Re: Introduction: Glogos - logic layer 0 for truth and coordination

Cryptographically verifiable CRDTs is something Dacument already does. It
mostly leaves out the extra hashing, since a signature already functions as
a digest. At this point, though, it is mainly an AI coded starting point, a
first working version. I am planning to parse it down into smaller, more
manageable pieces in the future.

I suggest we could collaborate on writing small, CRDT specific TypeScript
libraries, each with its own small specification that is easy to understand
and adopt. The goal would be to make single purpose CRDT primitives that
can be used in any project, instead of a fully managed solution.

On top of those, a Conflict Free Document Object that uses the primitives
and has a proven or agreed upon access model, asserted locally by each
replica.

I am not focused on AI agents. There is "agentic" decision making involved,
but z-base is meant to become a general foundation for building any web
application in a digital-sovereignty-enabling way.

For example, a similar architecture could be used to build something like
Gmail. A mail service would be only a transport layer, not an authority. It
would relay messages but never own identity, state, or history, and users
could move between services without losing anything. The service could
carry both end to end encrypted messages and ordinary TLS protected
messages. All user state would live in the private data spaces of the
participants.

SMTP would act as a blind carrier. Each inbox would be treated as a
resource, with messages as entries. Email addresses would only resolve to
inbox discovery metadata using an email <-> DID like concept from Amir.

For non email applications, recovery would rely on email in the specific
way described in the walkthrough video.
For an actual email service, that recovery method must obviously not be
used. Email would need its own recovery and continuity model.

This means you could build a z-base principled email system, with small
changes to the email protocol, and directly use z-base as a state
coordination layer for sovereign frontends.

But I think we will be discussing digital-sovereignty-enabling architecture
in more detail in the coming days ;)

ti 3.2.2026 klo 7.53 ap. Thanh M. Le <vnlemanhthanh@gmail.com> kirjoitti:

