MindFry vs. The World

MindFry vs. The World

MindFry occupies a unique position in the database landscape. This page clarifies what it is (and isn’t) by comparing it to familiar technologies.

Not a Replacement

MindFry is not designed to replace your production PostgreSQL or Redis. It’s a specialized substrate for cognitive applications that benefit from subjective memory behavior.

MindFry vs. Redis

Aspect	Redis	MindFry
Identity	Key-Value Cache	Cognitive Memory Substrate
Forgetting	Explicit TTL (time-based)	Organic Decay (access-based)
When forgotten	Bug / configuration error	Expected behavior
Relationships	None (flat namespace)	Synaptic Bonds (+/-)
Query side effects	None	Observer Effect
Use case	Session cache, rate limiting	Cognitive agents, adaptive AI

The Key Insight

Redis forgets because you told it to. MindFry forgets because no one cared.

In Redis, a TTL is a contract: “delete after N seconds”. In MindFry, decay is probabilistic and access-dependent:

Redis:  created_at + TTL < now → DELETE
MindFry: energy < threshold(mood) → DORMANT (still exists, inaccessible)

---

MindFry vs. Neo4j

Aspect	Neo4j	MindFry
Graph Model	Property Graph (static)	Synaptic Graph (living)
Edges	Named relationships	Polarized Bonds (+1, 0, -1)
Traversal	Cypher queries	Propagation cascades
Persistence	Permanent	Ephemeral (decay)
Query cost	Computational	Energetic (Observer Effect)

The Key Insight

Neo4j relationships describe. MindFry relationships act.

In Neo4j, an edge like [:KNOWS] is metadata. In MindFry, a bond actively transmits energy:

graph LR
    subgraph Neo4j
        A1[Alice] -->|KNOWS| B1[Bob]
        A1 -->|WORKS_AT| C1[Acme]
    end

    subgraph MindFry
        A2[alice<br/>E=1.0] -->|+1| B2[bob<br/>E+=0.5]
        A2 -->|-1| C2[enemy<br/>E-=0.5]
    end

Querying “alice” in Neo4j returns static data. Stimulating “alice” in MindFry changes the graph.

---

MindFry vs. Vector Databases

Aspect	Vector DB (Pinecone, Milvus)	MindFry
Similarity	Cosine / Euclidean (mathematical)	Synaptic strength (biological)
Embedding	Required (ML preprocessing)	None (raw keys)
Association	Nearest neighbors	Bond propagation
Dynamics	Static (re-index to update)	Living (continuous decay)
Personality	None	8-dimensional Octet

The Key Insight

Vector DBs find similar things. MindFry feels related things.

Vector similarity is computed: cos(embed(A), embed(B)). Bond strength is emergent: formed through co-access, strengthened by stimulation, and filtered by mood.

// Vector DB: mathematical distance
const similar = await vectorDB.query(embed('cat'), topK: 5);
// Returns: ['kitten', 'dog', 'pet', 'animal', 'fur']

// MindFry: synaptic association under current mood
await brain.lineage.stimulate({ key: 'cat', delta: 1.0 });
// In euphoric mood: 'kitten', 'purr', 'warm' activate
// In depressive mood: threshold higher, less propagation

---

When to Use MindFry

✅ Good Fit

• Conversational AI with memory
• Adaptive game NPCs
• Mood-aware recommendation
• Session context that should fade
• Experimental cognitive research

❌ Bad Fit

• Financial transactions
• Audit logs (must never forget)
• Simple key-value caching
• High-availability production data
• When “forgetting” would be a bug

The Bottom Line

If you want…	Use
Fast key-value cache	Redis
Graph traversal & queries	Neo4j
Semantic similarity search	Vector DB
Living, subjective memory	MindFry

---

Next: Understand Why Forgetting Matters — the philosophical foundation.