The Philosophy (Working Document)

A living document. Add to it whenever inspiration strikes. Raw, unedited captures live in raw/; this file is the distilled, evolving version. Eventually: a book. Its home on the web is freeholders.org.

Not a manifesto. A framework for using AI to empower people instead of control them. Its name, as of 2026-10-20, is Freehold.

The Thesis

Three times in living memory, a new technology promised to democratize power and give it to the people:

The Internet promised open information for everyone. Instead: mass surveillance, spy networks, everything monitored.
Social media promised instant human connection. Instead: fake connection, anxiety, loneliness, and then weaponized narrative control, disinformation, manufactured outrage to keep people fighting, and the burying of what people actually need to know.
AI is the third pivotal moment. It is happening right now.

Each time, the technology was captured by the few and turned into an instrument for controlling the many. As these systems scaled, the individual lost the ability to fight back.

The core question of this philosophy: how do we make sure AI doesn’t become the third great capture? How do we use this inflection point so that the people come together, empower themselves, and use it for the betterment of all humanity, instead of it going to the few who already have enough and only want control?

The Opportunity: The End of “No Free Lunch”

The thesis above is the threat. This is the opportunity: why now is different.

“There’s no such thing as a free lunch.” Lodged in the head since a first economics class senior year of high school, and it has held true for all of human history. Everything required effort. But a twofold change, circa 2024 to 2025 in the US and the West, is breaking that law for the first time:

Near-limitless personal energy. Solar panels (and increasingly batteries) have gotten cheap enough for ordinary consumers worldwide. “The power of the sun in the palm of my hand” (Doc Ock, Spider-Man 2). A family can redirect the monthly power bill into a durable asset that produces energy for a lifetime; roughly a 2,500 sq ft array could power a 2,500 sq ft home for the life of a family. That energy transfers to tools, vehicle propulsion, heavy machinery, or something as simple as powering a laptop.
Near-limitless digital labor. The AI revolution. Set aside the data-center debates (real, but not the point here): the honest, present-day value of AI is a dramatic increase in productivity. Admin work, accounting, and above all the ability to build purpose-built software without hiring an expensive developer.

Combine the two and you get the marker of a post-scarcity society, a term most people would file under Marxism or communism (labels most people, self included, couldn’t actually define by the book). The label doesn’t matter. The marker does: in the digital space, which is so much of the modern economy, speed, access, and resources are becoming effectively unlimited for an individual.

The consequence: the “high-hanging fruit,” things that were never economically feasible because labor cost too much, becomes feasible. Things a capitalist would only do for profit can now be done for human good, because the effort is so small there’s no point in extracting rent from it. We can focus on human good: efficiency and optimization for small businesses, for personal lives, and increasingly for the physical world as we digitize it (Blender models → machining and tooling → real objects).

Honest caveat: today’s AI pricing is subsidized. But we’re early, and if communities invested in the hardware themselves (the energy half already makes this possible), the same results are achievable with limited investment, no subsidy required. Own the panels; eventually, own the compute.

Evidence from lived experience

The two-subscriptions experiment. Running max plans from both Anthropic and OpenAI simultaneously, roughly half of one US minimum-wage worker’s monthly pay, feels not just limitless but unfathomable: dozens of personal projects, two large platforms built at once, this philosophy itself, house projects, all in parallel.
The compression of work. What used to take 2 to 4 weeks (research, planning, building) now takes a weekend, and not even an active weekend. Hand it the plan, go to sleep, wake up to it built.
The Arizona motorcycle club. Connected through Upwork; they’ve spent maybe $600 total (~$300 initial). For a freelancer that barely opens the door; it would be a loss-making simplified app at best. Instead, they collect monthly dues, put part toward improving their communication app, and it works. High-hanging fruit, picked. The empowerment endgame: with these tools they wouldn’t even need to hire anyone.
The license plate museum. A fun idea: an online museum of United States license plates, with generated replica images. Researching, designing, and building it would have cost literal months of a person’s time; the idea was effectively impossible. Now it’s simply a thought, absorbed into a $200/month plan.
The Hasan Minhaj question. “What is the actual value of AI?” A fair question when all most people see is creative slop and features nobody asked for. The answer this philosophy gives: productivity handed to individuals, not just corporations.

The Guiding Principle: Teach a Man to Fish

Give a man a fish and you feed him for a day; teach a man to fish and you feed him for a lifetime. (Not something Jesus literally said, but the principle descends from “I will make you fishers of men.”)

