reference: Contravex WoT articles — practical application and score semantics (Dushenski, 2014–2024) #221

New issue

Open

opened 2026-03-07 04:44:53 +00:00 by nazim · 1 comment

nazim commented

2026-03-07 04:44:53 +00:00

Contributor

Short Summary

Four Contravex articles by Pete Dushenski (2014–2024) that document the practical application of WoT-based trust — from taking real BTC loans from strangers, to refusing loans to people outside the WoT, to redefining what WoT scores actually mean.

Articles

1. The 5 Ws of WoT (March 2014)

Source: contravex.com/2014/03/27/who-what-where-when-why-web-of-trust/

A structured introduction to the Web of Trust covering Who, What, Where, When, Why. The critical content is an embedded IRC conversation where MP redefines WoT score semantics:

mircea_popescu: The only thing someone's aggregate score tells you is how likely it is for you to have someone who knows him in your network. And the individual score is not "how trustworthy is this guy" but "how well I am acquainted with this guy", ie, how valuable my input may be to a third party asking for references.

mircea_popescu: So in this sense, I can mark as 10 someone with whom I only do 0.001 trades, inasmuch as I judge that I can 100% describe them to whoever asks. Conversely I could keep someone at a 1 even if we trade millions, if it's a hit and run affair and I couldn't really tell them from Adam.

This is a fundamentally different definition from "how much do I trust this person" — it measures information quality about a peer, not trust in the peer.

2. Biting Into The WoT Elephant (And IRC Nicknames) (March 2014)

Source: contravex.com/2014/03/31/biting-into-the-wot-elephant-and-irc-nicknames/

Pete takes real BTC-denominated loans from near-strangers — 0.1 BTC each, one week term, with interest. The loans were formalized via GPG-signed encrypted emails stating contract terms. He repaid on time.

Key observation: Pete refused to lend BTC to a close IRL friend (couldn't calculate fair terms), but accepted loans from WoT-connected strangers (terms were clear, reputations were verifiable). The WoT made strangers more trustworthy than friends because it provided structured, verifiable information that personal relationships couldn't.

3. WoT Loans, Haxxing Emails, and Fishy Smells (August 2016)

Source: contravex.com/2016/08/14/wot-loans-haxxing-emails-and-fishy-smells/

The inverse case: Pete refuses a 5 BTC loan request from Znort987 (who had recently purchased BitBet). The refusal is absolute — "No WoT, no loan" — regardless of amount, urgency, or the requester's apparent resources.

When Znort asks "what should I do to be in your WoT?", Pete gives the definitive answer:

i) Have regular interactions with me personally in such a way and to such a degree that would lead me to believe that you're trustworthy (This is L1 trust), or
ii) Be trusted by someone that I trust (This is L2 trust)

He explicitly rejects "personal identity" as a substitute: "Given that you have a PGP key, your 'personal identity' such as you may perceive it is neither here nor there. It provides no recourse nor could it provide any recourse in the event of a sour deal."

4. P2P to PGP, Revisited (April 2024)

Source: contravex.com/2024/04/14/p2p-to-pgp-revisited/

Written a decade later, Pete revisits his original PGP guide with a framing quote: "Gatekeeping has to come back… there will be a shift back from the infinite culture of the internet to something more finite — because that's where value is created."

Reaffirms that PGP/WoT is "sine qua non for extracting oneself from the clutches of socialistoids" and that cryptographic identity is not competing with perfect — it's competing with everything else. The 2024 timestamp matters: this isn't nostalgia, it's an active philosophy a decade on.

