Mainframers are familiar with the myth that their systems are outdated, expensive, and in need of replacement. Other common refrains include COBOL being outdated and hard to learn, or that modernization means moving everything onto the cloud.
Misty Decker, product marketing director at Micro Focus, says everyone is biased because their view of the world and the effect of technological changes are based on their own education and experiences. This is true inside and outside of the mainframe space.
Tackling Unconscious Bias
In order to address tech biases, individuals need to be aware of their bias in the first place. "No one is unbiased, and there is danger in thinking that you're not biased," Decker says. Acknowledging your own biases can make it easier to fix or modify your thinking, challenging your assumptions not only about socioeconomic conditions, but also about the technologies and systems you use.
Although the human brain processes 11 billion bits of data every second, our conscious minds can only handle 40-50 bits of information per second, leaving much of our decision making up to the unconscious mind, according to a recent NPR Short Wave podcast.
Bias begins with the data sets technologists accumulate throughout their education and experience. It is up to the individual professional to expand their own data sets and train themselves to see beyond their current assumptions. "This is how your brain is trained," Decker says. "It's like an artificial intelligence bot. Our brains are wired to take the data we see and draw assumptions based on that, so we don't have to repeatedly figure things out."
Decker looked at it from a technology perspective to explain further. "There was a Twitter algorithm that would automatically crop out Black faces from photographs, because the data set it was primarily trained on used white faces to crop images," she says. "You can't say that the algorithm was filled with hate, but the data set used to train it was clearly biased."
In the world of technology, students and technologists are often learning about new software and programs, not legacy systems, older computer languages or other "old" technologies. In this way, she indicates many technology professionals are biased against the mainframe and COBOL, automatically viewing newer as better.
Steps to Address Unconscious Bias
Once enterprises and individuals can identify their technology biases, individuals and teams have to actively challenge those biases by educating themselves on unfamiliar software and tools. This should be done before making decisions about which solutions will solve issues the best. Part of that work will require technologists to reach out to experts on other software and platforms to familiarize themselves with those tools' strengths and weaknesses.
This scenario assumes that teams want to update software or systems in the first place. However, Decker says there also tends to be a bias against change among software engineers and others. Ultimately, biases can be dangerous to enterprises and their ability to fulfill the needs of their clients and customers, especially when customer data is involved, because it allows for security gaps and errors to occur.
For mainframers, partnering with the "other side of the fence" (non-mainframers) can help enterprises leverage the best of both systems and software, she advises. In addition, teams should review how they are rewarded in the workplace, because this too can create biases. For individuals who are rewarded for being the first to adopt a solution, they will be biased in favor of new applications, software, or systems. "This gives you an unconscious predisposition to choose new tech over the old," Decker says. "Even if nothing needs to change, you're still going to convince yourself that it does because subconsciously you know you'll get rewarded for it."
When technologists pay attention to their own biases, they can begin to actively listen to other experts that employ other solutions and platforms. Understanding the strengths and weaknesses of each technology provides a broader view of which solution will be most effective in meeting a business need. But each technologist has to come to the table with an open mind and understand that all solutions have boundaries to respect. "There is no one right tech solution for every problem," adds Decker.
At this point, each expert will learn about unfamiliar technologies and understand the strengths and weaknesses of those solutions, effectively changing their own brain's data set. They've expanded their knowledge and understanding, making conversations richer and helping others see where their biases may lie. From there, Decker explains that each individual needs to actively redefine normal, which will require the establishment of metrics to determine how far the normal course of business thinking has moved away from previous biases.
"When you're floating in the river on your inner tube looking up at the sky, you have no idea how far down the river you've gone until you look at the shore. After looking at the shore, you see that the dock is far behind you," Decker explains. "The current is always there. It's invisible like unconscious bias. It's the force that pushes things in the wrong direction."
Along the river, you'll need to position markers or metrics to determine how far from the path you've strayed. To keep on track and tackle bias, you have to "paddle" against the current. The goal should always be to remove barriers and give each technology solution the best shot at solving the business problem or meeting a need.
Actively Unravel Tech Bias for Business Success
Decker points out that those in the mainframe space have been previously rewarded for the system's reliability. "This emphasis on reliability is counter to the benefits of new solutions," she adds. To ensure the enterprise achieves its business goals, interdisciplinary teams with mainframers, cloud specialists, and other technologists need to come together with the same focus, reaching the business' goals.
Each member of the team should remove barriers/biases during discussions and assess each solution with a keener eye that keeps the business need in mind. This is where metrics are important in determining the best fit. Decker explains this is likely to be a combination of the mainframe partnered with other technologies. Additionally, she says these teams should be cross-functional and not report to one team (e.g., cloud team) or the other (e.g., mainframe team).
Projects undertaken by these interdisciplinary teams should tackle small changes to business systems. For example, a team could change a monthly reporting capability so it takes 30 minutes instead of 10 hours, Decker exaggerates. But the point is to have open and honest conversations with other technologists about the business value of changing that reporting capability, making it faster by determining the best tool to accomplish that goal.
Once the process is successful in uncovering biases, assessing technology based on metrics, and choosing the right fit, teams should market their strategy to others throughout the business. Without a focus on the business need in these discussions, teams can disintegrate or passively sabotage one another. This can be done by simply not sharing critical information and can lead to a project's failure.
Bias is everywhere. Tackling it starts with the individual, but through conscious effort, technologists and enterprises can work against the tech bias currents to find the right solutions and platforms for their business.