Applied here: don’t build things that do things for people. Build things that teach and equip people to do things for themselves. Education and capability, not dependence. The goal is empowering the people to empower the people.

The pencil corollary: judge the tool by its best users

A pencil is not a useless tool because one person can only draw stick figures with it. The same pencil in the hands of a great artist (or my wife) paints the most beautiful scenes. AI is the same: don’t judge the tool by the many who don’t yet know what to do with it, who’ve never thought with an entrepreneurial mindset. Judge it by the people using it well, and there are a lot of them.

And the aside that makes this a corollary of teach-a-man-to-fish rather than an elitist point: just because the masses don’t know how to use it yet doesn’t mean they can’t be educated to use it. That is the whole point of this.

Influences

Teddy Roosevelt: small-g and big-G government

From his autobiography (a formative book, including, hilariously, an entire chapter earnestly describing bird sounds):

The small-g government is the bureaucracy, the process, the thing people call “the swamp.”
The big-G Government is not an entity out to get you. The big-G Government is the people.
Therefore: if our big-G Government is failing us, we the people have an obligation, a duty, and, most importantly, the means to change it, if we choose to and believe so.

Roosevelt landed in the Presidency at the exact moment America was sliding toward oligarchy, and he fought for the everyday person. Yellowstone and the parks exist because someone thought of future generations, preserved not for utilitarian or extractive value, but for beauty and human enjoyment. American life was not supposed to be about grinding people into dust so capitalists could take everything. Life was about enjoyment. A laborer deserves a fair wage.

Roosevelt delayed the tide and gave the American people a fighting chance. Out of the world wars came FDR and a thriving middle class, because the power was in the people. And slowly, as is the way of nature, the oligarchs fought their way back, using technology and their means to sneak into positions of power and control the people.

The land thread (a connection drawn 2026-10-20, not something either man said). One idea runs through both touchstones: ordinary people owning their ground outright, and the commons kept out of the hands of the few. The Homestead Act (Lincoln, 1862) gave 160 acres to anyone willing to live on it and work it, so ordinary people could own land instead of renting from a landlord; Roosevelt lived that life personally as a Dakota Badlands rancher in the 1880s. Yellowstone is the shared-commons half; homestead is the individual-ownership half. “Freehold” is the legal word for the second one. That’s why the name belongs here.

Lincoln: the leader is not the movement

It was not Abraham Lincoln alone who preserved the Union. He had the belief, the means, the position, and he gave his life for it. But it was all the people who fought and believed that actually brought the Union together. One person, we the people, can do more than any single leader ever could. This philosophy should never depend on its founder.

The math of collective power

There are more of us than there are of them. When a company threatens to sue someone like Channel 5 / Andrew Callahan, 250,000 people donating $5 each, a tiny fraction of America, can throw the lawsuit away. Small contributions at scale beat concentrated wealth. That math is the people’s structural advantage; the technology just needs to make it usable.

Core Commitments

These are the non-negotiables that shape everything built under this philosophy:

Everything at cost. No such thing as a free lunch, but no rent-seeking either. The labor deserves its wage, and nothing more is extracted. No platform fees, no lock-in, no monetizing the users.
Empower, don’t do-for. Tools should build capability in the person using them (teach to fish), especially young entrepreneurs and young people in their communities who can take these tools and lift themselves up.
Not a walled platform. We’re not building a destination everyone must join and pay for. We’re building freely available capability people take with them.
Freely available, but curated. Open to all, but tended, because you don’t know who will do what, and people can get lost in the mess. Curation is stewardship, not gatekeeping.
Decentralized by design. The movement must outlive and exceed any individual, including its founder.
The human owns the product. Because the software is free, it is theirs: their data, their deployment, their choice of providers. Remove the middleman: the only person in the loop is the human, and the middle-person is the AI. They own the means.
No lock-in, by protocol. Because everything speaks MCP, every third-party service is swappable. Happy with your cloud provider? Great. They raise prices? Walk to whoever offers less. Interchangeability is what keeps the third capture from happening: you can’t capture people who can leave.

The Name: Freehold (decided 2026-10-20)

Owning property outright, versus leasehold, where you rent from a landlord. Subscription software is leasehold; Freehold software is yours. It’s a status anyone can earn by meeting a public standard, not a brand anyone owns, which is what makes it a movement word rather than a product name. The people are freeholders; the home is freeholders.org. The full definition, the naming pattern (Freehold + plain noun), and the rules that keep it from collapsing into one person’s project are in FRAMEWORK.md, Part 0.

