Algorithm
A fair draw, with nothing hidden
ScirDom combines a tested source of randomness with a fixed, published selection method to ensure that every draw is conducted fairly and can be independently checked.
The randomness
The randomness used in each draw comes from the active entropy block: a large block of certified random data generated by Eormen before being made available to the platform.
Every block is tested by three independent statistical test suites and must pass all three completely before it can be used. Those test results, along with the block identity and verification values, are published on the Transparency page for anyone to inspect.
The exact portion of the block used in each draw is recorded in the evidence pack, and the promoter receives a copy of the exact bytes consumed for that draw. This means the source of randomness for any given result can always be identified, traced, and independently checked.
The selection method
The method used to select winners is fixed, documented, and published. As the method is fully defined, any third party who has the locked participant register and the randomness used in a specific draw can independently reproduce and verify that result.
Every participant has an equal chance of being selected. Think of it as a physical draw: every name on an identical slip of paper, all placed into a bag, with the required number drawn out blind. Every possible combination of winners is equally likely to be drawn. ScirDom's method replicates this precisely, with a published mathematical proof to confirm it.
One additional safeguard is built in. When converting random data into a position in the participant list, a subtle numerical bias can arise if the conversion is not handled carefully. ScirDom's method identifies and discards any such conversion automatically, repeating until a valid, unbiased result is produced.
The Algorithm Specification Document (ASD)
The selection method is documented in full in the Algorithm Specification Document (ASD). The public specification is provided as a branded PDF rather than a raw source file, so the version a user downloads is the same formal publication referenced in evidence packs and verification records.
The document is written to be followed, not just read. Given the locked participant register and the exact bytes used in a draw, anyone can work through the steps and confirm that the winners shown are correct. Reference code is published alongside the document for reviewers who prefer to verify the result by running the method directly.
A separate public guide, ScirDom_How_To_Verify_A_Draw_v1_0.pdf, explains the independent replay process in plain language and is also included in every final evidence pack.
ASD v1.0 is the current published definition used by ScirDom. While the platform is still being finalised before Phase 7 go-live, corrections may still be issued to this version. Once ScirDom is live, any behavioural change will require a new ASD version so historical draws remain tied to the exact version recorded in their evidence pack.
ASD v1.0 – Public release bundle
These six files define the public ASD release bundle. They support public conformance checking of the published reference implementations, published test vectors, and offline evidence-pack replay. For a specific completed draw, the draw evidence package remains the replay artefact and supplies the draw-specific entropy bytes.
Ready to run a fair draw?
Every draw on ScirDom is backed by certified randomness, a published algorithm, and a sealed evidence pack.