hard to build. Why? If they get adopted, there is a lot of simultaneous traffic going on at once. Friendfeed decided to create a web application framework called Tornado to handle their traffic. They then were bought by Facebook, who decided to open source it for others to use on their social networking projects.

2.) Non-relational databases like Cassandra. A traditional database language like MySQL has a hard time handling all the calls back and forth from a web application. Remember the Fail Whale? That’s because when Twitter took off, there really wasn’t a lot of technology accessible to the average startup that could handle all the data going to and from their servers. As non-relational databases mature, it will be easier for social networks to handle heavier loads of data transfer at the same time.

3.) Cloud computing, cloud computing, cloud computing. If you build a solid application with a solid database, but your web hosting can’t handle the traffic, your site goes down. There is no such thing as unlimited hosting for a finite dollar amount. Cloud computing allows you to handle the spikes associated with the real time web without downtime or having to overbuy. If your site is pummeled by Twitter traffic, it will now be able to handle it.

4.) The software community builder that is the API. 2009 saw LinkedIn open its API. Facebook developers also gained ground into the walled garden that is Facebook and Twitter saw an explosion of applications utilizing their API. The easier we make it to digest social networks and use them the way we want (mobile or otherwise), the more they get used.

5.) Silos like Gnip. There is a lot of data going in and out of a social network. A single API can’t handle it, so applications can use silos to better handle all the people wanting access to data. For example, if you want to build a social network that pulls tweets about only certain topics, you would pull the data from Gnip who is pulling it from Twitter. This fuels niche social networks only looking for data about certain things.

Share and Enjoy:
  • Digg
  • del.icio.us
  • Facebook
  • Google Bookmarks
  • TwitThis
  • LinkedIn
  • Reddit
  • StumbleUpon
  • Technorati

Tags: ···

View Comments so far ↓

  • 1 Jason Cohen // Dec 20, 2009 at 7:52 pm

    Agreed. This isn’t a fad any more than “having a web page” was a fad of the late 90s.

    Corrections: Swarm and Hadoop aren’t databases; they’re distributed computing. Another good example of an open-source distributed non-relational database with commercial backing is CouchDB.

    Extra plus one on cloud computing. Not using cloud computing is like not using virtual machines in 2008. The objections are essentially nil and the advantages are many — even if you don’t need to scale!

  • 2 Michelle // Dec 20, 2009 at 10:59 pm

    Thanks, Jason. I’m still a bit deep in the stack for my comfort, but it’s really interesting!

  • 3 Oscar Del Santo // Dec 21, 2009 at 9:13 am

    Things are moving fast in the online world, and I have the great feeling that 2010 is going to be a great year for all of us.

  • 4 Dave Kellogg // Dec 21, 2009 at 7:30 pm

    Agree with you on the non-relational database angle, but remember that open source isn’t the only way to go. While key-value stores are great for some tasks (e.g., building an index of the entire web), other non-relational options such as XML databases/servers make very good platforms for web 2.0 applications as well.
    .-= Dave Kellogg´s last blog ..Quick Thoughts on the Oracle Hatchet Job on Mark Logic =-.

  • 5 Social Media – 17 Lessons I Learned In 2009 « NetWorks! Boise by Sales Results LLC // Dec 28, 2009 at 10:49 pm

    [...] Technology that Will Fuel the Real Time Web in 2010 (michellesblog.net) [...]

  • 6 Sency // Jan 11, 2010 at 7:03 pm

    Business Insider reviewed 7 real time tools for publishers

    http://www.businessinsider.com/7-free-real-time-web-tools-for-publishers-2009-12

    interesting to see a lot of hype behind this niche

Leave a Comment

blog comments powered by Disqus