Matt Asay, of OSBC and Alfresco, makes a very telling point in a recent post: One power of open source (if done right) is its suitability to interoperability and extensibility. As Matt states:
. . . let me give him/you an idea of what we’re already doing in this space. It’s not a question of what we might do, but what we’re already doing. You can get Alfresco integrated with Asterisk (VoiceRD from Novacoast) and SugarCRM (CRM) today. (And since our 1.4 Business Process Management release, we already have BPM in spades.)
Now extend this. Add some JasperSoft or Pentaho for Business Intelligence (perhaps reporting capabilities). Some DimDim for web conferencing. Some Zimbra or Scalix for email/collaboration. Want to scale this out on a grid? Get yourself some 3Tera. Etc. The great thing about all of this is that we don’t have to do all of it ourselves. In many instances, enterprises are already extending Alfresco (or these other projects) to meet these and other needs. Hence, when a large pharmaceutical/medical devices company wanted wiki functionality in Alfresco, it didn’t ask us. It just built it in.
One could certainly make the argument that first-generation open source like Linux was adopted for cost, risk and code-access purposes, and that second-generation open source like JBoss or Red Hat was adopted because of completeness and support across a broader portion of the stack. But I think what we are now seeing in third-generation open source efforts like Alfresco or LogicBlaze is the enterprise-scale integration and interoperability of components.
Open source combined with open standards avoid vendor lock-in and points the way to a very, very different application and deployment paradigm: identifying, evaluating and glueing, rather than baking the cake each time from scratch.