> Hi Jori,
>
> I took a look at z-base. The architecture is remarkably clean. Just one
> more step, Jori. You will Glogos is the missing piece of z-base.
>
> Regarding the mechanics: You nailed it. "Wall-clock" is a lie; only Causal
> Order is real. That is exactly why I implemented Vector Clocks in my
> interplanetary PoC
> <https://github.com/glogos-org/glo-mars/blob/main/CANON.md> (
> earth_mars_poc.py) to handle sync without consensus.
>
> This is where Glogos fits. I noticed z-base (specifically dacument)
> relies on CRDTs to sync state between devices. Glogos acts as the Blind
> Transport Layer to handle that synchronization via Entanglement & Causal
> DAGs.
>
> Here is how Glogos solves your two main concerns:
>
> 1. The "Lonely Writer" (Fabrication) Solved by Entanglement. As you noted,
> a lonely writer can lie. But z-base is designed for interaction. The
> moment your device syncs with another, your history is pinned. You can't
> rewrite it without conspiring with the network.
>
> 2. The "Prison" (Privacy) You asked: *"What if someone keeps a copy?"*
> This is the hardest problem. I invite you to check
> right-to-be-forgotten.ts
> <https://github.com/glogos-org/glogos/blob/main/examples/use-cases/right-to-be-forgotten.ts>
> in the repo. It demonstrates the "Toxic Asset" strategy (Crypto-shredding):
>
>    -
>
>    Action: Alice encrypts data locally. Glogos stores Hash(EncryptedBlob).
>    -
>
>    Deletion: Alice destroys the Key.
>    -
>
>    Result: The immutable log remains, but the content is mathematically
>    annihilated.
>    -
>
>       If an adversary keeps the data after the "Key Destroyed" signal
>       appears on Glogos, they are demonstrably holding contraband/dead data.
>       -
>
>       For AI Agents (your focus): Glogos is building infrastructure for 1
>       billion Edge AI agents (turning every smartphone into a local node). Humans
>       might be stubborn, but software is compliant. These agents check the Glogos
>       signal; if it says "Tombstone", they automatically drop the local copy.
>
> Conclusion
>
>    -
>
>    z-base handles the Sovereignty (Application Layer).
>    -
>
>    Glogos handles the Physics (Infrastructure Layer).
>
> We are building the same house. I'm just pouring the concrete foundation.
>
> Best regards,
> Thanh M. Le
>
> ---------------------------------------------------------------------------------------
> SHA-256("") — From nothing, truth emerges
> <https://github.com/glogos-org/glogos/blob/main/shared/artifacts/genesis-artifact.json>
> code · cel · cell · citizen · card · cluster · consortium · civilization ·
> cosmos
>
>
> On Tue, Feb 3, 2026 at 11:42 AM Jori Lehtinen <lehtinenjori03@gmail.com>
> wrote:
>
>> Hi once again Manh,
>>
>>
>> I went through your repo again because it kept bothering me that your
>> project still wasn’t clicking for me.
>>
>>
>> So I stripped the system down to what it mechanically guarantees. What I
>> think it really gives you is this:
>>
>>
>> Within a community or platform where your private key is already bound to
>> a known identity or role, you can publish a forward-only commitment log.
>>
>>
>> Others can verify:
>>
>>
>> only you could have written it
>>
>> the internal order is immutable
>>
>> nothing can be removed or rewritten after it is referenced
>>
>> and that each entry happened after the ones it references
>>
>>
>> But they cannot verify:
>>
>>
>> when the history actually started
>>
>> that it matches wall-clock time
>>
>> or that the early part wasn’t fabricated
>>
>>
>> Am I understanding this correctly?
>>
>>
>> Without a central, verifiable authority for time, timestamps can’t prove
>> real-world time. With clock skew and local control, I can set time=0 today
>> and increment it however I want. A trusted frontend does not help, since I
>> control my OS clock. A central Unix server could only prove causality
>> inside a specific service boundary, but not that the timestamp itself is
>> “real,” so I would not market this as a time proving system.
>>
>>
>> What remains is immutable ordering of immutable content. That can make
>> sense in a context where identity is already known, but I still struggle to
>> see where this is valuable outside a trusted environment.
>>
>>
>> If I am the only writer and others only read what I show them, I can
>> prove I wrote things in this order, but I can also fabricate the entire
>> history at any time. The only way to prevent rewriting is if others already
>> hold my full history. Otherwise I can just replace my zone and present a
>> new past.
>>
>>
>> Even then, I do not yet see a strong real-world need to prove that I did
>> not remove or change something between two events, unless multiple writers
>> or external constraints are involved.
>>
>>
>> I do think you are trying to solve a very hard problem, and if you truly
