Anti-social media

Thursday, June 19, 2008

I saw a post on TechCrunch this morning about problems with the Google App Engine.  The Google App Engine is an application development / hosting platform offered by Google for free (up to a pretty generous bandwidth / storage limit).   Like many other free offerings from Google, I love the philosophy behind Google Apps --- it is democratizing (meritocratising?) while at the same time offering a practical solution to real problems faced especially by small businesses.  In theory, a small group of smart people can develop the next killer app, offer it as a Web service and start a business, all without buying a single server.  A host of other companies, especially in the social media space, seem to offer similar benefits by providing a Web based service along with an API to let developers create applications around the service.

However, as the problems with the Google App Engine (and some other services such as Twitter and Flickr) make clear, there is a definite risk to developing applications that reside outside the walls of the enterprise.  I'm reminded of a 2005 article called Six Things you Should Know About Bubble 2.0 (Andrew Orlowski's ironic name for the Web 2.0 hype).  In point #6, Orlowski argues that most of the folks behind Web 2.0, social media and cloud computing applications in general are not up to the task of building reliable, scalable applications.  He goes so far as to suggest that the software engineers involved are not serious and not knowledgeable about system-level programming and robust, fault-tolerant design.  This seems to be me to be going a little far, in addition to being a needlessly ad hominem argument.

In searching for Orlowski's article in order to link to it, I stumbled across another whole set of sites and blog commentary that point out an entirely different set of issues with social media.  One of the key critiques of social media made by sites like AntiSocialMedia.net is that the democratic, participatory nature of most social media sites, along with their underlying assumption that most people have an honest contribution to make is highly vulnerable to small numbers of people who are highly motivated to be dishonest.  The focus of the pages that I read on AntiSocialMedia.net was on stock manipulation by fake comments left on financial site message boards, but I also saw links to content about fake Amazon.com reviews and other similar for-profit hoaxes. 

All in all, I think it would be a mistake to assume because of outages and scalability issues that the idea of cloud computing is invalid. Similarly, I think it's clear that software as a service (and applications built on social media APIs specifically) are here to stay.  The key will be for organizations to make smart decisions about how they make use of these opportunities.  Should a Fortune 500 company build mission critical applications on the Google App Engine?  No, not given the security and uptime concerns that those companies have.  Does it make sense for small or medium sized firms to build non-mission critical applications or even mission-critical but non-uptime critical applications that make use of free APIs and cloud resources? Sure, as long as they have thought through what they will do when those services and resources are not available.  In many ways, this is exactly like decisions that we've all been making for a long time about Web and email hosting --- do I do it myself, or outsource?  If I outsource, what are my contingency plans?  As long as those are in place, expectations are set reasonably, and there is a plan to immediately communicate with users about outages, I would not hesitate to use the free services that are out there. 


Posted by: Mark Reichard at 11:46 AM
Tags: Social media

Comments: