There’s a well-known essay by Dan McKinley called “Choose Boring Technology.” We didn’t just read it. We built a company around the principle.
The Problem with Exciting Technology
Every new framework, every new database, every new language comes with a cost that’s easy to underestimate:
- Unknown failure modes. You don’t know how it breaks until it breaks.
- Smaller community. Fewer Stack Overflow answers, fewer battle-tested patterns.
- Migration risk. Today’s exciting technology is tomorrow’s legacy system.
- Hiring complexity. Finding engineers who know the tool and know its sharp edges.
None of this means new technology is bad. It means new technology is expensive in ways that don’t show up in the feature comparison chart.
What “Boring” Actually Means
Boring doesn’t mean old. Boring means:
- Well-understood. We know how it fails, and we know how to fix it.
- Battle-tested. Thousands of companies have already found the bugs.
- Well-documented. When something goes wrong at 2 AM, the answers exist.
- Stable. The API won’t change in a breaking way every six months.
PostgreSQL is boring. It’s also one of the most capable databases on the planet. Linux is boring. It runs most of the internet.
When We Choose “Exciting”
We’re not dogmatic about this. Sometimes the boring option genuinely doesn’t solve the problem:
- When the problem domain is new (AI/ML, for instance)
- When performance requirements rule out conventional tools
- When the boring option has a known, fundamental limitation
But the burden of proof is on the new technology, not the proven one. We don’t adopt new tools because they’re interesting. We adopt them because they’re necessary.
The Real Test
Before we add any technology to a project, we ask:
- Can we solve this with something we already use?
- What’s the operational cost of adding this to the stack?
- Who maintains this if the original team moves on?
- Is this technology serving the product, or serving our curiosity?
If the answer to #4 is “our curiosity,” we put it in a side project, not a client’s production system.
The Payoff
Projects built on boring technology share some characteristics:
- They ship on time more often
- They’re cheaper to maintain
- They’re easier to hand off
- They still work a year later without emergency patches
That last one matters more than people think. Software that quietly works is the highest form of engineering. It’s just not very exciting to tweet about.
At Medhaksha Labs, we choose technology the way we’d choose building materials: for reliability, not for Instagram.