HUNT.ZA — putting Universal Acceptance in front of South Africa's next engineers
South Africa has its own domains — .africa, .joburg, .capetown, .durban — and many of the country's biggest platforms quietly reject email addresses that use them. The fix is easy. The hard part is that almost no one knows the problem exists. So we ran a hackathon at the University of Pretoria to change that.
The problem
Universal Acceptance is the principle that every valid domain and email address should work everywhere — including newer and non-Latin extensions. It stays broken for a mundane reason: a validation pattern somewhere caps the top-level domain at three or four characters, and .africa never makes it through.
The bug is invisible until someone with a .africa address tries to sign up and is told they don't exist. The hard part isn't the fix — it's that almost nobody knows the problem is there. When we started the day, most of the students assumed it had already been solved.
Context
South Africa has its own ICANN-delegated domains — .africa, .joburg, .capetown and .durban — yet a surprising number of local banks, government portals and insurers reject email addresses that use them. These are valid addresses. The validation code is simply out of date. It is a textbook Universal Acceptance failure, hiding in plain sight across the country's digital infrastructure.
What Tyto built
We hosted HUNT.ZA at the University of Pretoria alongside ICANN, the UA Steering Group, Registry Africa and ZARC. Twenty-eight students, eight teams, one day, two halves.
The landscape hunt. Teams went looking for real UA failures across live South African platforms and scored them by severity — 50 points for a site rejecting a .*.za address, 30 for .africa, .capetown, .joburg or .durban, and 15 for other UA issues. Multipliers stacked on top: 2× if the failure locked a user out entirely, 1.5× for genuinely unusual breakage, and 1.3× for government, banking or insurance platforms. The result was a sharp, evidence-based picture of where the country actually stands.
The validator build. In the afternoon, every team built an email validator from scratch — no libraries, no AI, just the problem and a blank file. We tested each one against a suite of valid and invalid UA addresses to see what held up.
What this proves
One team took best validator. Another won the report category and, having also placed second in the build, walked away overall winners.
But the scoreboard wasn't the point. Twenty-eight people walked in assuming UA was a solved problem and walked out able to spot the bug, explain why it matters, and write code that doesn't reproduce it. That's twenty-eight future engineers who will never quietly lock a .africa user out of a signup form. For a problem whose root cause is simply that too few people know it exists, that's the work that matters.
- 28 students — across 8 teams, from 2nd year to Masters
- 2 disciplines in a day — landscape research and from-scratch validator engineering
- 0 → fluent — most students had never heard of UA that morning; all left able to explain and detect it
- 4 partners — ICANN, UA Steering Group, Registry Africa and ZARC, hosted at the University of Pretoria
Your business has a story like this waiting
It usually starts the same way — better analytics requested, a data foundation problem found.