The Architecture Idea (early sketch)

The technical center of gravity: AI + the MCP protocol + open Git repos.

The entire barrier to entry for a person should be:

An AI key, so they pay for their own usage, at cost. That’s it.
An MCP server (or servers) they connect their agent to.
A curated set of Git repos wired into that MCP: the “pillar functions,” the capabilities.

The shape is almost a no-code solution: a person with zero infrastructure connects an agent to the MCP, and the MCP handles the hard parts (signing into accounts, authorization handshakes, deploying something online), and the person walks away with a real capability they now own.

The owned stack (added 2026-07-30)

The marketplace isn’t just skills the MCP can reach. It’s a core set of free, enterprise-grade software and information people deploy and own themselves.

There is no middleman. The only “middleman” is the person who owns the server, which is the person themselves. Their own deployed instance is the hub; desktop and mobile connect to it. Sync isn’t a service someone runs for them; it’s a capability the software carries with it, set up by the software itself.

The protocol requirement. Anything built for this marketplace must satisfy a standard: the app must be deployable everywhere, or at minimum ship with releases already built for every target, so that all a person does is have their agent take it and deploy it. The person brings their own means (any cloud provider, or their own hardware if they have the ability); we deliberately don’t tackle infrastructure, because it’s too much of a problem and it’s not ours to own. Once deployed, they hook their agent into it, and the server runs on its own and syncs on its own. Cloud is an option, not a requirement: you put it on the cloud only if you need it everywhere you go, which matters now that AI is available on the phone.

Properties of the stack:

Deploy anywhere, handle no infrastructure. The software is freely available and deployable to any cloud server the person chooses. We don’t run infrastructure for them; the AI helps them set it up on whatever provider they pick. Easily customizable by design.
The server carries its own MCP. Once deployed, their instance is MCP-enabled, so their agent can reach it from anywhere.
Sync across desktop, mobile, and cloud. Their data follows them; they choose where it lives.
The local-data problem. AI can teach you the knowledge (e.g., the tax rules you need when starting a business), but you still need somewhere yours to track the actual records. That gap, a trusted, owned, local-first home for your data, is one of the main issues this stack exists to solve.
Zero management burden. Because digital labor is cheap, versioning and upkeep can happen automatically, or pin one version forever and never think about it again. Ownership without the sysadmin tax.
Cost always sits with the consumer, transparently. They pay their cloud provider and their AI key directly. Nobody in between takes a cut.

The QuickBooks example (the clearest picture of the whole stack so far). A new business owner needs accounting. QuickBooks is $20/month, money they may not have. Instead: their agent deploys the free accounting app to wherever they choose, and from then on bookkeeping is just talking. “I bought a $300 device, add it to the database so it’s ready when I file taxes.” The agent reaches the app through MCP, locally on the desktop or on their cloud instance from their phone, and the record is kept, in a database they own, with no subscription. The AI supplies the tax knowledge; their owned stack supplies the trusted place to track it. That’s the local-data problem solved and the middleman removed, in one everyday scene. This app is now Freehold Ledger.

The giving model

Pricing is intrinsically free. Sustainability comes from goodwill: crowd-sharing, and the ability for anyone to donate to the account of whoever built or maintains a thing, if they want to. But donations are a side benefit, never the goal, and never depended on: resources are near-limitless enough that they aren’t required. The goal is good entering the world. The real riches are the positive impact of empowering other people.

Core Pillars (to be built out)

The central open work of this philosophy: identify the core facets of empowerment. Not specific apps, but categories of capability that repos/MCP servers should exist for. Candidates so far:

Education: the original spark (“educational use, the platforms, and MCP”). Learning that builds capability.
Deployment / making: getting something you built into the world with nothing but an agent and a key (the one concrete example so far).
Collective action: the 250,000-people-times-$5 math, made frictionless. (implied by the Channel 5 example; to confirm/develop)
(more to come; add them as they arrive)

The Roadmap

Ramble. Capture the ideas raw as they come (this repo; we are here).
freeholders.org. The movement’s home on the web, in two parts (shape set 2026-10-20):
- The philosophy side: one main page holding the work-in-progress philosophy (this document, published), plus a live feed, blog-post style, where Marcus engages the philosophy and thinks out loud as it evolves (the raw captures are the seed of that feed).
- The marketplace side: the curated registry of Freehold software, starting with Freehold Ledger.
- Originally planned as an arm of jippylong12.xyz; now its own site from the start, with the personal site linking out.
Book. Hone the ideas into the finished argument.

