What it does
EPPE takes public utility infrastructure filings and readings and anchors their hashes, so that a figure published today can be shown, later, to be the same figure that was published today. It runs on CCP.
The problem it addresses is narrow and real: in a rate case or a permitting fight, the party holding the records is a party to the dispute. Anchoring does not make the data true. It makes silent revision detectable.
In the code
# Anchor a commitment, never the payload. The document stays wherever it
# lawfully lives; the chain carries only enough to prove it has not moved.
commitment = sha256(canonical_json(record)).hexdigest()
anchor(commitment, source_url=record.source, observed_at=record.fetched_at)Illustrative shape. Canonicalization before hashing is the part that is easy to get wrong: two byte-different serializations of the same record must produce the same commitment or the whole exercise proves nothing.
How this differs from the ordinary version
Civic, not commercial
The instrument exists because El Paso is deciding large questions about data centers, water and grid capacity with numbers supplied by the parties who benefit from them. It is a public-interest tool, and the research page on data centers is the argument it belongs to.
In the field
Data centers and the El Paso grid
The regional data-center buildout is a live public question about power and water. Whatever the answer turns out to be, it should be argued from figures that cannot quietly change between the hearing and the record.
Questions
- Does this make the underlying data accurate?
- No. It makes revision detectable. Accuracy is a separate problem and the instrument does not claim to solve it.