Recently, I have been working a couple of projects that have pitted commercial software against Open Source Software (OSS) during an evaluation phase to decide which to use in an enterprise CRM application. Initially, the commercial software won because of reluctance to change the selection process, misunderstanding of the real completeness of a COTS product (most COTS products require some degree of customization – the amount depends on the enterprise architecture and the complexity of the API), an over reliance on the face value of information provided during the presales phase of acquisition, and finally, the focus of the Open Source developers to provide an engineering framework and not a finished product (Although this is changing as more as more OSS developers recognize the commercial value of their product e.g., MySQL, SugarCRM, and Talend).
The adoption of OSS in the government is gaining momentum. OSS is reaching significant penetration well beyond its traditional IT infrastructure domain and is moving into applications, business intelligence and customer relationship management (CRM). Drivers of OSS adoption are also changing, witnessing increasing user maturity. The lower cost and local economic development benefits that drove early adopters have been replaced by total cost of ownership (TCO) considerations and the desire to overcome procurement complexity.
Currently, there is little expectation that open source will really free users from vendor dependence. However, an important aspect is the emergence of collaborative development efforts between agencies that use an open-source development process. The other side of the coin is; have you ever acquired a COTS product for an enterprise application that did not require engineering or significant “customized development”? Even such stalwarts as Microsoft are designing there signature products (such as Office) as platforms capable of being engineered, configured, and developed into enterprise applications.
FYI…. The commercial product first chosen, had to be abandoned due to cost and non-adaptability to the overall architecture…
