Just because an organization has decided it's time to modernize its old applications doesn't mean it will be smooth sailing from there on out. It has to consider a variety of factors, such as cost and time, before diving into an app modernization project. Companies keep maintaining applications on the Mainframe because it will be too expensive to re-write them. Moreover, lots of companies have based their core system on COBOL applications that rewriting would be a high business risk. It can be expensive and time-consuming if a project roadmap isn't carefully devised and followed.
Several generations of developers have coded those applications with the applications knowledge often faded along with the developers. Consequences are: code duplication, fear to modify working apps, and regressions. Most of the code base assessed for modernization would have posed one or the other challenge to understand the business flow or conversion to a different technology.
Recently there was a request to modernize a client’s 30+-year-old applications Model 204. In spite of being from the Legacy background, many of us hadn’t heard of it. FORTRAN was one of our main subjects during our engineering academics in 1992, but now we don’t even have people to claim they know how to modernize. Subsequently, a large pension Insurance company in Switzerland, with a 30-year-old COBOL application and heavily batch-oriented as well as Online CICS, made many attempts to replace the system in ORACLE Forms, Java, JEE, C#. Hence, it is critical that you find some of the best practices to rationalize legacy applications and identify cost-effective methods to continue to support legacy applications until they are retired, which itself can be a first giant step. Compared to other languages, COBOL is much advanced and very much present in the market. Other previous contenders such as PL/1, C, Pascal, VB, and C++ are all in a much steeper decline than COBOL itself. A mismatch between the supply and demand of legacy programming skills could down the line have serious repercussions for businesses.
In 10-15 years we might be talking about ABAP/SAP in the same way we're talking about COBOL now. The interesting part is that there are still companies investing millions of dollars for many consultants to build systems that will bury them the same way COBOL did 20 years ago.
Year 9999 AD
I feel that the above image of one of my cartoons that was published in 1999 is very much representative of this situation. COBOL was introduced in the 1950s and quickly became the premier language for writing the much-needed business software, but the switch from the Mainframe computers to desktop PCs saw challengers like Java and .NET overtake it as the preferred language. Despite its dwindling popularity, huge amounts of the modern world’s infrastructure is still based on the language. This is a cause of concern for businesses, as many academics running university IT courses around the globe do not have COBOL programming as part of their curriculum.
Illustrations - Vishwesh Bhat
Graphics - William Ebenezer