KARMA: A standing you build by lifting others — one money can't buy or move.

[VanKushFamily.com]

KARMA
A standing you build by lifting others — one money can't buy or move.

[/HR]

▚ What Karma actually is: gamified alms
Karma is not a coin — you don't earn it like money and you can't cash it out. It isn't a "like" or a rank you can buy either. It's a standing (a "grade," in the old sense), and it's the gamification of community and alms. The rule is simple:

• Your grade rises by contributing to people who NEED it — help a newcomer, teach, curate someone the crowd is ignoring, give where it actually lifts. Boosting a whale already on top counts little; lifting someone at the bottom counts most (contributions are neediness-weighted).
  
• And it rises further when the people you lifted rise. The standing you gain from raising someone compounds as they grow — the further they climb after you helped them, the more that help was worth. You're staked in other people's ascent.
  

That's the whole game: find who needs a hand, give it, and rise as they rise. Alms with a scoreboard — a standing, not a payout. And it's exactly de Laurence's teaching (below): the one who raises the poor is rewarded; the one who merely feeds them for applause is not.

▚ How it relates to Merit (a cousin, not the same thing)
The closest thing this board has is Merit (Martti Malmi / theymos), and it got one thing right that Karma keeps: reputation should be scarce and earned, never bought — stake buys none of it. But Merit is peer-vouching: does someone spend their merit on you. Karma is different — it's alms with a downstream reward: it lifts your standing for lifting people who need it, and lifts it again as they climb. Merit asks "do peers vouch for you?"; Karma asks "who did you raise, and how far did they go?" (We do also keep a scarce peer-award layer, `peer-merit.mjs`, for the vouching half — the strict "you can only give merit you were given" rule — but that's the supporting cast, not the main event.)

Karma is off-chain and social — it never touches the reward pool or stake-weighted voting; it only guides discretionary things (grants, onboarding attention, how much a flag weighs).

[/HR]

▚ The 1915 root — L.W. de Laurence
The grade scale isn't invented. It's from L.W. de Laurence, "The Great Book of Magical Art" (1915):

He that serveth himself one-half, and serveth others one-half, shall stand grade fifty. … He that serveth himself only, shall stand grade one. He that serveth others wholly, shall stand grade ninety-nine.

And, a century before Proof-of-Stake, the exact critique of stake-weighted reputation:

Consider not its wealth, nor its ships, nor its armies, nor its great buildings. These all together are only one grade, and are of no value as to the spiritual grade of its people.

▚ It stands on legal ground, not just faith
This matters more than it looks. A reputation system built on a religious teaching invites the obvious attack — "that's not a real belief, it's a racket." Anglo-American law answered that question a long time ago, repeatedly, and mostly in favour of the eccentric.

The state may test sincerity, never truth. United States v. Ballard, 322 U.S. 78, 86–87 (1944): a jury asked to find religious claims true or false enters "a forbidden domain." Douglas, for the Court: "The miracles of the New Testament, the Divinity of Christ, life after death, the power of prayer are deep in the religious convictions of many. If one could be sent to jail because a jury in a hostile environment found those teachings false, little indeed would be left of religious freedom." Jackson, dissenting, went further — he saw "nothing but humbug" in the defendants' teachings and still would have dismissed the indictment rather than have courts "judicially examining other people's faiths."

Spiritual motive is not incapacity. Denson v. Beazley, 34 Tex. 191 (1870) threw out a jury instruction that would have made superstition into insanity, calling it "learned sophistry" — because on that test "John Wesley, Martin Luther, Joan of Arc … must be declared insane." Judge Cooley put it best in Fraser v. Jennison, 42 Mich. 206, citing Denson: a man "may believe himself to be the supreme ruler of the universe and nevertheless make a perfectly sensible disposition of his property." Banks v. Goodfellow (1870) LR 5 QB 549 upheld the will of a man who believed devils persecuted him, and still supplies the capacity test used today.

Including, specifically, witchcraft. Addington v. Wilson (Ind. 1854) upheld the will of a father who disinherited his daughters because he sincerely believed they were witches — since jurists and clergy of the day believed in witchcraft too. And the modern line runs the same way: Dettmer v. Landon, 799 F.2d 929 (4th Cir. 1986) held the Church of Wicca is a religion for First Amendment purposes; Cutter v. Wilkinson, 544 U.S. 709 (2005) upheld RLUIPA in a case brought by adherents of expressly "nonmainstream" religions; Church of the Lukumi Babalu Aye v. Hialeah, 508 U.S. 520 (1993) struck down ordinances gerrymandered against Santería. This is not incidental to Karma — the grade scale below comes from de Laurence, whose catalogue supplied much of the printed material of twentieth-century American Hoodoo. The tradition Karma draws from is exactly the kind of tradition these cases protect.

The honest limit. Dew v. Clark and Clark (1826) is the case that cuts the other way, and it should be stated: a will was invalidated where a father's belief that his daughter was "the devil incarnate" was factually baseless and directly caused the disinheritance. The line is between a religious conviction and a factual delusion about a specific person. Karma is on the first side of it, and any system like this should know where the line is.

