Hacker News Reader

Comment by adrianN | Hacker News Reader

Until jobs doing maintenance pay as well or better and have better career prospects than jobs doing greenfield development, the rational choice is to avoid jobs where you don't get to use the latest fad technology. Being the greybeard that keeps the legacy stuff running is an honorable position, but these people usually have a hard time keeping their jobs when the legacy tech is finally replaced.

View on HN

All your banking transactions run over old mainframes using COBOL. Probably on systems maintained for 20 years. They will not even be replaced as most already have tried this and not succeeded.

I know of companies training COBOL developers to make sure they can maintain these legacy systems. Mostly they are also the mission critical systems.

The web applications at these financial institutions will be in more modern crappy frameworks. But rest assured the expert COBOL developer gets paid much more for doing a lot less :)

Sure, but what prospects will those newly minted COBOL developers have when the next recession hits and their department decides most of them aren't that critical after all or when they realize they're in a dead-end job and want to move on? The market for mainframe programmers is small and getting smaller every day. There are no new customers in that market, only the existing ones and the list keeps getting shorter every year. The situation is even worse for anyone stuck on legacy tech from the PC era on.

The department cannot get rid of mission critical maintenance. If you do that, your company dies right there and then once you're even minimally unlucky. The newcomers have a delay where they need to be trained, which is typically measured at at least a year.

The new guys get removed first. Ones working on shiny less critical processes. Shiny and critical generally happens in startups.

When legacy is being replaced, old timers generally see the writing on the wall and move on as well as retrain.

Agreed. I liken it to the natural world where a species can be extremely specialized to exploit a particular ecological niche, but that niche is getting destroyed. Times are good when you're the only animal that can get to a particular fruit but when that fruit goes extinct, you're screwed.

And the people maintaining that COBOL usually make the same money they could make managing a McDonald’s.

My understanding is that COBOL is paying a lot right now.

Some niches are of course different from the mainstream, but spending five years maintaining a VB6 application is unlikely to further your career a lot.

More than being an inexperienced programmer with shiny keywords.

An experienced programmer with all the shiny keywords is rare.

Maybe as if you're one of the very best COBOL programmers around and work as a consultant. However, anecdotally, working a normal 9-5 COBOL programming job at a bank or insurance company doesn't pay well at all.

Reply on news.ycombinator.com