How Will Automation Manage Legacy Codebase?

How Will Automation Manage Legacy Codebase?

The subject of automation is one that scares the average employee, over time, millions of jobs have been automated so much so that the average person has to really consider if the contribution of the average person in the workforce will not be reduced to mere labour. Sometime ago, I saw a question on Quora (I can't find the link) that asked how software engineers will survive automation seeing as they're (software engineers) some of the people championing automation.

image.png

A RELATABLE EXPLANATION

Yes, It's possible to automate the job of software engineers (to an extent). Sometime ago I wrote about Github copilot and how its controversial release had developers and aspiring developers scared about the viability of the software engineering field. It's worth noting that there are a ton of skills out there that are somewhat similar in scope to software engineering, not exactly hard to learn but people just don't feel pushed to learn them. There's also the issue of trust and humans like having people to hold responsible.

What I want to dwell on more today is how automation will affect legacy codebase. Anyone who has the privilege of going through the codebase of somewhat established organizations will realize that most organizations don't document code. Undocumented code is hard to manage, even harder than poorly written (yet documented) code. While I see automated software helping in the development of new software solutions (in the not too distant future).

image.png

FINALLY

The average netizen doesn't grasp how code becomes old to the point of being stale. For the average netizen, when a new app comes out you can easily just update and it works as usual, with codebases, upgrading to the latest instalment of the software framework or the programming language can't be done at the click of a button. You need a team of developers to manually upgrade a codebase and even the process is fraught with a lot of uncertainties. The management of a large organization is unlikely to wager on a machine, in the same way, the financial sector still uses traders rather than some software.