ITGumbo: spicing IT up

Since it seems as though questions, which two to three years back centered around whether open source is here to stay, are now focused on what's next for the open source adoption wave. While there are certainly differences that exist cross-industry, there are a number of absolute realities that will govern how fast/solidly open source continues up the application stack.

As adoption strategies mature and more open source products prove to be production quality, I expect further saturation and increased penetration along the lower (categories 1,2 and parts of 3 below) and higher parts (pieces of category 3 and 4) of the application stack, respectively. However, I also expect that the penetration rates for this "upper echelon" of the application stack will be exponentially lower in comparison. For one, I have a hard time seeing how the JBoss model (lower cost, open source disruption) will be duplicated in this phase of open source growth...if only for the simple fact that the technologies involved traditionally take shape as horizontal/industry specific solutions. So commoditization won't play as much a factor as it did lower down the stack.

1. Web/network layer: Server operating systems, Web servers, File systems/sharing, etc.
2. Application platforms: Servlet containers, app servers, development tools
3. Infrastructure: Business Intelligence, content management, databases, ECM
4. Business apps: CRM, ERP, industry specific apps

Still, the following will remain steady:

• Open source continues to emerge as a strategic play. As awareness regarding open source shifts, the interest level in how to further exploit it as a competitive advantage has also grown. This coincides with higher levels of interest in using open source technologies for application and data services (see category 3 above). Something which bodes well for open source BI, ESB and portal players.
• Lower costs will continue as a main driver. There's no need to expound about this one. So while cheaper doesn't always equal better, cheaper + same level of quality = more value.
• Open source consumption (by enterprises) will dwarf production. A number of individuals, including myself, have voiced the value that contribution has within the context of an open source ecosystem. Unfortunately, I don't see the typical consumer-focused stance of organizations changing drastically in the near term.
• The desktop remains its own domain. At best I feel that the first significant wave of open source enterprise desktop initiatives are at least 3 to 5 years away. The challenges in this domain center around bringing a full stack of software that will run on user desktops. Additionally, the prospect of migrating users and installations from existing Microsoft productivity tools to open source alternatives remains overtly daunting. So while a number of local and central public authorities across EMEA are open to national and EU objectives which call for increased use of open source in requests for proposal (RFPs). I'm not sure if the open source desktop will experience any residual effects of increased adoption levels within the other categories listed above.

Nonetheless, even as open source moves up the stack, the most challenging aspect of open source adoption is not technical. Simply put, it is the shift that it brings to the enterprise software business model. As concerns about security, intellectual property and support continue to keep a more sustained progression up the application stack at bay. Since at the end of the day, clean code compiles don't overrule the fact that technology adoption is swayed by political and even cultural concerns.