2006 - The Year of User Generated Content

It is hardly surprising to call 2006 the Year of User Generated Content, what with rapid adoption (and rich valuations) of YouTube, MySpace and the like - consumer-generated content. Couple that with business user generated content - where the content is commercial , even market oriented, and its impact on the year is even stronger. I would even go so far to say that electronic markets are based on user-generated content - where the content are commercial offers - ranging from lists of offers (e.g. Craigs List) to bids and offers, from ebay to Wall Street.

The technology impact of this is important - namely:

• I/O: Frequent updates as opposed to frequent reads add real architectural challenges. Typically scale out and caching solves the frequent read problem pretty well. Frequent writes requires a whole another approach - with risks of dirty data everywhere (particularly if performance is an issue - and where is it not?)
  
• Code segregation: For social networking applications likeMySpace, this enables users and contributors to self-style their pages. In the 1st generation of Web 2.0 this means segregating UI from styling (via CSS for example) but for 2nd generation Web 2.0 (should this be Web 3.0?), this concept extends further - to business logic, to data base queries, etc. 2006 was for 1st generation - expect 2nd generation to be significantly more complex - scaling to enable user styling was hard enough.
  
• Data synchronization: Related to the 1st issue. With high write rates, keeping data fresh is pitted against keeping it fast.
  

From a business perspective, what matters is monetization - and how does this content generate revenue? This need not be evil nor underhanded - infrastructure enabling such content generation and distribution is not free. For electronic markets, the content makes the market - the market generates revenue for the operator. For non transactional systems, revenue comes from leveraging the network effect. This network effect benefits everyone - users, aggregators and those that want to sell to the aggregated users alike.

And 2007? Is Social Commerce Coming?

I see very many of the same things happening in 2007 in the area of Social Commerce - where Web 2.0 meets E Commerce. While the details and challenges differ, my guess is there will be distinct technology and business model challenges.

But first, a Social Commerce overview:

While originally popularized by Yahoo Shoposphere and currently being pitched by Gartner and others, it seems to be a combination of social technologies, such as blogs and wikis and networking technologies and commerce, commercial transactions. Some categories are pretty well established:

• Affiliates and referrals (popularized by Amazon)
• Network marketing

while other are emerging, such as:

• Group Commerce, whether within groups (physical or virtual) or between them
• Social mashups involving commercial transactions

But however it surfaces commercially with whatever monetization strategy, in terms of technology, look for heavy borrowing from Web 2.0 (wiki's, blogging, Ajax) plus:

• Distributed transactions - like high frequency/volume of updates, this poses distinct scalability challenges - both in terms of managing state and transaction locks but also propogating transaction state information across a wide number of participants and systems. Under such conditions, a potentially huge number of events and messages could be generated by system and participants alike.
  
• Complex User Interaction models: Along with the potential need to coordinate transactions across independent participants, comes the need to coordinate the participant collaborations as well.
  
• Greater code segregation models: As discussed above, advanced Web 2.0 (Web 2.0 Gen 2) needs modularization beyond UI layout, it needs segregation of business logic from UI and business logic from data access. This is true for social commerce as well. In fact, social commerce is part of Web 2.0 Gen 2.
  
• More nimble UI's: UI's need to be aligned with the user category and be able to be adjusted as user categories change and new ones come on line. Consumer interfaces are obviously different than those for enterprise and SaaS (Software as a Service). This is nothing new - what is becoming increasingly important however are for applications and services to service multiple different user segments. Look to code segregation - discussed above - to be a key enabler.