The attached article reports that the Covid-19 pandemic is exposing aging, inflexible computer systems at the heart of the U.S. economy, as well as a shortage of experts to fix the problems. More specifically, government agencies computer systems are struggling to keep up with online services due to legacy technology and applications running COBOL.
I programmed a lot of COBOL back in the day. I have now gathered about 8 articles where states are having a significant problem with Unemployment and other systems.. This is not a forum to sell, but we have an AI system that can analyze the code and reduce the need to hire experts to support it. And when people retire, you don't have to worry. If you are interested, you can contact me. About 50% of the code out there is Cobol and I don't know many people who can and have the desire to do that kind of work. (noted by the fact that the last word in the title is spelled wrong as well as a NJ official saying they have a COBALT problem. :)
We have a legacy cobol ERP that runs the rail operations. However, we ported it off the IBM mainframe 8-9 years ago to micro-focus hosting and DB2 for linux. Still works pretty, and is technically ahead of most US Railroads. But we have had a legacy modernization underway for a couple of years now - to provide a more robust application suite. Although, probably not more secure, because we currently are not a web-application. Even our pen-test vendor says hacking into Cobol is pretty rare. The current crisis has had no impact yet on this modernization program or its prioritization level. Our projects being impacted are those that require some level of onsite vendor support for project phases.
We don’t have a legacy system but we do have systems made with modern framework that restricts our growth because of their inflexibility.
There’s great value in architecting code well for longevity and flexibility. The current crisis doesn’t appear to impact vendors in improving their systems significantly. Their goal appears to only change it marginally to get through the current issue. They may be right about cost-value proposition. I don’t know.
High level- we have legacy systems that have a lot of tree rings. We have RPG and COBOL development and maintenance roles. Those things are not an issue. What ends up being a big drag on progress are all the ancillary systems that talk to the mainframes. Some things that are homegrown. Some things based on inflexible configurations. Small-ish user bases so there is reliance on vendor support of tribal knowledge.