.png)
The bar for creating useful tools that help out with day to day tasks has gotten lower and I keep wondering how low it will go before I’m creating my own custom e-mail client or spreadsheet application.
Building tools instead of downloading utilities
What got me thinking about this was that I’ve gotten into the habit of creating my own simple tools to do things that used to mean downloading something off the internet. For example I wanted to convert a bunch of m4a audio files to mp3 files so I could play them on this tiny old mp3 player I use when working out. Yeah, sometimes I use a good old fashioned mp3 player that isn’t connected to the internet. No alerts, texts, calls, etc… interrupting my playback.
Searching around for free software led to an hour of downloading a variety of different software programs riddled with ads, asking for personal information, or only doing what I wanted to do if the full version was purchased. It ended up only taking me 30 minutes of iteration with my preferred LLM to create a python script that did the conversion. When I needed to add a bunch of files to a Gemini project and hit the file limit I whipped up a couple of scripts that concatenated all the docx files, pdf files, and pptx files so instead of hitting the file limit I only ever needed three.
Open source projects can be bent to fit a specific business need
That experience and reading about people vibe coding expensive enterprise quality tools and releasing open source versions got me thinking. Now technical and tech savvy folks can rapidly create custom tools and software solutions. And open source projects can be bent to fit a specific business need without the army of experts it used to take.
This opens up all kinds of possibilities for small, midsize, non-profit, or really any business looking to save on costs. Granted open source projects have always offered a great alternative to many expensive solutions. But like custom enterprise software, it took experts to customize them, and in open source’s case, to keep them running. The bar for building your own customizations or from scratch tools is getting lower quickly.
Maintenance and security still matter
While the bar may be lower for building custom solutions, the things everyone always seems to forget about have not. Maintenance, security, monitoring, and updating still remain to be considered when looking to build tools that are more than just a stand alone single purpose solution sitting on my local computer. So while the cost of building has gotten cheaper the cost of owning still remains.
When I worked in digital mapping we used to say that mapping the world once was relatively easy compared to keeping everything updated. I learned this lesson when I used to create apps for the original iPhone (yeah the 2G version). In the first couple of years I created 12 or so game apps as a side project. They sold ok and I made some decent pocket money but over time the hardware evolved and the effort to keep them updated and working became too much for the small amount of revenue they produced.
I’ve been at more than one company that had a revenue dependency on outdated software that no one wanted to touch or look sideways at for fear of it failing. Sadly, I have been part of what I call kicking the can down the road. All to save money in the short run instead of biting the bullet and rebuilding something that does the same thing but on modern systems. Until of course it gets abandoned due to budget cuts, ends up not being maintained again, and then needs to be moved again. Rinse and repeat.
Considerations when it comes to creating and maintaining your own software
So there is more to consider than just fast and cheap when it comes to creating and maintaining your own software.
- Build it if it’s for you, but think twice if it’s for everyone.
- All software needs an owner because all software has an expiration date.
- Free is never free, so pick the bill you’re prepared to pay. That may be a monthly fee, a part-time maintainer, consulting services, or your own time. Software that isn’t being maintained becomes a bigger problem the longer it goes unmaintained.
- Think hard before rebuilding commodity software. If you can swap it out when something better or cheaper comes along, let someone else carry it. I’d love a custom e-mail client that works exactly the way I want, but maintaining it sounds like a full-time job I’d rather pass on.
I still don’t know how low the bar will go. But I do know what a good tool looks like. It’s sitting in my gym bag with a dead battery, ready to play the same forty songs it’s played for years. No ads, no updates, no subscription, nothing to maintain. Just doing the one job.