>> solve it, it would be significant.
>>
>>
>> I suggest you watch Martin Kleppmann’s lecture series on distributed
>> systems it explains well that wall time is an illusion in distributed
>> systems, and that logical time (causality) is the best we can do. Unix time
>> becomes a liability.
>>
>> A DAG can show “after-ness,” but not real-world time, and not that a
>> whole history was not swapped unless someone else already holds it.
>>
>>
>> I wish wall time was provable. I wish identity just from a name was
>> provable. That would let us say that someone truly invented something at a
>> certain moment, just from singed content but I do not see how that can be
>> done.
>>
>>
>> GitHub has “immutable” commits and relies on server time (likely
>> synchronized across multiple sources), so at least it can prove unix timed
>> order inside its own system. But even there, I could delete a repo and
>> recreate one with the same name and new history so it is not immutable to a
>> namespace in that sense.
>>
>>
>> If I rely on an external service to store my history and want to show
>> that what I am presenting now was not tampered with, then this starts to
>> make some sense. But even there, I have nothing to verify against if I am
>> fully dependent on that service. At the end of the day, I can only prove
>> that the artifact I show right now is internally consistent, not that it
>> was not created just a second ago.
>>
>>
>> So for now, I do not see “truth emerging from nothing.”
>>
>>
>> I do enjoy technical debate, and I like when my ideas get challenged,
>> it’s the only way to strengthen them. I’m still waiting for real pushback
>> on my Digital-Sovereignty-Enabling Architecture claim. 😎
>>
>>
>> If the video is a bit much, I’ve added other resources to the z-base
>> organization profile:
>>
>> https://github.com/z-base
>>
>>
>> Regards,
>>
>> Jori Lehtinen
>>
>>    -
>>
>>
>>
>> su 1.2.2026 klo 6.15 ip. Jori Lehtinen <lehtinenjori03@gmail.com>
>> kirjoitti:
>>
>>> Hi Manh,
>>>
>>> I want to be honest: I don’t think immutability matters for digital
>>> sovereignty.
>>>
>>> What matters is that the user or entity decides what exists in their
>>> digital space, not a protocol, not a ledger, not an immutable history. From
>>> that perspective, a system that makes all attestations permanent feels like
>>> a direct violation of sovereignty rather than a foundation for it.
>>>
>>> I do see that there may be valid use cases for **immutable**,
>>> cryptographically verifiable histories. I’d be genuinely interested to hear
>>> what those are in concrete, real-world terms. not framed as abstract
>>> inevitabilities or distant futures.
>>>
>>> For contrast, here is a long and awkward walkthrough of a
>>> digital-sovereignty-enabling architecture that is implementable today:
>>>
>>> https://www.youtube.com/watch?v=GIFVetIC8X0
>>>
>>> Regards,
>>> Jori Lehtinen
>>>
>>>
>>>
>>> su 1.2.2026 klo 15.10 Thanh M. Le (vnlemanhthanh@gmail.com) kirjoitti:
>>>
>>>> A Thought Experiment
>>>>
>>>> Imagine the year 2526. A mathematician creates a cryptographic
>>>> attestation claiming the existence of an odd perfect number—one of
>>>> mathematics' oldest unsolved problems. The attestation uses a post-quantum
>>>> hash algorithm developed in 2487, not SHA-256. It references the universal
>>>> anchor (GLR) and carries a valid signature.
>>>>
>>>> {
>>>>   "zone":
>>>> "8f7e6d5c4b3a2918f7e6d5c4b3a2918f7e6d5c4b3a2918f7e6d5c4b3a2918f",
>>>> // Who: hash(public_key)
>>>>   "subject":
>>>> "3a1b2c3d4e5f6a7b8c9d0e1f2a3b4c5d6e7f8a9b0c1d2e3f4a5b6c7d8e9f0a1",     //
>>>> What: hash("Paper: Odd Perfect Number Exists")
>>>>   "canon":
>>>> "b4c5d6e7f8a9b0c1d2e3f4a5b6c7d8e9f0a1b2c3d4e5f6a7b8c9d0e1f2a3b4",        //
>>>> How: hash("opt:math:theorem:1.0")
>>>>   "time": 17556729380000,
>>>>                                                           // When
>>>>   "refs": [
>>>>                                                                          //
>>>> From where:
>>>>
>>>> "03b426423c8a7f3fe1d1204e564efcc9415f1f5d524b3e2fe7dfa78f38756546",
>>>>               //   -> Upgrade post-quantum attestation ID
>>>>     "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"
>>>>                   //   -> GLR
>>>>   ],
>>>>   "proof":
>>>> "c1d2e3f4a5b6c7d8e9f0a1b2c3d4e5f6a7b8c9d0e1f2a3b4c5d6e7f8a9b0c1d2..."    //
>>>> Binding: signature
>>>> }
>>>>
>>>> Fast forward to 3026. A proof emerges showing odd perfect numbers
>>>> cannot exist. The 2526 claim was wrong.
>>>>
>>>> Yet the attestation remains perpetually valid.
>>>>
>>>> This is not a bug. This is the foundation we need to build upon.
>>>>
>>>> The Core Insight
>>>>
>>>> Current web architecture conflates two distinct concepts: the validity
>>>> of a statement's structure and the truth of its content. Existing systems
