By Herring on Apr 05, 2009
There is a great saying that it took a long time to invent the wheel, but replicating it is easy! I think the main challenge with proprietary stacks is exactly this -- paying for a business model on a wheel that in its day was revolutionary, but today is old and not effective. Does your car still drive on a granite wheel -- didn't think so! The open source "wheel" has taken the best designs from the proprietary world and enhanced them, and in doing so created some great new solutions at lower costs.
What I am not suggesting is rip-and-replace! But I am suggesting you look at your whole application needs and see when open source alternatives like GlassFish will work.We have seen some really interesting cost savings by using GlassFish over the proprietary alternatives -- savings of up to $3M in some instances.
Now before you think that open source is just for those hot web 2.0 garage startups.. consider that T-Mobile is using GlassFish with some great results.
"High availability allows us to meet our stringent uptime requirements and the Sun GlassFish Enterprise Server enables us to cost-effectively deploy new services while meeting our performance and availability requirements," said Erez Yarkoni, vice president, T-Mobile, USA.
The other issue to consider with proprietary vendors is lock-in. Basically for the proprietary vendor lock-in is key to success. With lock-in they become a sole source vendor for your needs and they can charge what they like -- and may have!
The combination of high-cost, proprietary products and vendor lock-in frequently constrains businesses from embarking on new software initiatives -- something that customers I speak to can't afford to have happen.
Additionally, the capital expenditures and associated financial risk required to deploy proprietary products can either delay the profitability gains of new software initiatives or simply prevent enterprises from attempting innovative ideas to drive new revenue streams.
For example, proprietary products from BEA Systems/Oracle are far more expensive to acquire than open-source alternatives and don't offer enterprise developers the flexibility to customize to fit the business' changing needs. Although developers can submit feedback, the cycle of development is long and, with a limited number of engineers working on the software, new features and updates are delivered much more slowly than in open source.
To ensure that problems can be addressed as quickly as possible and to reduce operational costs, enterprises can chose a comprehensive open-source platform backed by an established commercial entity that provides support and understands the interdependencies not only of that platform but of other third-party products that are already in the enterprise's IT infrastructure.
I invite you to share your story, or read other stories from real users: http://blogs.sun.com/stories/