Legacy IT

The scary truth about modernization

Placeholder Image for Article Template

Jon Dittmer is vice president and general manager of the defense sector at Array Information Technology.

Fifteen years ago, I was a defense program manager challenged with modernizing a legacy Cobol supply chain system. Today, I sit on the industry side of the fence awaiting the release of a solicitation for modernizing that same system.

As part of that solicitation, the Defense Department is trying to convert 40-year-old Cobol code to Java. Imagine what that Cobol code looks like “under the hood.” It has been modified by hundreds of developers over the years without a major architectural upgrade. How agile do you think software maintenance is in that environment? Add an aging workforce to the equation (including one employee who is over 80 years old) and eventually a major crisis will occur. It’s inevitable — and probably all too familiar.

In the past 15 years, there seems to have been little to no progress on the majority of the government’s many legacy modernization efforts. IT systems are just as critical to the mission today, if not more so. So why can’t we seem to get legacy modernization right?

Organizations that fail to modernize will become unresponsive to customer and constituent needs, and they will ultimately not be competitive in the marketplace. Perhaps most important, the gap between where they are and where they need to be will only widen, which causes a much more painful, expensive and scary future.

Unfortunately, we are facing incredibly tight and shrinking budgets, more so today than in a long time. But if agencies do not take the appropriate steps toward modernization, they will almost certainly incur increased costs related to performance and maintenance. With a modernized platform, organizations can add capabilities and enhance overall performance while reducing their electronic footprint.

Inconsistent and overly complicated IT systems are the proverbial skeletons in the closet of legacy modernization.

However, IT organizations often fails to grasp those fundamental risks and rewards in terms of real business or mission impact — or in terms of the costs associated with the cultural or operational changes that might be required.

Furthermore, we often find ourselves in the middle of positive change and modernization, only to find that someone loses patience, new executives change direction or 10 reasons surface for why we should stop modernizing — for example, the actual or perceived cost or a newfound faith in the old system. Those reasons create inconsistent and overly complicated IT systems that are outdated and ineffective. They are the proverbial skeletons in the closet of legacy modernization. Organizations can’t get rid of them, and over time, the IT organization loses credibility and the ability to effect positive change.

We need to rid ourselves of those skeletons before they cripple our organizations. When a system no longer supports your business processes or when there is an off-the-shelf solution that meets your needs, consider a complete overhaul or replacement. However, modernization will usually be the right approach when your business rules still align with the system and your biggest issues are tied to legacy cost and lack of agility. Some systems will benefit from a minor platform tweak to enhance performance and lower the cost of operations, while others will need to be fully redesigned.

All modernization is not created equal, but delaying the effort will only put more skeletons in the closet — and increase the size, scope and scariness of the modernization that’s required down the road.

Who's Fed 100-worthy?

Nominations are now open for the 2015 Federal 100 awards. Get the details and submit your picks!

Featured

Reader comments

Mon, Dec 16, 2013 Kenneth Shafer Indianapolis, IN

Mr. Dittmer's piece is long on platitudes and short on substance. Also omitting that the reason many Legacy Systems are still in place, still running, and still performing their mission, is that previous attempts at "modernization" failed. Not in small part because the languages-du-jour are not simply up to the heavy lifting that CoBOL performs with reliability. And where is the justification that systems modernized by "replacement" are any simpler to maintain, and any less expensive to maintain? Maybe Legacy Systems are complicated because they represent the accretion of literally *decades* of business rules changes and enhancements. Would a "brittle" system have that kind of track record? Most, if not nearly all, of "modernization's" needs can be accomplished by retaining the "core" of a Legacy System, turning it into "services" by means of a Service Oriented Architecture, and then wrapping-around a Java / web-enabled front end.

Tue, Dec 10, 2013 RayW

From the bottom feeder point of view, while I see plusses and minus to massive upgrade programs, there is one issue that Mr. Dittmer only hints at. We spend 15 years getting a system to the point where it is now performing flawlessly, only to be told that it is obsolete and we will get a new system because the old is banned and will be turned off in X days. Now potentially another 15 years of pain to get to the point we are today only to repeat this then?

But since we are in the Federal system and tied to Microsoft who likes to break existing software with each facelift release, Mr. Dittmer has many valid points that are best addressed early (if we can find funds) instead of later when it impacts both funds, productivity, and customer relations. Unfortunately, accountability to congress is one big issue in preparing ahead of time. Why spend money trying to replace what works when you will not have to do that for an election cycle or two? And today that is a BIG issue. Plus, if you start early, is the system you are planning for going to be valid when you need it or will it be on the bad list like your current program? Another cost to wory about.

Mon, Dec 9, 2013 Suparna Rao

Completely agree ! While technology executives try to push out modernization plans as much as they can, what happens when outdated systems start costing its business their customers ? This is the premise of our latest blog, in case you were interested in reading. http://mlabs.boston-technology.com/blog/is-your-legacy-application-holding-you-back

Please post your comments here. Comments are moderated, so they may not appear immediately after submitting. We will not post comments that we consider abusive or off-topic.

Please type the letters/numbers you see above