>>>> often bind identity to location (who controls the server) or authority (who
>>>> issued the certificate) rather than cryptographic proof of history.
>>>>
>>>> Glogos proposes a fundamental separation. An attestation's
>>>> cryptographic validity is independent of its semantic truth. The protocol
>>>> verifies that Zone X committed to Statement Y at Time T through Method M,
>>>> without judging whether Y is true, false, or undecidable.
>>>>
>>>> This separation is not philosophical indulgence. It is architectural
>>>> necessity.
>>>>
>>>> *At the foundation of this architecture lies the Glogos Root (GLR) with
>>>> the value
>>>> "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855". This is
>>>> not merely a soulless hexadecimal string. It crystallizes a design
>>>> philosophy that respects minimalism, objective truth, and physical
>>>> constraints. By choosing the empty string, we declare independence from
>>>> human bias—no language, no culture, no institution owns the hash of
>>>> nothing. By choosing 256 bits, we commit to truth's presence on every
>>>> device, from microcontrollers in remote sensors to data centers processing
>>>> international trade. This anchor point is simultaneously everywhere and
>>>> nowhere, belonging to everyone and no one. It is the only honest starting
>>>> point for a system claiming universality.*
>>>>
>>>> Four Critical Distinctions
>>>>
>>>> 1. Structure versus Semantics. The attestation about odd perfect
>>>> numbers remains structurally valid five hundred years later regardless of
>>>> mathematical truth. Its zone derivation from the public key is verifiable.
>>>> Its signature proves possession of the private key. The protocol supports
>>>> cryptographic agility, verifying proofs according to the algorithm version
>>>> specified in the attestation (e.g., v1:Ed25519, v2:Post-Quantum). This
>>>> ensures the structure remains durable even as the underlying mathematics
>>>> evolves. Its timestamp establishes causal ordering in the directed acyclic
>>>> graph. Its reference to GLR anchors it to the universal root. These
>>>> structural properties are immutable facts about the attestation's creation,
>>>> completely independent of whether the content claim is ultimately proven or
>>>> disproven.
>>>>
>>>> 2. Method versus Purpose. Glogos provides a substrate—a set of
>>>> primitives for creating causally-ordered, cryptographically-bound
>>>> commitments. It does not prescribe how those primitives should be used.
>>>> Scientific communities might build citation graphs. Supply chains might
>>>> track provenance. Legal systems might record contracts. Each application
>>>> layer interprets the same structural elements differently based on
>>>> domain-specific canons. The protocol remains neutral, like TCP/IP
>>>> transmitting packets without knowing whether they contain email, video, or
>>>> financial transactions.
>>>>
>>>> 3. Trust versus Truth. Trust is a relationship between agents. Truth is
>>>> a property of propositions. Glogos facilitates trust by making commitments
>>>> non-repudiable, histories immutable, and provenance verifiable. Users then
>>>> decide whom to trust based on historical accuracy, witness confirmations,
>>>> and consistency with independent sources. The protocol provides raw data
>>>> for informed decisions without claiming to arbitrate truth. This is the
>>>> only sustainable approach in a world where truth claims vary across
>>>> cultures, evolve over time, and remain contested in many domains.
>>>>
>>>> 4. Logic versus Consensus. Traditional blockchains use consensus to
>>>> impose global ordering, requiring all nodes to agree on a single history.
>>>> Glogos uses logic to validate each attestation independently. When two
>>>> attestations conflict, the protocol does not automatically resolve the
>>>> dispute. It stores both. Applications at higher layers decide resolution
>>>> strategies—legal systems might prioritize timestamps, scientific systems
>>>> might require reproducibility, commercial systems might invoke arbitration.
>>>> The absence of forced consensus is not a limitation but an acknowledgment
>>>> that disagreement is a fundamental feature of complex systems, not a bug to
>>>> be eliminated.
>>>>
>>>> The Neutrality Imperative
>>>>
>>>> Many protocols attempt to be intelligent by evaluating content
>>>> validity, preventing "bad" information from being recorded, or enforcing
>>>> specific business rules. This approach creates three problems that compound
>>>> over time.
>>>>
>>>> First, who decides what constitutes "bad" content? Standards change
>>>> across cultures, evolve over time, and depend on context. A protocol that
>>>> embeds current values becomes obsolete when those values shift. The odd
>>>> perfect number example illustrates this perfectly—what seems obviously
