DOS-era tech still runs factories. It's also your next SaaS opportunity.

I stumbled on this Hacker News post from mlaux asking who's still keeping a DOS machine up because the business depends on it. It's a simple question, but it opened a rabbit hole I didn't expect. The thread is full of stories about CNC mills controlled by ISA cards, spectrometers that require a parallel port dongle, and dBase applications running on 30-year-old hardware.

It's easy to read that and think, "Wow, that's a dying niche." But our data at PainSignal tells a very different story. We track thousands of problems across industries, and legacy system pain is not only alive—it's growing. And where there's pain, there's opportunity for builders.

The problem isn't just DOS. It's everything stuck in the past.

mlaux's post focuses on DOS-era RAD environments like Clipper and Clarion, plus industrial instruments tied to obsolete hardware. But when we look at the problem data in manufacturing, the issue is much broader. We're tracking 181 distinct problems in manufacturing alone, with an average severity around 4 out of 5. That's not a niche. That's a screaming market.

For example, one recurring problem is equipment downtime caused by the inability to integrate modern monitoring tools with legacy machines. Operators are forced to manually record readings, or worse, leave machines unmonitored because the software can't talk to the hardware. That's a direct consequence of using 30-year-old tech.

But here's the thing: the pain isn't limited to manufacturing. Healthcare is a parallel universe. We're tracking 1,040 problems in healthcare, and many of them stem from the same root cause—outdated documentation systems and equipment that can't communicate. One case that stands out involved a nurse placed on unpaid leave because manual glucose logs didn't match the glucometer's internal audit trail. That's a severity 5 problem, and it exists because the hardware and software were never designed to talk to each other.

The common thread? Data capture and integration. Whether it's a CNC mill or a glucometer, these systems generate critical data that's trapped in silos. The businesses that run them are desperate for a bridge to the modern world.

So what can you build?

The obvious answer is emulation or compatibility layers. Software that lets modern machines run old DOS apps or interface with ISA/GPIB cards. That's a real product, and some people are already doing it. But here's where it gets interesting: our data shows there's a much bigger opportunity in the software layer above the hardware.

We're seeing a trend in manufacturing where the top opportunities are not just about keeping old machines running, but about capturing data from them and feeding it into modern analytics dashboards. One of the highest-scoring app ideas in our database is a universal data connector for legacy industrial equipment. It's not a hardware emulator; it's a reporting and monitoring layer that sits on top of whatever ancient tech is underneath.

Think about it. The factory owner doesn't care if the DOS machine runs in a VM or on original hardware. They care about knowing whether the production line is up, whether quality is within spec, and whether maintenance is due. They want a dashboard. They want alerts. They want compliance reports.

And the same pattern appears in healthcare. The app ideas with the most traction aren't about replacing old glucometers—they're about creating a unified audit trail that reconciles device data with manual logs. It's not a hardware problem; it's a workflow and compliance problem.

The market is bigger than you think.

mlaux's post assumes that businesses using DOS and legacy hardware are a niche, possibly a dying market. Our data says otherwise. First, there's the sheer volume of problems: 181 in manufacturing, 1,040 in healthcare. Second, these aren't low-severity annoyances. The average severity is around 4/5, which means people are losing money, time, or even patient safety. Third, the problem is not going away on its own. The hardware might be old, but the demand for it is still high because the alternatives are either too expensive or too risky.

And here's a kicker: the trend lines for these opportunities are rising. In both manufacturing and healthcare, the top app ideas have scores above 60 out of 100 and are trending upward. That's a clear signal that the market is heating up.

If you're a vibe coder looking for your next project, don't ignore the dusty industrial floor. The next SaaS unicorn might be a dashboard that talks to a floppy disk drive.](https://painsignal.net/explore?industry=manufacturing)

How to get started

You don't need to become a retro-computing expert. Start by picking one narrow pain point. From our data, two stand out:

Equipment downtime monitoring for legacy machinery – A lightweight sensor/software combo that can watch an old machine's status through its existing serial or parallel port and push alerts to a phone.
Automated compliance reporting for manual logs – A tool that ingests photos of handwritten logs or CSV exports from old software and generates audit-ready reports.

Both ideas address severe, recurring problems in industries that are willing to pay. And both can be built by a solo developer in a few months.

The HN thread is full of anecdotes that prove the problem is real. Our data proves it's profitable. The only thing missing is you building the solution.

Check out the original Hacker News thread to see the stories yourself.