Ideas Inbox

Drop raw ideas here the moment they arrive, no matter how rough. Promote the good ones into pillars above.

2026-07-29: had concrete lifetime examples of pillar functions but they slipped away in the moment. Capture them here next time they resurface.
2026-07-30: the license plate museum, an online museum of US license plates with generated replica images. Both a fun project in its own right and a proof-case for “ideas that were impossible are now just a thought.”
2026-07-30: community-owned compute. Pair the solar argument with AI. If a community invests in its own hardware the way a household invests in panels, the subsidized-pricing caveat disappears. “Own the panels; eventually, own the compute.”

Open Questions

What are the core pillar functions? (The big one.)
What does curation look like in practice? Partly answered: curation is certification against the Freehold Standard, not taste. Still open: who maintains registries, and how do multiple registries coexist?
How does “at cost” work mechanically? (People bring their own AI key; what about hosting, domains, other real costs?)
~~What’s the name?~~ Resolved 2026-10-20: Freehold. Movement home: freeholders.org.
How does this become a book? What’s the spine of the argument? (Working answer: Threat → Opportunity → Method.)
How do you keep an empowerment network from being captured, the way the Internet and social media were? Working answer: interchangeability by protocol. You can’t capture people who can leave.

Log

2026-07-29: Founding capture. Verbatim: raw/2026-07-29-founding-thoughts-verbatim.md. This document created from it.
2026-07-30: Post-scarcity capture. Verbatim: raw/2026-07-30-post-scarcity-verbatim.md. Added “The Opportunity” section (solar + AI ending the no-free-lunch era), the pencil corollary, lived evidence (two-subscriptions experiment, motorcycle club, license plate museum), and the three-step roadmap (ramble → website → book).
2026-07-30 (second capture): Owned-stack capture. Verbatim: raw/2026-07-30-owned-stack-marketplace-verbatim.md. Added commitments 6 and 7 (the human owns the product; no lock-in by protocol), “The owned stack” architecture section (deploy-anywhere software with built-in MCP, desktop/mobile/cloud sync, the local-data problem, zero management burden), and “The giving model” (intrinsically free, goodwill donations as side benefit, impact as the real riches).
2026-07-30 (third capture): No-middleman clarification. Verbatim: raw/2026-07-30-no-middleman-clarification-verbatim.md. Resolved the sync question: the person’s own instance is the hub, sync is set up by the software itself, no one runs anything for anyone. Added the protocol requirement (apps must ship deployable-everywhere releases; the agent performs deployment; the person brings the means; cloud optional) and the QuickBooks example.
2026-07-30 (project setup): Made the folder self-contained for any future agent: created README.md (orientation + workflow) and FRAMEWORK.md (the 10-point standard, the 8-step order of operations, and the scoped first application: simple accounting, v1 = local core + MCP surface). Linear project created for the app; cloud descoped to a future stage (Desktop → Cloud → Mobile).
2026-10-10: First public articulation. A Reddit reply draft in drafts/2026-10-10-reddit-reply.md; seed copy for the site.
2026-10-20: Named it. Verbatim: raw/2026-10-20-freehold-name-and-site-verbatim.md. The movement is Freehold, the people are freeholders, the home is freeholders.org (purchased on Cloudflare the same day, $8.50 first year; freehold.com/.org/.dev/.app/.io were taken). The freehold repo is the philosophy and the website only (site in site/); apps like Freehold Ledger get their own repos. Follow-up once the site is live: a Freehold section on jippylong12.xyz linking out. Added FRAMEWORK.md Part 0 (the definition, the naming pattern, the movement rules), the land-thread note under Roosevelt, and the two-part site shape under the Roadmap. Moved everything out of jippylong12xyz/philosophy/ into its own repo at ~/Programming/Personal/freehold. Linear project renamed Freehold Ledger. Em dashes scrubbed from the distilled docs per Marcus (raw captures untouched by design).
2026-10-20 (site): Decided how freeholders.org gets built and started building it in site/: Astro with Vue islands, Tailwind, static output on Cloudflare Pages at $0 hosting. The philosophy page renders this document from the repo at build time, so the site never holds a copy; /standard renders FRAMEWORK.md the same way. The feed is one Markdown file per post with RSS. The marketplace is a registry of certification records against the 10 points, rendered as a checklist and published as /registry.json so other registries and agents can read it. Licenses set: CC BY-SA 4.0 for the writing, MIT for the site code. Follow-up once live: the Freehold section on jippylong12.xyz.