Healthcare.gov is in the news all over these days. I have my own opinion about Affordable Care Act ( aka Obamacare) but that is not what I want to talk about. There are enough conversations on it already. I want to talk about technical aspect of Healtcare.gov. I am IT professional and feel obligated to give my opinion on this topic. There is bigger lesson to be learned here and it is important to take notice of this across the corporate world.
Healthcare.gov is built completely ground up using mostly opensource software by number of companies jointly. There are several issues with this approach that resulted in the technical disaster of Healthcare.gov. Lets discuss some of these
Some of you might remember when Airbus was making A380 jumbo jet and it made parts in different countries. When they tried to put it together wings did not fit!! It was blamed on design software incompatibility. Think about the parallels between the situations. All these different companies built different components of software and government tried to put it together as single product. Government acting as system integrator; Really!!
Open source software is greatest thing happened to IT. Companies like Facebook, Google, twitter etc. really showed the world how to use these technologies effectively. All these companies have hundreds and thousands of highly skilled technologists who are encouraged for their creativity and most of the time designing things without requirements. Whenever some of these companies tried to build something with half-baked requirements, it failed. Remember Google built platform for Electronic Health Records? What happened? Open source may appear to be cheap alternative but it is not for everyone and everything. I am sure some of you remember how ERP’s came about. Before ERPs, most companies built there own systems using ‘N’ number of technologies and maintained it themselves. It worked great in the beginning but what happened eventually? It became impossible to maintain. ERP’s came and took the requirements of building and maintaining away from the corporation. Traditional ERP vendors still did not completely succeed. That is when Cloud vendors came in and fix what traditional vendors were unable to fix. Get businesses out of creation and maintenance of software. Now with so much talk about open source, are corporation (Government being one of the largest corporation and healthcare being one of the largest department) going to restart what they abandoned years ago? Are they going to go back and develop their own software and maintain it? Is open source for Non IT focused companies? I don’t know answer. I have opinion but do not have answer. My opinion is, there is still value in working with proprietary software vendors like Microsoft, Oracle, IBM and SAP. Wit IT Focused companies continuing to pressure these companies, these companies will continue to evolve the software and reduce the margins. Think about what powers Salesforce cloud? Its proprietary Oracle Software!!! Salesforce did not rely on open source to scale. Some of you might know this, but twitter ran into scalability issue with their open source implementation and finally decided to move away from it. Now this is the company that is IT focused. They have the talent and ability to attract top IT talent. Why does anybody think this open source movement should be adopted by corporations?
Let me be clear. Open source is one of the greatest movements against monopoly in the software industry. Firefox was first one that started chipping away market share from Internet Explorer. It was complete product and how many companies take the code of Firefox and try to build their own browsers? That is what we are talking about when we talk about using open source to build solutions. It sounds good but the hidden costs of open source over the life of software are high. Some of you remember Sugar CRM? It is not even in top ten vendors in this space, which covers over 75% of the market share. I worked for a startup company and we implemented Sugar CRM in one of our product branch. It worked great. 10 yrs fast-forward, we got successful but we could keep tool up to date. Every hour spent on upkeep of this tool was time taken away from development of core product, which was making company successful. Finally we decided to go with leading vendor product. The moral of the story is, it is very important to understand pit falls of adapting open source software. It sounds very attractive but you need to make sure you understand long-term strategy to maintain these. If you make one wrong decision today, you will end up paying now and later when you decide to finally switch over. Open source is great option for some of the companies but not for most. Negotiate well with your vendors leveraging open source as a lever. Large IT system integrators love the open source movement because that is the next new thing. Do not confuse between open source and cloud. Some of us do. Cloud is not necessarily open source.
What should government do now with healthcare.gov? It is not simple in this case. They need to make some tactical fixes for now. There is no choice but to put Band-Aids. Once it is stabilized, they must analyze the architecture carefully and start replacing components of architecture with proven technologies. Use COTS where available. Government should not be in the business of developing software. It is good business for system integrators but not for the citizens of this country. Long term these Band-Aids are not going to work. So it must act fast and in parallel. I have done similar things in my experience and it requires leadership, decisiveness, and lot of luck. It is like changing tires on moving truck. But it can be done and it must be done. Weather you are for Obamacare or not, it is irrelevant to this discussion.
So, Open source or not to open source? It depends!!!