plants: hundreds of species in citadel genesis and batuka

animals: dozens of species in citadel genesis and batuka

fungi: dozens of species in batuka

system of tagging

  • abundance: yes, limited, trial, none, gone
  • supply: yes, later, wishlist, no
  • margin: high, mid, low, none
  • autonomy: support, staple, extra

species/all

species/research

sets

encoding

practical spec for encoding the botanical knowledge graph into cyber

one species = one particle

each of the 205 species pages in this graph:

  • has content: description, ecology, uses, observations, images
  • gets content-addressed via IPFS → CID
  • becomes a particle in Bostrom
  • can be cyberlinked to anything: other species, locations, compounds, observations

example: coffea arabica

particle: QmXk7f...  (IPFS CID of the species page)

cyberlinks from this particle:
  coffea arabica → "family" → Rubiaceae
  coffea arabica → "grows_at" → cv.land
  coffea arabica → "needs" → shade
  coffea arabica → "companion" → calliandra calothyrsus
  coffea arabica → "produces" → caffeine
  coffea arabica → "observed_by" → [neuron address]

observation cyberlinks

every time a neuron observes a species in the field:

neuron → "observed" → photo_cid
photo_cid → "depicts" → species_particle
photo_cid → "location" → gps_cid
photo_cid → "timestamp" → block_height

the observation is permanent, verifiable, and linked to the knowledge graph

what 205 species create

with ~10 cyberlinks per species (conservative):

  • 2050 cyberlinks encoding ecological relationships
  • a queryable biological knowledge graph inside Bostrom
  • rank computation reveals: which species is most connected (ecologically central), which location has highest biodiversity, which compounds appear across most species

search queries that become possible

  • "nitrogen fixing tree" → ranked list of species by relevance
  • "companion for coffea arabica" → species connected by "companion" cyberlinks
  • "medicinal fungi" → intersection of fungi particles and medicine cyberlinks
  • "what grows at 1500m elevation" → location-linked species subgraph

bulk encoding

the 205 species pages can be batch-uploaded:

for each .md file in pages/ where tags contain "species":
    cid = ipfs.add(file)
    cyberlink(neuron, cid)                    // "created" link
    cyberlink(cid, genus_cid, "belongs_to")   // taxonomy edge
    cyberlink(cid, location_cid, "found_at")  // geography edge

cost: ~600 cyberlink transactions. at current bandwidth rates, achievable with moderate CYB stake via investmint

from graph to protocol

this graph is a prototype. the species pages, the [[wiki-links]], the tags — they ARE a knowledge graph. the step from markdown to Bostrom is mechanical:

  • markdown page → IPFS CID → particle
  • [[wiki-link]]cyberlink
  • tag → typed edge
  • the graph is already built. it just needs to be committed to the protocol

Local Graph