01 / PROOF LABTelemetry
LOGVERITY
Telemetry economics and evidence control
Measure eligible work, defensible savings and records that must remain preserved.
INDEPENDENT RESEARCH · ENGINEERING & TECHNICAL ASSURANCE
Engineering claims that survive scrutiny.
A research portfolio in deterministic execution, integrity-preserving state, operational efficiency and verifiable assurance. Explore the public evidence boundary first. Request deeper technical material under controlled disclosure.
Research identity: Loris Longarini · Italy · Independent asset owner. No implied institutional affiliation, customer deployment or third-party certification.
01 / INDUSTRIAL PROBLEMS
Start with the problem, not the logo.
Three questions connect the public research to measurable operational value.
01 / COST
Which work is worth paying for?
Eligible telemetry, protection of evidence and actual financial impact.
LOGVERITY ↗02 / LOADWhich transformations can be avoided safely?
Exact-output reuse with the event stream preserved.
T.A.R.T.A.R.U.S. ↗03 / INTEGRITYWhat must remain true under failure?
Duplicate actions, stale generations, replay and explicitly unknown outcomes.
T.H.O.R.A.N. ↗02 / PUBLIC LABORATORY
Thirteen ways to inspect the questions.
Six operational Proof Labs and seven assurance profiles. Each link leads to a public demonstration with a deliberately bounded claim.
02 / PROOF LABAdmission
T.A.R.T.A.R.U.S.
Stream work avoidance and exact-output reuse
Preserve each event while avoiding repeated transformations when the output can be reused safely.
03 / PROOF LABState
T.H.O.R.A.N.
State fencing and replay integrity
Guard duplicate actions, stale generations and ambiguous outcomes with explicit decision states.
04 / PROOF LABMemory
C.E.R.B.E.R.U.S.
Memory and allocation control
Expose allocation pressure and cost rather than hiding it behind average throughput.
05 / PROOF LABData path
C.Y.C.L.O.P.S.
Serialization and copy accounting
Locate the data movement and encoding operations that add hidden pipeline work.
06 / PROOF LABRecovery
N.E.M.E.S.I.S.
Replay and generation guard
Model stale replay, repeated events and generation mismatches at a guarded boundary.
ASSURANCE SERIES / Ω
Explore the Ω series Seven additional boundary-focused demonstrations.
Timing, sensing, filtering, invariants, correlation, industrial I/O and agreement.
03 / UNDERLYING WORK
Two independent portfolios. Forty-eight research identities.
The public demos are an entry point. The asset portfolio is broader, organised by source lineage, purpose and disclosure scope.
PORTFOLIO A / SOVEREIGN
30
Execution, governance & system integrity
Deterministic execution, state, orchestration, evidence and specialised research systems.
Explore Sovereign →PORTFOLIO B / FRONTIER18
Verification, isolation & trusted interfaces
FZ / FF / N5 lineages covering verifiability, controlled boundaries and recovery.
Explore Frontier →04 / RESEARCH PUBLICATIONS
Four essays. No empty archive.
Long-form technical writing on the questions that guide the engineering. These are research notes, not peer-reviewed journal articles.
Research note / 01 · 09 NOV 202601
The Economics of Unnecessary Work
Why eliminating invisible processing is useful only when correctness, auditability and the customer's invoice agree.
Read article →Research note / 02 · 09 NOV 202602State Integrity Under Failure
Recovery is meaningful only when authority, transitions and evidence remain intelligible after interruption.
Read article →Research note / 03 · 09 NOV 202603From Nanosecond Decisions to Physical Evidence
Decision logic, network paths and external hardware timing are separate measurement domains.
Read article →Research note / 04 · 09 NOV 202604An Evidence-First Research Portfolio
How to make 48 research assets legible without pretending that 48 assets share the same maturity.
Read article →EVIDENCE RELEASES / OWNER-APPROVED REGISTER
Latest scoped results.
Published findings update from a read-only release register, independently of the website deployment. Each entry retains its source date, measurement boundary and limitations. Private raw artifacts remain with the researcher.
Hardware RX/TX timestamp pairs: 64/64
p50 136,837 ns · p99 163,625 ns. Owner-reported server-side NIC timing interval; restoration PASS. Not sub-microsecond, not a full two-host end-to-end latency result.
Request original provenance ↗Read-only identity check
Authenticated access, JTAG IDCODE match and SHA-256 precheck reported PASS; no new pin-to-pin timing result established by this check.
13 demonstrations and bounded reference tests
13/13 pages and 149/149 resources checked in R10; static R10 documentation remains available for exact historical provenance.
R10 public dossier ↗Last publicly reviewed: 09 November 2026. The register updates after a source-bound result is reviewed and explicitly released; an unreviewed test is never auto-published.
05 / DILIGENCE
Public proof first. Deeper evidence by request.
Reproduction-sensitive architecture stays private; the boundaries of a claim should not.
01
Inspect
Use the public demonstrations and read the relevant evidence scope.
02
Define
Specify workload, version, acceptance criteria and rights.
03
Validate
Arrange a scoped technical evaluation or funded qualification when needed.
04
Decide
Consider licensing, collaboration or no fit against measured value.
RESEARCH UPDATES / OCCASIONAL DISPATCH
Follow the evidence, not the noise.
Get new research notes and carefully scoped qualification updates. No automatic exposure of private documentation.