◇ Values Commons

For fundersGrant brief, revised 29 Sept 2026.

Values Commons works. Nobody outside the project has used it yet.

I built it alone. You can open the app today and rank a coffee, a bank or a messenger by what you care about, with a source for every scored fact. What I cannot show you is a stranger using it: I have no record of one. I am asking for $35,000 for six months to find out whether it helps people who did not build it, and to publish what I find either way.

Want to kick the tires first? The tour takes two minutes and uploads nothing.

Take the 2-minute tour Open the app Read the full written brief

Plate IFig. 1. What is built and working. The hatched marks are things nobody has done yet.

What exists, and what does not

Conscious Consuming is the first app. It lets a person set rules they keep, make a trade-off between what they care about, and compare real products, services, media, organisations and causes. It is a set of static files, so it installs like an app, works offline and keeps your values on your device.

  • The app, liveConscious Consuming ranks options by your weights. Nobody can pay for a place in the order.
  • Two more apps on the same engineKosplora, a learning atlas, and a guide to choosing a messenger. Neither needed an edit to the engine.
  • The open standard the apps shareYour values as a file you keep, facts as files anyone can correct, and group decisions made by combining files in the browser. The standard, and how the files travel.
  • A source for every scored factMore than 120,000 scored facts, and every one names where it came from. Where the evidence is too thin, the awards ledger withholds the mark and says so.
  • A stranger using itNone on record. I keep a sheet for sitting a first-time visitor down for ten minutes and writing down where they stop. It is still empty.
  • A group deciding something real with itThe group tools work in the browser. No group outside the project has used them.
  • Another community building its own appThe app builder page drafts a starter guide in the browser. Nobody outside the project has made one.
  • Outside money$0 received from anyone. The funding ledger is public and has no entries.

built and workingnot done by anyone yet

Counted 29 Sept 2026140+ decision categories covered24,000+ entries140+ published guides3,600+ comparison pages300+ measures withheld400+ top marks

Counts read from the app's own data files and the awards ledger. Money: the funding ledger.

I publish one weakness myself. Only about a quarter of the entries are backed by more than one independent source. The others have one, often the seller's own website. A source is a place you can check, and it is not always a second opinion.

Honest information about everyday choices has no business model

Tools that help people choose are usually paid for by advertising, affiliate commissions, brand deals or the users' data. Each of those gives someone a reason to push the ranking one way. The labels are not much help either. The European Commission reports that 53% of green claims it examined gave vague, misleading or unfounded information, and 40% had no supporting evidence.

Source: European Commission, Green claims, citing its study Environmental claims in the EU (2020).

This project refuses all four ways of paying for itself, so somebody has to pay for the time instead. The refusal is written down before any money arrives. The ledger caps money from any brand, maker or service the project rates, or might rate, at $0, forever, and it caps any single grant at $35,000 in a year.

What the money buys

$35,000

Over six months. It pays for the first use by people outside the project: my time, the outreach to find first users and one or two groups, the watching and the writing up.

It buys

  • 20 to 50 people making real decisions with the app, and telling me what helped, what confused them and what was missing.
  • One or two real groups taking a shared decision from their own rules to a plan, with any disagreement left visible.
  • One outside community planning an app for a subject it knows better than I do.
  • Fixes for the problems those people run into, and for nothing else.

It does not buy

  • Servers. The site is static files and costs almost nothing to serve.
  • Features nobody asked for. The app already has far more features than it has users.
  • A say over any ranking. No funder gets placement, coverage, wording or data. The ledger says so in writing.
  • A good result. If nobody finds it useful, the report will say that.

What I would test

  • One person. Can someone who did not build it use it for a real choice, and say why they picked what they picked?
  • One group. Can a group reach a shared plan without losing the people who disagree?
  • One new subject. Can another community plan a useful app for something it knows?

Plate IIFig. 2. Six months of work, one job at a time.

Six months of work

Each stretch has one job, and the last one is to publish what happened.

  1. Month 1Invite the first usersName the decisions, invite 20 to 50 people, and ask for feedback without accounts or tracking.
  2. Months 2 to 3Watch real choicesSit beside each person for ten minutes without helping, write down where they stop and what they think each part is for, in their words, and fix what the page failed to explain.
  3. Months 3 to 4Try one group decisionWork with a real group, then publish the blockers, the disagreement and the reasons behind the decision.
  4. Months 4 to 5Plan one new appHelp an outside community describe its subject, starter facts, privacy limits and correction route before any code.
  5. Month 6Improve the adoption guideRevise the public guide using what the first users and the outside community needed. Read the current guide.
  6. CloseoutPublish what happenedFindings, the group's result, the new-subject brief and the open questions, with the failures left in.

What's happening.Six months of people outside the project using the app for the first time. It starts with people making real choices on their own phones and ends with a public write-up, including whatever did not work.

Why it matters.Everything solid in Plate I was built by me and tested by me, on me. A tool for choosing by your own values is only worth funding if it helps someone who did not build it, and that is the one check none of my audits can run.

Devil's advocate.Twenty to fifty volunteers is a pilot, and a small one. People who agree to try a values app already like values apps, so the first cohort will flatter it. The method is also hard on me in particular. It asks me to sit beside a stranger using something I built and say nothing for ten minutes.

Rabbit hole.Run the first test on yourself. The tour follows one choice from what you care about to a group's plan, on the app's real messaging data.

Sit someone down with it for ten minutes and write in with where they stopped. It would be the first entry on that sheet.

What I will report back

  • Whether the app helped a first cohort choose more by their own values, in their words, including the people it did not help.
  • At least one real group decision made with the tools, or a plain account of what blocked it.
  • A brief for one new subject, written with the community that knows it.
  • Corrections sent back to the open-data sources wherever the first users found errors.

Who is asking

I'm Bentley Moon, a solo maker. I designed and built the app, the two smaller apps, the engine under all three, the guides and the group tools. I also run an AI-evaluation lab at bentleymoon.com under the same rule as the withheld marks here. When my own tests refute one of my results, the result stays in the record, marked as refuted, with the date.

To talk it through, write to futurisminstitute@gmail.com. I'll give a live walkthrough, or answer diligence questions in writing.