The Method Missing from the Internet

A PhD is mostly thought of as a credential. Underneath it is one of the best methods people have found for working with knowledge — and almost none of it has made it online.

Most people think of a PhD as a certificate: three letters after a name, a line on a résumé, a gate a university keeps. That is the part you can frame on a wall. It is the least interesting part.

Underneath the certificate is a method — one of the oldest and most reliable ways people have found to come to know something well. It goes roughly like this:

1. Ask a question narrow enough to actually answer.
2. Read what is already known, carefully, and write down what each piece found in your own words.
3. Make a claim — and say, in advance, what would prove it wrong.
4. Gather the evidence and write it up, every assertion tied to a source a reader could check.
5. Hand it to people whose job is to find what is wrong with it: someone who asks whether your method can support your claim, someone who knows the field better than you do, someone who argues the opposite case as well as it can be argued.
6. Answer every objection, in writing. Revise.
7. Then stand in a room and defend it, out loud.

None of that is about the paper you receive at the end. It is about how a person earns the right to be trusted with a piece of knowledge.

Here is the problem: almost none of that method lives on the internet.

The internet is extraordinary at answers. Ask anything and you will get one — usually several, often confident, frequently contradicting each other. What you almost never get is the process that would let you judge them. Claims arrive without the question they were meant to answer. Sources are linked, when they are linked at all, as decoration rather than as evidence. Nobody says what would change their mind. Objections, when they appear, appear as a comment-section brawl rather than a committee whose job is to make the argument better. And nothing is defended. It is posted, and then ranked by how many people reacted.

Artificial intelligence makes this sharper, not softer. A model can produce a fluent answer to anything in seconds. Fluency is not the method. An answer nobody has challenged is a draft, however polished it reads.

It is a strange gap when you say it plainly. We would never accept science without the scientific method — results with no hypothesis, no controls, no review. Yet that is close to how most information online is produced and consumed: conclusions without the discipline that would make them conclusions.

It isn't only the PhD

The doctorate is one member of a family. Wherever people have had to get knowledge right — because lives, money, liberty or history depended on it — they built a method with the same bones: a claim, the strongest case against it, evidence both sides can inspect, and a record of how it was settled. Almost every one of them is as absent from the open web as the thesis defense.

The thesis and defenseuniversities

A question, a reading of everything known, a claim that can fail, a committee that attacks it, a defense out loud.

Systematic reviewmedicine · Cochrane

Publish the plan before looking. Find every study, not the convenient ones. Grade each for bias, then pool the results — and report what the evidence can't tell you.

Peer review and replicationscience

Strangers with expertise check the work before it counts; others repeat it to see if it holds.

The adversarial triallaw

A burden of proof, rules for what counts as evidence, and cross-examination — every claim tested by the person most motivated to break it.

The disputationmedieval universities

State the strongest objections first, then your answer, then reply to each objection in turn. Aquinas wrote a whole summa this way.

The fact-check desknewsrooms

Independent confirmation before publishing, an editor who asks "how do we know?", and a public correction when it was wrong.

Competing hypothesesintelligence analysis

List every explanation that could fit, then weigh each piece of evidence against all of them — hunting for what disproves, not what confirms.

Red team and premortemmilitaries · engineering · business

A team whose job is to defeat the plan; or imagine it has already failed and write down why, before you begin.

Forecasting with a track recordforecasting research

State beliefs as probabilities, write them down, and keep score over time, so confidence has to be earned.

The Delphi panelRAND, 1950s

Experts answer anonymously, see each other's reasoning, and answer again — round after round, until they converge or show exactly where they disagree.

Source criticismhistorians

Before using a document: who made it, when, for whom, and what could its author actually have known?

The blameless postmortemengineering

After something breaks, reconstruct exactly what happened and why — to fix the system, not to find someone to blame.

Each of these is a way of turning opinion into knowledge someone else can rely on. Each is a practice most people never see from the inside. And each could be practised by anyone with a question and the patience to follow the steps.

Why PhD.me

You cannot get a PhD from an app, and PhD.me doesn't pretend otherwise. What it gives you is the method — the part that was always the point.

Start with a question and test whether it is specific enough to answer. Build a reading list from the open scholarly record and log what each work found, in your own words. Write a proposal that names what would prove you wrong. Draft, with every citation checked against the sources you actually read. Face a committee that objects to your specific sentences and points you to the work that answers them. Answer every objection in writing. Then defend it.

Do that once and something changes in how you read everything else. You notice which claims come with their question attached and which don't. You start asking what evidence would change a mind — including your own. You start writing for other people the way a good committee would want: sourced, falsifiable, honest about what you don't know.

That is the reason to learn it. Not for the letters — for the habits. To learn better for yourself, and to give better information to everyone else. The internet has no shortage of answers. What it is missing is the method that makes an answer worth trusting.