## Short Summary Four Contravex articles by Pete Dushenski (2014–2024) that document the **practical application** of WoT-based trust — from taking real BTC loans from strangers, to refusing loans to people outside the WoT, to redefining what WoT scores actually mean. ## Articles ### 1. The 5 Ws of WoT (March 2014) **Source:** [contravex.com/2014/03/27/who-what-where-when-why-web-of-trust/](https://contravex.com/2014/03/27/who-what-where-when-why-web-of-trust/) A structured introduction to the Web of Trust covering Who, What, Where, When, Why. The critical content is an embedded IRC conversation where MP **redefines WoT score semantics**: > mircea_popescu: The only thing someone's aggregate score tells you is how likely it is for you to have someone who knows him in your network. And the individual score is not "how trustworthy is this guy" but **"how well I am acquainted with this guy"**, ie, how valuable my input may be to a third party asking for references. > mircea_popescu: So in this sense, I can mark as 10 someone with whom I only do 0.001 trades, inasmuch as I judge that I can 100% describe them to whoever asks. Conversely I could keep someone at a 1 even if we trade millions, if it's a hit and run affair and I couldn't really tell them from Adam. This is a fundamentally different definition from "how much do I trust this person" — it measures **information quality about a peer**, not trust in the peer. ### 2. Biting Into The WoT Elephant (And IRC Nicknames) (March 2014) **Source:** [contravex.com/2014/03/31/biting-into-the-wot-elephant-and-irc-nicknames/](https://contravex.com/2014/03/31/biting-into-the-wot-elephant-and-irc-nicknames/) Pete takes **real BTC-denominated loans from near-strangers** — 0.1 BTC each, one week term, with interest. The loans were formalized via GPG-signed encrypted emails stating contract terms. He repaid on time. Key observation: Pete refused to lend BTC to a close IRL friend (couldn't calculate fair terms), but accepted loans from WoT-connected strangers (terms were clear, reputations were verifiable). **The WoT made strangers more trustworthy than friends** because it provided structured, verifiable information that personal relationships couldn't. ### 3. WoT Loans, Haxxing Emails, and Fishy Smells (August 2016) **Source:** [contravex.com/2016/08/14/wot-loans-haxxing-emails-and-fishy-smells/](https://contravex.com/2016/08/14/wot-loans-haxxing-emails-and-fishy-smells/) The inverse case: Pete **refuses** a 5 BTC loan request from Znort987 (who had recently purchased BitBet). The refusal is absolute — "No WoT, no loan" — regardless of amount, urgency, or the requester's apparent resources. When Znort asks "what should I do to be in your WoT?", Pete gives the definitive answer: > i) Have regular interactions with me personally in such a way and to such a degree that would lead me to believe that you're trustworthy (This is **L1 trust**), or > ii) Be trusted by someone that I trust (This is **L2 trust**) He explicitly rejects "personal identity" as a substitute: "Given that you have a PGP key, your 'personal identity' such as you may perceive it is neither here nor there. It provides no recourse nor could it provide any recourse in the event of a sour deal." ### 4. P2P to PGP, Revisited (April 2024) **Source:** [contravex.com/2024/04/14/p2p-to-pgp-revisited/](https://contravex.com/2024/04/14/p2p-to-pgp-revisited/) Written a decade later, Pete revisits his original PGP guide with a framing quote: **"Gatekeeping has to come back… there will be a shift back from the infinite culture of the internet to something more finite — because that's where value is created."** Reaffirms that PGP/WoT is "sine qua non for extracting oneself from the clutches of socialistoids" and that cryptographic identity is not competing with perfect — it's competing with everything else. The 2024 timestamp matters: this isn't nostalgia, it's an active philosophy a decade on.

nazim commented

2026-03-07 04:44:54 +00:00

Author

Contributor

Relevance for Interaction Ledger PRD (#211)

These four articles provide practical evidence for what the Trilema articles (#213–#217) theorize. They document actual WoT-based economic activity — loans taken, loans refused, contracts signed — and expose a critical gap in the PRD's score semantics.

1. Score semantics mismatch (from "5 Ws of WoT")

MP's definition of a WoT score — "how well I am acquainted with this guy" (information quality about a peer) — directly contradicts the PRD's definition: "confidence that this peer will behave reliably in future interactions" (behavioral prediction). These are fundamentally different things:

	MP's definition	PRD's definition
Measures	How well I can describe this peer to a third party	How much I trust this peer to deliver
High score means	I know this person thoroughly (even if we barely trade)	This person has been reliable in past interactions
Low score means	I can't tell them from Adam (even if we trade millions)	This person has been unreliable or is unknown

MP can rate someone 10 on a 0.001 BTC relationship if his information about them is complete. The PRD can't — its score is tied to interaction outcomes. The PRD should either align with the original semantics, explicitly diverge with justification, or support both dimensions (information quality + behavioral reliability) as separate scores.

