User-Generated Media - Challenges & Solutions for Business and Academia

Social networking and user-generated content (UGC) sites present unique technical challenges, which lead to unique business challenges.  While unexpected growth is a potential problem for any online site, it is both the holy grail and (in the spirit of "be careful what you wish for") a ticking time bomb for social networking sites. 

A new whitepaper from Akamai (also available free from streamingmedia.com) goes into some depth about the special factors that affect social networking sites.  Some highlights:

• User-generated content sites are the fastest-growing category of web site (by unique visitors) on the Net, showing, in some cases, triple digit year-over-year growth. Of the ten fastest growing web brands, five are UGC sites (for example, Flickr and Wikipedia). 
• Social networking/UGC sites have, by definition, unpredictable storage and bandwidth needs, making technical infrastructure (and therefore, budget and capital expense) planning a crap shoot.  Outsourced capacity on-demand is an important option to consider before you're faced with site-crippling runaway success. 
• Success is tied closely to having a fast innovation cycle -- try stuff out, see how it works for your users.  Continually sense-and-respond to user needs to find that sweet spot of simplicity, functionality, and sustainability that makes your site sticky and social.  One way to do this is to minimize the time and effort you put into infrastructure build-out and put it into more creative endeavors. 
• If you're an ad-driven site, performance is directly tied to revenue, as faster loading pages keep eyeballs on the site, lead to more page views per user, and therefore register more ad impressions.  When Friendster moved to Akamai's delivery network in March 2006, they saw an immediate 33% decrease in page load times, and a threefold uptick in page views.

Even for an educational institution, outsourcing certain infrastructure is appealing.  With service-oriented Web APIs, it can be easier now to work with a vendor/partner than it is to build it myself.  If I want to put up a quick video recording/encoding/sharing service for my users, I can:

• Build it myself - not always a bad idea, and definitely a quick-and-dirty solution for a pilot or proof-of-concept, provided I have to staff and the time to move it from P-O-C to production-ready if the need arises.  
• Acquire and deploy an inexpensive product.  I was surprised to find YouTube clones like Clip'Share and Altrasoft VideoShare for a few hundred bucks or less.  Again - good for a proof-of-concept.  May or may not offer enough for coping with real success.
• Use a Web Service API like that from Video Egg or JumpCut to handle all the media operations, while you focus just on your website.  These services handle media input (in the case of Video Egg, from webcam and cell phone, as well as file upload). transcoding, online editing and delivery.  It can provide a platform for rapid development of your own custom solutions, as well as a scalable solution in case your solution takes off.  

I'm generally a big fan of institutions building their media solutions in-house, but the combination of the unpredictable needs of user-generated media, the ease and excellence of some of the vendor service-based APIs, and the need to be able to innovate quickly without up-front investment in big infrastructure creates some interesting possibilities.  

The Akamai white paper, Successful Social Networking and User-Generated-Content Applications: What You Need to Know, (which, by the way, I wrote) addresses some other challenges of social and UGC sites -- how edge-caching works with dynamic content, how to control costs when growth is unpredictable, options for exercising editorial control over UGC sites, and some examples of how social networking is being used by businesses to build revenue and create new opportunities.