I walk the line.
During a recent conversation I had with two savvy enterprise architects at a major corporation, we found common ground on this as a key goal.
What is the line? It’s the one between choosing open source or commercial software products.
We were discussing the relative strengths and weaknesses of both, and how they fit into an overall enterprise architecture at a large firm that regularly acquires companies to boost growth and market share.
Both architects had worked in different industries at other large firms, helping design, build, manage, and transition their enterprise IT estates, and had lived this problem for a long time.
And what does this have to do with ? Well, large companies engage in mergers and acquisitions (M&A), and, so does government, in a way. Of course, in the public sector, it’s not called M&A. But it might be called data center consolidation, agency interoperability or agency reorganization. The words are different, but the requirements are identical with the commercial world.
(Learn more about )
Their big challenge today was to allow acquired companies to run independently of the acquirer. This meant acquired companies could continue operating and selling without pausing to integrate, but it yielded a sprawl of petty IT fiefdoms. Effectively their firm is a holding company, and this fragmented structure will not give them the scale to compete globally with the tech giants now moving in to their industry.
As senior architects, their job is to design a unified IT model for their independent units, and then start moving these units towards that design. The complexity, rigidity and vendor dependence of the current model led them to consider open source as the way out. They could design, write, and own APIs and then move the independent units to that standard corporate architecture.
The two architects felt open source APIs would offer more control and less lock-in and dependence on vendors, but I pointed out that approach sometimes requires they write a lot more of their own code,; besides, open source communities can be just as rigid as commercial vendors when it comes to modifying their systems for customers. Yes, vendor lock-in can be frustrating when vendors stop innovating, but being locked into an unresponsive or subpar internal development team can be just as frustrating, and even more expensive.
So is "finding the line" really about where you use open source versus commercial software?
No, when you consider that getting the petty fiefdoms on board with the new corporate architecture is going to require top sponsorship and buy-in. That is going to require a business case, including potentially lower costs and innovation that impacts the business's top and bottom lines and keeps the firm competitive against integrated, global tech giants. The line is really about what you buy (could be open source or commercial software) and what you build, and why. Products like Oracle Integration Cloud can add a middle ground of integration between open source and commercial products to match a specific public sector or commercial workflow.
I recently faulted AWS for replacing Oracle in some of their consumer businesses with fragmented cloud services: it didn't lower costs; required an expensive, extended development effort; and yet it very likely subtracted business value with less performance, availability and features.
Agencies need to determine what to build and what to buy, from a palette that should include open source or commercial cloud products that optimize for the business goal while simplifying and streamlining integration across IT fiefdoms. Find that buy-versus-build line based on meeting that business goal while optimizing the balance between your company's software development skills and existing software and cloud services.
Learn about Oracle Government Cloud.