My new paper: what a permissioned ledger can and cannot prove
DEV Community

My new paper: what a permissioned ledger can and cannot prove

Today I published a research paper: "What a Permissioned Ledger Can and Cannot Prove About Business Documents." It is open access on Zenodo, the research archive run by CERN, with a permanent DOI. Read it here: doi.org/10.5281/zenodo.24438467

Why I wrote it

"It's on the blockchain" is one of the most overused sentences in enterprise technology. It gets treated as if it meant "it's true", "it's legally binding" or "nobody can dispute it".

None of those follow from recording a fingerprint on a ledger. But the narrower things a ledger does prove are genuinely valuable, and they get lost once the hype is stripped away. I wanted to write down, precisely, where the line is.

What a ledger proves

The paper states five guarantees, each with the condition it depends on:

  • Integrity: the document hasn't changed since it was anchored.
  • Existence by a time: it existed in this exact form no later than a given moment.
  • Attribution: a specific account approved it.
  • Ordering: which approval came before which.
  • A shared view: every party checks against the same record.

What it doesn't prove

And seven things it does not establish: the truth of the content, the person behind an account, anything before anchoring, completeness, privacy for guessable documents, legal validity on its own, and the availability of the file itself.

Six design rules

From those gaps come six practical rules: anchor at the moment of approval, check permissions before anything is signed and keep keys out of apps, salt fingerprints of guessable documents, treat document storage as its own system, use eIDAS qualified signatures where legal weight is needed, and tell users plainly what was proved.

The paper walks through a full multi-party document lifecycle, from purchase order to delivery confirmation, correction and invoice, and ends with a simple conclusion: a ledger proves less than people claim, but what it proves is precise and worth having.

Links

  • Paper (DOI): doi.org/10.5281/zenodo.24438467
  • Zenodo record: zenodo.org/records/24438467
  • ORCID: orcid.org/0009-0009-3536-0667
  • More of my work: pratyushshivam.com

I'm Pratyush Shivam, CTO of Konfidio in Berlin. If you build or buy systems on permissioned ledgers, I'd like to hear where you think the line sits.

Read on DEV Community ↗ ← Back to News

Comments

No comments yet. Start the discussion.