___ __ ___ __ ___ __
|\ \ |\ \|\ \ |\ \|\ \ |\ \
\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \
\ \ \ __\ \ \ \ \ __\ \ \ \ \ __\ \ \
\ \ \|\__\_\ \ \ \|\__\_\ \ \ \|\__\_\ \
\ \____________\ \____________\ \____________\
\|____________|\|____________|\|____________|
___ __ ________ ________ ___ ________ ________ ________
|\ \ |\ \|\ __ \|\ __ \|\ \ |\ ___ \ |\ __ \|\ _____\
\ \ \ \ \ \ \ \|\ \ \ \|\ \ \ \ \ \ \_|\ \ \ \ \|\ \ \ \__/
\ \ \ __\ \ \ \ \\\ \ \ _ _\ \ \ \ \ \ \\ \ \ \ \\\ \ \ __\
\ \ \|\__\_\ \ \ \\\ \ \ \\ \\ \ \____\ \ \_\\ \ \ \ \\\ \ \ \_|
\ \____________\ \_______\ \__\\ _\\ \_______\ \_______\ \ \_______\ \__\
\|____________|\|_______|\|__|\|__|\|_______|\|_______| \|_______|\|__|
___ __ ___ ___ ___ ___ ________ _____ ______ ___ __ ________ ________ ________ ________ _________ ________ ________ ___ __
|\ \ |\ \|\ \|\ \ |\ \ |\ \|\ __ \|\ _ \ _ \ |\ \ |\ \|\ __ \|\ __ \|\ ___ \|\ ____\|\___ ___\\ __ \|\ ____\|\ \|\ \
\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \|\ \ \ \\\__\ \ \ \ \ \ \ \ \ \ \|\ \ \ \|\ \ \ \_|\ \ \ \___|\|___ \ \_\ \ \|\ \ \ \___|\ \ \/ /|_
\ \ \ __\ \ \ \ \ \ \ \ \ \ \ \ \ \ __ \ \ \\|__| \ \ \ \ \ __\ \ \ \ \\\ \ \ \\\ \ \ \ \\ \ \_____ \ \ \ \ \ \ \\\ \ \ \ \ \ ___ \
\ \ \|\__\_\ \ \ \ \ \____\ \ \____\ \ \ \ \ \ \ \ \ \ \ \ \ \ \|\__\_\ \ \ \\\ \ \ \\\ \ \ \_\\ \|____|\ \ \ \ \ \ \ \\\ \ \ \____\ \ \\ \ \
\ \____________\ \__\ \_______\ \_______\ \__\ \__\ \__\ \__\ \ \__\ \ \____________\ \_______\ \_______\ \_______\____\_\ \ \ \__\ \ \_______\ \_______\ \__\\ \__\
\|____________|\|__|\|_______|\|_______|\|__|\|__|\|__|\|__| \|__| \|____________|\|_______|\|_______|\|_______|\_________\ \|__| \|_______|\|_______|\|__| \|__|
\|_________|
Two things people are unusually bad at valuing:
being kind, and being educated.
Both are cheap. Both compound. Both have more evidence behind them than almost anything else you could choose to do with a life, and both are routinely treated as decoration on top of the serious business of getting ahead.
So the first half of this page is the case for them, made with numbers rather than sentiment.
The failures are at the bottom, and they are last for a reason. They are the easy part to write, the satisfying part to read, and the least useful. Every one of them was preventable with something already published before it happened, which only means anything once you know what was on the shelf.
Not as sentiment. Kindness is the most examined behaviour in social science and it keeps producing the same result: it is good for the person receiving it, measurably good for the person doing it, and structurally stable in a way selfishness is not.
Education is the intervention with the widest documented downstream effects of anything humans do deliberately. Almost every measure a society cares about moves with it, and several of the effects are large enough to be visible in mortality data.
Most of what people believe about studying is wrong, and the corrections have been sitting in the literature for decades. The methods that feel effective are usually the ones that are not, because fluency feels like knowledge and is not.
Learning styles are a myth. Neurodivergence is not, and conflating the two does real harm in both directions. The difference is that one is a preference people report and the other is a difference in how attention, working memory and language processing operate, with accommodations that have evidence behind them.
The section before this one is about how people learn differently. This is what you owe them once you build something. Almost all of it is cheap, all of it is specified in public, and the interesting part is how often a thing that passes the standard is still unusable.
Before the catalogue of people who took shortcuts, the people who did not. None of these had the resources the situation seemed to require, and the interesting part in every case is the specific method rather than the determination.
This site's argument is that the standard usually existed before the failure. That holds at every scale, including the largest one. These are the things the species has actually signed — not aspirations, but instruments with ratification counts, and in several cases measurable results.
The skill that makes all the others work. Everything in the catalogue at the bottom of this page had a moment where somebody could have updated and did not, and the reasons people fail to update are well studied and depressingly consistent.
What follows is the other half of an education: the shortcuts. Each of the cases at the bottom of this page is somebody skipping a step that was already written down, and they sort almost entirely into three words.
don't be stupid, lazy, or greedy
None of the failures catalogued below were caused by a person being bad at their job. Each was a system that made the wrong outcome easy and the right one effortful. These are the standards that already existed when it happened.
Security is not a feature that gets added late. It is a set of assumptions about who is already inside, made early and written down.
Every system here sat between an organisation and a person who could not see inside it. That asymmetry is the whole ethical content of the work.
Everything above was known before it was needed. That is the uncomfortable shape of this whole site.
55 cases in four groups. Each names what happened, the source, and the published standard that would have caught it. The standard always predates the failure. That is the only editorial position this page holds.
12 cases · Equifax, Capital One, Marriott / Starwood, SolarWinds, Log4Shell, and more
10 cases · Wells Fargo, Volkswagen, Boeing 737 MAX, Uber, Wirecard, and more
22 cases · Replit agent, Zillow Offers, Mata v. Avianca, Air Canada, DoNotPay, and more