>>>> false today might be proven true tomorrow, or vice versa.
>>>>
>>>> Second, how do you update rigid rules embedded in protocol? Hard-coded
>>>> policies become legacy technical debt. The protocol either breaks backward
>>>> compatibility when rules change, or maintains obsolete restrictions that
>>>> constrain future innovation.
>>>>
>>>> Third, what happens to unforeseen use cases? A protocol that judges
>>>> semantics inherently limits creativity. The most important applications are
>>>> often those the designers never imagined. TCP/IP succeeded partly because
>>>> it remained agnostic about payload content, enabling innovations from email
>>>> to streaming video to distributed computing.
>>>>
>>>> Glogos chooses radical neutrality. It accepts all cryptographically
>>>> valid attestations regardless of content, reputation, or meaning.
>>>> Interpretation remains the responsibility of applications built on top.
>>>> This is not abdication of responsibility but recognition that a
>>>> foundational layer must outlive the assumptions of its creators.
>>>>
>>>> Consider the historical parallel. The printing press did not judge
>>>> which books deserved publication. It provided a substrate enabling the
>>>> scientific revolution, the Reformation, the Enlightenment, and countless
>>>> other movements—many contradicting each other. The technology's neutrality
>>>> enabled its civilizational impact.
>>>>
>>>> A Question of Architecture
>>>>
>>>> Does the web need a physics layer? Consider the properties of physical
>>>> reality that digital space currently lacks.
>>>>
>>>> In the physical world, causality is absolute. An event occurring later
>>>> cannot have caused an event occurring earlier. Time flows in one direction.
>>>> These are not social conventions but natural laws. Digital systems have no
>>>> such guarantees. System clocks can be manipulated. Event logs can be
>>>> rewritten. History can be retroactively altered without leaving detectable
>>>> traces.
>>>>
>>>> In the physical world, information is conserved. What happened cannot
>>>> unhappen. You might forget, records might decay, but the past does not
>>>> simply vanish. Digital systems routinely delete history. Websites
>>>> disappear. Databases are reset. The past is mutable, contingent on the
>>>> whims of server administrators.
>>>>
>>>> In the physical world, matter is self-identifying through its
>>>> properties. Chemical composition determines identity regardless of
>>>> location. Digital data lacks this property. Two files with identical
>>>> content but different locations are treated as separate entities. Identity
>>>> derives from position rather than intrinsic characteristics.
>>>>
>>>> Glogos provides digital analogs to these physical properties. The
>>>> directed acyclic graph enforces causal ordering through hash chaining—each
>>>> attestation cryptographically commits to its predecessors, making time
>>>> reversal mathematically impossible. The content-addressed subject field
>>>> makes identity intrinsic—the same content produces the same identifier
>>>> regardless of storage location. The immutable event log preserves
>>>> history—past attestations remain verifiable even as new attestations extend
>>>> the graph.
>>>>
>>>> The Odd Perfect Number Remains
>>>>
>>>> Return to our thought experiment. The attestation from 2526 claiming an
>>>> odd perfect number exists will remain cryptographically valid in 3026
>>>> regardless of whether mathematicians prove or disprove the claim. Its
>>>> perpetuity derives not from the correctness of its content but from the
>>>> integrity of its structure.
>>>>
>>>> This is precisely the foundation we need. Not a system claiming to
>>>> arbitrate truth, but a system making the history of claims transparent,
>>>> unalterable, and verifiable. In a world flooded with misinformation,
>>>> deepfakes, and historical revisionism, clarity is more valuable than
>>>> authority.
>>>>
>>>> The protocol does not create truth. It creates the conditions under
>>>> which truth can be distinguished from falsehood through accumulated
>>>> evidence over time. An attestation proven wrong does not disappear but
>>>> becomes part of the intellectual history—a record of what was believed
>>>> when, why it was challenged, and how understanding evolved.
>>>>
>>>> This is how science actually works. Newton's physics was not deleted
>>>> when Einstein's relativity emerged. It remains a valid description within
>>>> its domain, now understood as a special case of a more general theory. The
>>>> history of wrong but structurally sound claims is as important as the
>>>> history of correct ones.
>>>>
>>>> Realizing the Experiment
>>>>
>>>> Glogos is an attempt to answer these architectural questions. It acts
>>>> not as a competitor to existing standards but as a reference implementation