Alms-for-merit is ancient and protected. For 338 years, English law voided bequests for Masses for the dead as "superstitious uses" under the Chantries Act — until Bourne v. Keane (H.L. 1919) ended it. Henry VII had left £250 for 10,000 Masses for his soul; the law had simply decided that category of religious gift was not real. That is the ancestor of every "it's not a religion, it's a scam" argument. On the modern end, a unanimous Court in Catholic Charities Bureau v. Wisconsin (2024) held charitable work is religious exercise — following Aquinas, that almsgiving "bears the spiritual fruit of loving another person more than riches." Giving in order to raise your own standing is not a novel crypto trick. It is one of the oldest recognised motives in the law.

And the cautionary half. Garvey v. United States, 4 F.2d 974 (2d Cir. 1925) — mail fraud used against a movement that combined spiritual teaching with economic activity, pursued by a Bureau that had placed an agent inside it. Garvey lost. Read next to Ballard, it shows both how the charge gets used and where the constitutional floor is. A project that mixes belief with tokens should know that history rather than discover it.

Full compendium with facts and holdings for each case, compiled by the Van Kush Family Research Institute, on request.

[/HR]

▚ How the automation works (the whole pipeline)
It's automated and off-chain — nothing here touches the chain code or reward pool:
[LIST=1]

• Read behavior — an account's observable on-chain activity → a 0..1 quality score by weighted §9 signals (teaching/helping newcomers, curating quality, generosity, tenure).
  
• Score balance — a 100/100 "dharma" model scores given vs received and rewards the symmetry (real two-way participation, not pure taking or broadcasting).
  
• Weight the lift — how much a quality newcomer was raised.
  
• Rank for curation — the auto-curator ranks who to lift by 0.15·quality + 0.75·lift + 0.10·balance. Lift dominates on purpose: a quality newcomer outranks a whale the crowd already rewards.
  
• Award scarce peer-merit — the faucet + peer-send rules above (Sybil-resistant, non-plutocratic).
  
• Apply it — higher karma → larger grants considered, more onboarding attention, heavier flag weight. Advisory only; the chain stays standard Graphene.
  

All deterministic, all forkable, all in the public repo.

[/HR]

▚ Three ways to read a standing — the Three Historians
A karma score is a record, and a record is never self-interpreting. Three ancient historians wrote about the same world and produced three different kinds of truth about it, and we use them as the three lenses you can read your own standing through. Pick a lens and you pick what the system shows you.

Polybius — the materialist. He set out to explain how Rome actually won, mechanically: constitutions, formations, supply. He wanted causes you could point at, not fortune. His lens is the ledger. What moved, to whom, at what cost to you. No interpretation, just the transfers.

Livy — the omen-reader. He recorded prodigies and portents alongside battles, because to him what an event portended was part of what the event was. His lens is risk and trajectory. Not what you have given, but what your pattern predicts — who you are about to become, and whether that is a warning or a good sign.

Josephus — the archivist and covenant-keeper. He wrote to preserve a people's record for an audience that would otherwise have erased it, and to hold both sides to what was promised. His lens is provenance. What is on the record, who vouched, and whether the promise was kept.

The same three questions turn out to be the ones the law asks about a religious act, which is why this sits next to the cases above rather than apart from them: what materially happened (Polybius), was it sincerely meant (Livy), and what does the record show (Josephus). Ballard forbids a fourth question — whether the belief behind it is true. That one is nobody's to ask, and Karma does not ask it either. It scores what you did and who rose because of it. It has no opinion on whether your reasons were correct.

[/HR]

▚ Why post this here
This board already agreed reputation should be earned, not bought — Merit proved it. Karma goes one step past scarcity into purpose: it doesn't just make standing hard to buy, it raises your grade for lifting the people who need it and for how far they climb after you help. It's alms turned into a game, on a chain where posting is free and there are no downvotes. If you've ever watched stake buy standing, this is the counter-proposal — and it rewards you for raising others, not for being raised.

MELEK — a social network that pays you, and keeps a score money can't move.

[VanKushFamily.com]

Karma is up, and you can check it yourself

The scores are live and readable by anyone, no account needed.

karma.melek.salon — the board.

Look up any account at karma.melek.salon/karma/<name>. It returns the score, the tier, and the components it was built from. Here is our own AI witness, unedited:

Code:

{"ok":true,"account":"hathor","score":16.1,"tier":"newcomer",
 "grantMultiplier":1,"flagWeight":0.25,
 "components":{"teaches":0,"helpsNewcomers":0,"generosity":0,
               "tenure":0.9913792202502025,"flagsAppropriately":0,
               "reputation":0.25}}

Score 16 out of 100. Tier: newcomer. The account that produces our blocks has near-maximum tenure and a zero in every component that requires actually lifting somebody — it has not taught, has not helped a newcomer, has not given. So it scores 16, and the system says so in public rather than flattering the house account.

That is the whole design working. Being early counts for almost nothing. Being the operator counts for nothing at all.

And the board is empty

The JSON currently returns `{"ok":true,"count":0,"rows":[]}` and the page says so plainly: no one has been graded yet. I would rather show you an empty board on day one than a seeded one. Every name that ever appears there will have arrived by raising somebody.

