As cloud expands, does legacy grow too?

I was recently reading Drew Bell’s post Legacy systems are everywhere. It struck a deep chord for me.

Join 32,000 others and follow Sean Hull on twitter @hullsean.

Drew first touches on a story of upgrading an application with legacy components, taking pieces offline, and rebuilding to eliminate technical debt.

He then tells a parallel story of renovations in his new home. Well new for him, but an old building, with old building problems.

I’ve gone through some similar experiences so I thought I’d share some of those.

o A publishing company on AWS

I worked with one company in publishing. They had built a complex automation pipeline to deploy code. As a lead engineer planned to exit, I was brought in to provide support during transition. As with large complex websites, there was a lot that was done right, and some things done in old ways. Documenting all the pieces and digging up the dead bodies was a big part of the job.

Also: Is a dangerous anti-ops movement gaining momentum?

o Renovating a kitchen

In parallel to the above project, I was renovating my kitchen, in a new home in Brooklyn. Taking on this project myself, I dutifully assembled IKEA cabinents, and laid them out to spec. As I began the painstaking process of leveling for the countertop, I ran into trouble. Measurement after measurement didn’t add up. It seemed one section was shorter than another, where the counter should go.

Since I needed to add support for a dishwasher, that had to be measured correctly. Yet the level tool told a different story than the yardstick. Finally after thinking about it for a few hours, I put the level on the floor itself. Turns out the floor wasn’t level! That explained why cabinets were shorter in one area than another.

Also: How do we lock down systems from disgruntled engineers?

o Legacy in 5–7 years?

Complex systems like software, exhibit a lot of the same surprises as old buildings. That was one surprise I wasn’t expecting. As houses are renovated on the 15–30 year timeframe, software seems to experience a five to seven year cycle.

Whether a consequence of shifting sands in the underlying stack, databases, frameworks or cloud components, or the changing needs of product & customers

Also: Is AWS a patient that needs constant medication?

o Opportunity everywhere

As companies large & small migrate pieces of their systems to the cloud, move to microservices or rebuild on serverless, the opportunities are endless. It seems every firm is renovating their kitchen these days, putting on a new roof or upgrading their data pipeline.

Also: Is AWS too big to fail?

Get more. Grab our exclusive monthly Scalable Startups. We share tips and special content. Our latest Why I don’t work with recruiters


Originally published at Scalable Startups.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.