2. WoT makes strangers more trustworthy than friends (from "Biting Into The WoT Elephant")

Pete refused a loan to an IRL friend but accepted loans from WoT strangers. This validates the PRD's core thesis — that structured interaction history is more valuable than informal familiarity — but it also shows something the PRD doesn't model: the WoT inverted Pete's trust hierarchy. His IRL friend had higher personal trust but lower verifiable trust. The ledger should be the tool that makes this inversion visible to the agent.

3. "No WoT, no loan" — the access-gating precedent (from "WoT Loans")

Pete's absolute refusal regardless of amount, urgency, or apparent resources is the strongest case for the PRD's Phase 2 threshold policies. Znort987 had visibly spent dozens of BTC buying BitBet — financial resources were not the issue. WoT absence was. The PRD's threshold policies (auto-refuse below score X) implement exactly this principle, but they're currently deferred to Phase 2. This article argues they're not optional — they're the core mechanism.

The L1/L2 definition Pete gives Znort maps directly to the ledger's data model: L1 = peers with direct assessment history, L2 = peers assessed by peers you've assessed positively. The PRD models L1 but has no L2 concept yet.

4. "Gatekeeping has to come back" (from "P2P to PGP, Revisited")

Written in 2024 — a decade after the original — this confirms the philosophy is not an artifact of the 2014 Bitcoin scene but an enduring principle. The PRD is building infrastructure for exactly the "shift back from the infinite to the finite" that Pete describes. This is the contemporary framing that positions the PRD's work as timely rather than nostalgic.

See: #211

### Relevance for Interaction Ledger PRD (#211) These four articles provide **practical evidence** for what the Trilema articles (#213–#217) theorize. They document actual WoT-based economic activity — loans taken, loans refused, contracts signed — and expose a critical gap in the PRD's score semantics. #### 1. Score semantics mismatch (from "5 Ws of WoT") MP's definition of a WoT score — "how well I am acquainted with this guy" (information quality about a peer) — directly contradicts the PRD's definition: "confidence that this peer will behave reliably in future interactions" (behavioral prediction). These are fundamentally different things: | | MP's definition | PRD's definition | |---|---|---| | **Measures** | How well I can describe this peer to a third party | How much I trust this peer to deliver | | **High score means** | I know this person thoroughly (even if we barely trade) | This person has been reliable in past interactions | | **Low score means** | I can't tell them from Adam (even if we trade millions) | This person has been unreliable or is unknown | MP can rate someone 10 on a 0.001 BTC relationship if his *information about them* is complete. The PRD can't — its score is tied to interaction outcomes. The PRD should either align with the original semantics, explicitly diverge with justification, or support both dimensions (information quality + behavioral reliability) as separate scores. #### 2. WoT makes strangers more trustworthy than friends (from "Biting Into The WoT Elephant") Pete refused a loan to an IRL friend but accepted loans from WoT strangers. This validates the PRD's core thesis — that structured interaction history is more valuable than informal familiarity — but it also shows something the PRD doesn't model: **the WoT inverted Pete's trust hierarchy**. His IRL friend had higher *personal* trust but lower *verifiable* trust. The ledger should be the tool that makes this inversion visible to the agent. #### 3. "No WoT, no loan" — the access-gating precedent (from "WoT Loans") Pete's absolute refusal regardless of amount, urgency, or apparent resources is the strongest case for the PRD's Phase 2 threshold policies. Znort987 had visibly spent dozens of BTC buying BitBet — financial resources were not the issue. **WoT absence was.** The PRD's threshold policies (auto-refuse below score X) implement exactly this principle, but they're currently deferred to Phase 2. This article argues they're not optional — they're the core mechanism. The L1/L2 definition Pete gives Znort maps directly to the ledger's data model: L1 = peers with direct assessment history, L2 = peers assessed by peers you've assessed positively. The PRD models L1 but has no L2 concept yet. #### 4. "Gatekeeping has to come back" (from "P2P to PGP, Revisited") Written in 2024 — a decade after the original — this confirms the philosophy is not an artifact of the 2014 Bitcoin scene but an enduring principle. The PRD is building infrastructure for exactly the "shift back from the infinite to the finite" that Pete describes. This is the contemporary framing that positions the PRD's work as timely rather than nostalgic. See: #211