Article Response 33 (Technical Inertia)

Article

In this post, the author discusses talks about the difficulties of using the Class E IPv4 address space in the modern internet. He does a very good job of outlining what would make this so difficult but the article as a whole does tend to get into the networking weeds a lot. Not a bad thing per se, but you will need to brush up on your networking knowledge a bit while reading through this.

Show any senior dev that quote at the top and they'll nod vigorously in agreement. Technical inertia is real and it is a royal PITA. Personally, I've had the (mis)fortune of starting a few codebases and then handing them off for maintenance/extension or receiving a codebase and having to extend it. One thing I've quickly is that the early technical decisions have far reaching consequences and even the minor things may end up becoming not so minor. 

Once I had this realization, the obvious next step was to decide that I would always start with good defaults. But how do you decide what a good default is? Like most important questions in life, the answer is "It Depends". The more novel your app, the less likely you'll know what good defaults are because no one would have walked the path and vice versa. This is why more senior devs do system design interviews, the idea being to make sure that they know what good defaults are for the most common app categories and/or paradigms. However, regardless of how novel your app is things like using a DB, caching etc, etc will always come into play at scale, so at the very least you should know what they best practices for those are so you can focus on the more esoteric portions of the application.

The idea I'd like you to leave with from this post is that technical inertia will always occur, it's on you to make sure that it works for you, not against you.

Thanks for reading and as always, all comments, critiques and questions are highly appreciated. Here's a link to the previous article response.

Brought to you by mulVid.