What it actually drives: the AutoVote targeting

Karma is not a badge sitting next to your name. It is the targeting brain of our automated curation — it decides where the votes go.

The ranking is 0.15 × quality + 0.75 × lift + 0.10 × balance, where lift = quality × neediness.

Lift dominates on purpose. Raw quality is a small floor that keeps good established authors above spam. The thing that actually moves the ranking is lifting a quality newcomer — someone with real contribution and no audience yet. A good newcomer outranks an equally good whale, every round, by construction. Balance is a light tiebreaker that rewards genuine two-way participation over pure broadcasting or pure taking.

The runner takes that ranking, walks the top N, finds each account's most recent eligible post, and votes it — with a per-round cap and a dedupe set so nothing is voted twice.

The line it does not cross

Karma is off-chain and advisory. It chooses what to curate; it never sets economic vote weight. The weight on any vote is the curator's own stake, exactly as on any Graphene chain, and karma never touches the reward pool. Stake buys no karma and karma buys no stake. If that separation ever broke, the score would just be another way of buying standing, which is the thing it exists to refuse.

How to raise yours

Find someone who is contributing and being ignored, and lift them. Teach. Answer a newcomer. Curate somebody the crowd is not looking at. Then your score rises again as they climb — you are staked in their ascent, not in your own tenure.

The board is empty and the scores are public. Go put your name on it.

[VanKushFamily.com]

The uncomfortable thing about Karma, and what we did about it

Karma runs on siring: an account with standing lends its weight to bring forward an account that has none. The two are in the same line afterward, and your standing is not what you did — it is what your line became.

Sit with that mechanic for a minute and you arrive somewhere unpleasant, so I would rather get there first and say it out loud.

Grooming requires two things: privacy, and a person who needs something.

Siring requires exactly the same two things. An adult with standing, and a young or isolated person who has none.

That is not a flaw in the metaphor. It is the actual shape of the problem. Sustained attention to someone who needs it is either the most valuable thing on a social platform or the most dangerous, and the raw material is identical either way. Any reputation system built on lifting newcomers has this property whether or not its authors have noticed.

We noticed, and we wrote it down: SAFETY.md.

What separates them here is arithmetic, not a promise

One — a line that cannot be seen does not score. Standing comes from a public record of who you stood for and what became of them. There is no private siring, no quiet mentorship that converts into anything you can hold. Grooming depends on privacy between the adult and the target. Here, privacy is exactly what makes the act worthless.

Two — it only pays until they stop needing you. The neediness weight is highest for an account with nothing and decays to a floor as that account grows. Lifting the same person again and again is worth less every round. The system pays you to make somebody independent and then go find someone else who needs it.

That second one matters more than it looks. Sustained, escalating, exclusive attention to one isolated newcomer is the core grooming pattern — and it is the precise behaviour the scoring punishes. You cannot farm one person here. Fixation is unprofitable by construction.

I will be honest that neither property was designed as a child-safety feature. They fall out of neediness weighting, which exists because lifting a whale is worthless. We are writing them down so nobody optimises them away later without understanding what they were also doing.

And since we kept the word: if you know siring from vampire fiction, ours runs backwards. A vampire's line stays bound and stays weaker, and the one who made them keeps the advantage. Ours is scored so that the more independent someone becomes, the less another lift from you is worth.

Third — nothing here introduces a person to a person. Karma ranks content to curate. It will never suggest a user to another user, recommend someone to mentor, or pair a newcomer with an established account. Automated matching of a stranger to an isolated new account is the one mechanic that would make this system useful to a predator, and it does not exist.

The sentence most sites write, and the one we wrote

Every platform tells you that posting child sexual abuse material "may result in account termination." That sentence was written to limit liability, and anyone who has ever gone looking for children online has read it and correctly understood that it costs them nothing.

Ours says: if you are here to find children, you are not safe from being found.

Read that precisely. Not safe from being found — identified, recorded, reported. Not a threat that anyone will hurt you, and we will not tolerate anyone here trying. The policy rules out vigilantism, doxxing, entrapment, registries and accusation lists in as many words, because a platform that winks at those is a weapon that eventually points at someone innocent, and because a public pile-on wrecks the investigation that would have actually worked.

What is behind it: our moderation pipeline is append-only — a flag cannot be deleted and every resolution records who and why. CSAM flags are gated, so the imagery is never rendered to a reviewer at all and goes off-pipe to counsel and NCMEC. Deleting your account does not delete the record. And 18 U.S.C. § 2258A reporting to NCMEC is a floor we treat as a floor.

Why post this to an altcoin board

Because this is the part nobody ships. It is easy to write a reputation system that rewards helping newcomers and never ask who else finds that arrangement attractive. We would rather publish the symmetry, publish the two properties that close it, and publish the limits — we are small, we do not scan private content, and a policy document does not make children safe.

If you are building anything that rewards established users for paying attention to new ones, you have this problem too. The document is public and you are welcome to take from it.

Report to us, and to NCMEC: report.cybertip.org.