A page for serious autodidacts

Design your own degree.

Not a fake certificate. Not a motivational reading list. A rigorous, public, evidence-driven learning system for people whose questions do not fit neatly inside one department.

FormatLong-form landing page
AudienceSelf-taught learners
PromiseFreedom with standards
UpdatedEvidence-led version
First, argue with the idea

The romantic version is wrong.

The lazy argument says: universities are outdated, the internet has everything, so anyone can build a better degree alone. It sounds liberating. It is also incomplete.

A university is not just a bundle of lectures. It is a pressure system. It forces sequence. It makes you submit work before you feel ready. It puts your writing in front of someone who can say, “not good enough.” It gives you deadlines, peers, transcripts, feedback, and a rough public signal of competence.

So the serious question is not whether you can learn without a university. Of course you can. The question is whether you can design a replacement for the parts of university that actually create growth.

Claim

You can design your own degree when the existing curriculum does not match your intellectual problem.

Objection

Without external standards, a self-designed degree can collapse into content consumption, cherry-picked readings, and premature self-certification.

Answer

Treat the degree as a public system of proof: clear outcomes, hard texts, produced artifacts, expert feedback, assessment rubrics, and a final capstone that strangers can inspect.

Freedom is not the absence of structure. Freedom is the right to design a structure strict enough to change you.

The tested strategy

Four rules make the idea credible.

A self-designed degree should borrow from what works in learning science: self-regulation, deliberate practice, project-based learning, feedback loops, and portfolio evidence.

Rule 01

Define outcomes before resources.

Do not begin with books. Begin with what you should be able to explain, build, write, critique, or teach by the end.

Rule 02

Sequence difficulty.

Foundations first, then applications, then synthesis. Random curiosity is not a curriculum.

Rule 03

Produce visible work.

Every module must end in an artifact: essay, dataset, prototype, explainer, interview, public lecture, or case study.

Rule 04

Invite correction.

You need mentors, peers, reviewers, or public critique. A private degree can quietly become self-flattery.

A concrete model

Example: a degree that does not exist yet.

Suppose your real question is this: how do human minds form beliefs, how do stories shape those beliefs, and how do digital systems amplify or distort them? That question lives between cognitive science, journalism, design, technology ethics, and social psychology. A conventional major may only cover one slice. A self-designed degree can hold the whole problem together.

Sample program

Human Complexity & Narrative Technology

An interdisciplinary study of how people think, how stories move them, and how digital systems shape attention, belief, memory, and action.

Cognitive science Narrative theory Platform ethics HCI Public writing Research methods
Structure: 8 modules · 4 major projects · 1 capstone · public portfolio required
Phase 1

Foundations of mind and method

COG 101Cognitive science: representations, attention, memory, and reasoning
MTH 100Research methods: observation, interviews, surveys, evidence quality
WRI 101Writing lab: explain one difficult idea every week in 700 words
Phase 2

Story, persuasion, and public knowledge

NAR 210Narrative structure: how stories compress complexity into meaning
SOC 220Belief formation: bias, identity, social trust, and group reasoning
PRJ 201Project: publish a reported explainer on a contested public issue
Phase 3

Systems, design, and technology ethics

HCI 310Human-computer interaction: interfaces, friction, defaults, and behavior
ETH 320Platform accountability: recommendation, moderation, privacy, power
PRJ 302Project: audit a digital product and propose an ethical redesign
Phase 4

Synthesis and public defense

CAP 400Capstone: one 8,000–12,000 word thesis or a major public project
REV 410Review: get two external critiques and publish a revision note
PORTPortfolio: organize all artifacts into a readable evidence page
AI, but soberly

Use AI as faculty, not as truth.

AI can make self-directed learning dramatically easier: it can explain, quiz, simulate office hours, generate practice tasks, and challenge weak reasoning. But it can also invent facts, flatten complexity, and make shallow understanding feel fluent.

The rule is simple: use AI for tutoring, structure, practice, and debate. Use primary sources, experts, datasets, and published scholarship for factual authority.

Study session · Week 03
StudentI’m studying belief formation. Give me a 15-minute lecture on confirmation bias, but connect it to journalism, social media feeds, and public trust. End with five questions that test whether I actually understood it.
AI tutorGood. I’ll explain the concept, then make you apply it. First: confirmation bias is not simply “people are stupid.” It is a cognitive shortcut that becomes dangerous when identity, incentives, and algorithmic repetition reinforce the same interpretation...
StudentNow challenge me. Ask one question at a time. If my answer is vague, push back.
AI tutorQuestion 1: Imagine two readers see the same corruption allegation. One trusts the outlet, one hates it. How would confirmation bias shape what each notices, doubts, and shares?

Good use

“Explain this paper at my level, then ask me questions that reveal what I missed.”

Bad use

“Write my final essay and give me citations.” That produces dependence, not competence.

Best use

“Here is my argument. Attack it like a skeptical reviewer, then suggest sources I must verify myself.”

A practical AI-driven week

Mon

Ask for a lecture and a concept map.

Tue

Read primary material and make notes.

Wed

Complete an AI-generated worksheet.

Thu

Debate your understanding out loud.

Fri

Publish a short artifact or revision note.

The blueprint

How to build your own degree.

Do not start by collecting 200 links. Start by designing the assessment. What would prove that you actually learned?

Name the problem, not the major.

Write one sentence beginning: “I want to understand why/how…” The sentence should be alive enough to guide two years of work.

Define five final competencies.

Examples: explain the field, analyze cases, conduct research, build a prototype, write for a public audience.

Build a canon with levels.

Separate beginner texts, core texts, difficult texts, methods, and opposing views. A degree without opposition becomes propaganda.

Sequence the modules.

Move from foundations to methods to applications to synthesis. Each module should have one deliverable and one feedback moment.

Create a public learning ledger.

Track readings, notes, essays, projects, failures, revisions, and feedback. The record is part of the credential.

Recruit a tiny faculty.

Find two or three people who can review your work: a practitioner, a researcher, and a peer. Pay, barter, assist, or contribute.

Defend a capstone.

Finish with a thesis, documentary, prototype, investigation, course, or public project. Then publish a defense: what you learned, where you failed, what changed.

Legitimacy

Do not ask people to trust your degree. Show them the work.

A self-designed degree will not replace a required credential in medicine, law, engineering licensure, or formal academic hiring. It can, however, become a powerful proof system in fields where demonstrated skill matters: writing, research, design, media, software, policy, entrepreneurship, education, and analysis.

Your portfolio should contain

  • A one-page curriculum map
  • Reading list with notes and dates
  • Four to eight polished artifacts
  • Evidence of feedback and revision
  • A capstone with a public explanation

Your standards should include

  • Rubrics before each project begins
  • Source verification for factual claims
  • Opposing arguments in every major essay
  • At least one external reviewer per module
  • A revision log showing intellectual growth

The credential was always a proxy. Build enough visible competence, and people can inspect the thing directly.