>>>> of this missing substrate. By separating structure from truth and anchoring
>>>> to external entropy, it seeks to provide verifiable history without the
>>>> constraints of global consensus.
>>>>
>>>> The winter solstice 2025 genesis was a first step—a proof that these
>>>> concepts can function in reality, verified by formal logic and tested
>>>> against the constraints of the physical world. It demonstrates that trust
>>>> can exist without central servers, and global systems can function without
>>>> continuous connectivity.
>>>> This shifts the economics of verification from expensive institutions
>>>> to accessible mathematics, and extends the social reach of the web to
>>>> places where the signal fades but the need for truth remains.
>>>>
>>>> This is about completing the picture—filling the gap for applications
>>>> that demand decades-long verifiability.
>>>>
>>>> The Invitation
>>>>
>>>> The odd perfect number experiment isn't just about architecture. It
>>>> forces us to ask what happens when trust becomes cheap.
>>>>
>>>> The question isn't where this technology fits in your stack. The
>>>> question is: what does society look like when verification is free and
>>>> available to everyone? When we stop relying on central authorities and
>>>> start relying on mathematical history, things change. Costs drop. Friction
>>>> disappears. The web reaches places it couldn't reach before.
>>>>
>>>> If we want a global economy based on sustainable digital trust, the
>>>> need for this layer is obvious.
>>>>
>>>> The web you have built connects humanity. It should also preserve
>>>> humanity's knowledge. That requires foundations as durable as stone tablets
>>>> but as flexible as the digital future.
>>>>
>>>> From nothing, truth emerges—not because we decree it, but because we
>>>> create clarity.
>>>>
>>>> Thanh M. Le
>>>>
>>>> ---------------------------------------------------------------------------------------
>>>> SHA-256("") — From nothing, truth emerges
>>>> <https://github.com/glogos-org/glogos/blob/main/shared/artifacts/genesis-artifact.json>
>>>> code · cel · cell · citizen · card · cluster · consortium ·
>>>> civilization · cosmos
>>>>
>>>>
>>>> On Thu, Jan 15, 2026 at 2:45 AM Manh Thanh Le <vnlemanhthanh@gmail.com>
>>>> wrote:
>>>>
>>>>> Hi everyone,
>>>>>
>>>>> Following the discussion on did:cel and Steven's insightful questions
>>>>> about content integrity, I am opening this thread to introduce the
>>>>> underlying protocol: Glogos.
>>>>>
>>>>> https://github.com/glogos-org/glogos
>>>>>
>>>>> 1. The essence: a logic substrate (L0)
>>>>> Glogos is not just a storage system or an identity method. It is a
>>>>> logic layer 0 for truth and coordination, designed to anchor digital
>>>>> reality without reliance on central authorities.
>>>>>
>>>>> The entire system is anchored to a single, universal mathematical
>>>>> constant: the universal anchor (GLR), which is simply SHA-256(""). From
>>>>> this empty string, we build a directed acyclic graph (DAG) of attestations.
>>>>> This creates a deterministic, tamper-evident history that relies on math,
>>>>> not servers.
>>>>>
>>>>> 2. The vision: from truth to coordination
>>>>> Why do we need a "logic layer"?
>>>>>
>>>>> To reduce information asymmetry and coordination friction.
>>>>> In the functioning proofs of concept, Glogos is powering:
>>>>>
>>>>> *   Public goods: where communities fund shared resources without
>>>>> intermediaries (reducing friction).
>>>>> *   Commitment devices: where individuals signal credibility to
>>>>> overcome trust gaps (reducing asymmetry).
>>>>> *   Resilient trade: where supply chains coordinate even when
>>>>> disconnected (the "oceanic gap").
>>>>>
>>>>> Glogos provides the social physics for these interactions. The
>>>>> protocol treats a "promise" or a "contribution" as a physical
>>>>> object—anchored in the DAG, unforgeable, and universally referenceable.
>>>>>
>>>>> 3. Conclusion
>>>>> By integrating identity (who) with a logic substrate (how we
>>>>> coordinate), the community can ensure that public goods are created and
>>>>> sustained globally.
>>>>>
>>>>> Glogos is offered as a contribution to this shared vision—providing
>>>>> the mathematical grounding needed for resilient digital trust.
>>>>>
>>>>> I welcome discussion on how this logic substrate can serve the broader
>>>>> goals of the Verifiable Credentials community.
>>>>>
>>>>> Best regards,
>>>>> Mạnh Thành Lê
>>>>>
>>>>> ---------------------------------------------------------------------------------------
>>>>> SHA-256("") — From nothing, truth emerges
>>>>> <https://github.com/glogos-org/glogos/blob/main/shared/artifacts/genesis-artifact.json>
>>>>> code · cel · cell · citizen · card · cluster · consortium ·
>>>>> civilization · cosmos
>>>>>
>>>>