A confirmed bug becomes a standing check
How a confirmed finding becomes a regression check, and why a check only counts as protection once it has been seen to fail.
Blog
Writing from nth Labs on testing software that ships faster than people can check it, including what we have measured about our own tool.
Product
Testing behind a login, without your password
A run can reuse the session your own browser already has, so SSO and 2FA still work and nothing on our side sees a password or a login form.
How the persona panel picks its testers
How the panel draws six simulated testers from ten thousand records to match the audience you describe, and what it does not model.
Research
Research note
Software production is booming. Who is checking it?
Five published studies, read together: code output rose by a fifth to a quarter in a year, while every measurable sign that somebody maintained the code fell.
Articles
36 out of 36 is not 100%
Why the textbook confidence interval gives impossible answers at small sample sizes, with real coverage as low as 3.9% at 16 trials, and the one-line fix.
Which accessibility issues can be automated?
An automated checker ships around a hundred rules. Counted against the spec they land on 29 success criteria, and 29 of those rules cover one criterion between them.
You are not your user, in five numbers
In a research dataset of ten thousand simulated users, 27% write code, 12% are digital natives and one in five arrives trusting your software.
How to tell a website was built by AI
Seven signs of the house style many AI-built sites now share, in the order a visitor meets them, each one visible without developer tools.