By offering on-the-job training programs that incorporate both legacy systems programmers and emerging tech developers, agencies can fill in the skills gaps and encourage the cross-pollination of expertise, which will ensure a much smoother modernization of IT systems.
While legacy systems such as mainframes and their monolithic applications have supported the complex, mission-critical workloads of government for decades, the continued reliance on their rigid and cumbersome technology has increased risks – whether around system failures, subpar customer engagement, operational challenges or the general inability to remain agile in today’s digital world.
One of the biggest challenges for mainframe applications comes with the shrinking talent pool of those well-versed in supporting them. When COVID-19 hit and many state unemployment platforms crashed, the legacy programming language COBOL was often cast as the sole culprit. In reality, there are hundreds of obscure languages hiding in mainframes around the globe, and COBOL is actually one of the more universally supported and understood procedural languages in the mainframe arsenal. With their expensive licensing fees, languages such as Natural, CA Gen, CA Telon, PL/I, ADS/Online and Assembler, pose even greater threats to organizations with miniscule talent pools.
Legacy applications have, over time, become "black boxes" of information for future IT managers. Millions or even billions of lines of entangled code has been written by various developers who, in many cases, have moved on from their posts or retired. Passing through different hands over many decades, often without proper documentation, the code is inscrutable to new employees lacking knowledge of its purpose, functionality and history who are tasked with deciphering it. And since most employees that do have legacy skills are now on the verge of retirement (18.8% of COBOL developers are 55+ and 60.5% are 45+), they’re poised to take their institutional knowledge with them and leave quite a mess behind if unaddressed before then. In fact, a 2018 Forrester study called “Modern Mainframe KPIs Are Key to a Successful Digital Strategy” found that enterprises lost around 23% of specialist mainframe staff in the five years prior, and 63% of open positions were unfilled at the time.
While talent experienced with DevOps and emerging technologies are much easier to employ today and are key to modernization, those skills aren’t always enough. Without historical knowledge of how application functionality has evolved, or accurate and current documentation, kicking off any sort of application modernization project without the help of experts and their tools will be inherently slower, filled with headaches and rife with opportunity for failures.
It’s therefore critical that organizations successfully bridge the gap between old systems and talent and the new now. Part of addressing this challenge requires stronger on-the-job training programs that incorporate both legacy-systems programmers and emerging tech developers. By implementing this strategy, organizations fill in the skills gaps and encourage the cross-pollination of expertise, which will, in turn, ensure a much smoother modernization of IT systems.
For example, by training COBOL programmers to be Java developers, agencies can leverage their legacy knowledge of applications’ functionality to make better and faster changes. This also helps improve staff attitudes toward modernization by showing there’s a place for their skills moving forward. Ultimately, in-house training shows how different skills can be leveraged to meet constant changes in app development, helps retain a company’s most valuable resources and promotes an “always-be-learning” culture.
Training tech talent to close the skills gap starts with four essential elements:
- Legacy and DevOps expertise. Training is most successful when there are experts on both sides of the skills gap. This includes those versed in legacy technologies – the COBOL programmers – who understand the entire lifecycle of legacy systems as well as workers trained in new skills, who live and breathe DevOps. Both have expertise to share that will help agencies more quickly bridge the gap between legacy systems and a true DevOps practice.
- Repeatability. Agencies should ensure training is continuous and that they’re capturing each employee’s expertise by establishing a central repository for information sharing. This includes recording training sessions to capture screen shares or any white-boarding or creating a company Wiki to make it easy to find documentation. Confluence and SharePoint are two common platforms organizations use to store knowledge and keep it accessible.
- Scalability. Training material should be organized into several tracks based on employees’ roles and goals. Continually refining the tracks and training and formalizing a training system that scales with agency needs are key. This process requires maintaining open communication across functional groups.
- Mentorships and cross training. A formal mentorship program ensures everyone is on the same page with overarching strategies and DevOps culture. It also promotes adherence to best practices and broadens everyone’s skillsets. Best of all, it makes onboarding faster and easier while ensuring a knowledge base that will help manage contingencies and unforeseen demand.
Overall, the skills gap between those versed in legacy technologies and those trained in a cloud and DevOps world can complicate an agency’s application modernization plans. By cross-pollinating skillsets and building a robust in-house training programs, agency leaders can avoid the pitfalls resulting from scarce legacy system talent and leverage existing investments as they further modernize IT.