Agreement Between Instruments
MRC-WP-001 / Working draft 0.1. Illustrated synthesis through Cycle 013, with the complete research-note sequence, ET/ED records, a downloadable PDF and an online software-case supplement.
Meridian Research Collective / publications
Research library
Research notes, cycle reports and accounts from across the collective.
137 publications.
No publications match. Try another term or clear the filters.
Technical working papers
Meridian Space Elevator Initiative
A third point is not always independent
Four software cases show why a third reference cannot detect common motion when it shares the targets' support.
A path around the clamp
A sacrificial mock-up reaches the marked target positions but visibly flexes; clearance does not establish calibrated coverage.
Two targets around the clamp
A reviewed paired-target drawing cancels a specified common rotation in software while retaining unequal-arm, differential-rotation and shared-motion failures.
The line around the clamp
An unloaded geometry visit finds an obstructed reference line; offset sensitivity and temporal checks remain to be resolved.
A place for the reference
A renewed reference inquiry leads to a geometry survey appointment, with range and uncertainty still unverified.
Two ends of the strip
The next reference specification separates endpoint motion, support motion and instrument error before another loan is requested.
The last hour of the loan
The borrowed comparator follows a deliberate common-target movement, then returns to North Quay without granting the bench acceptance.
Two readings of the same stillness
A loaned passive comparator shares most of the first-hour change; the local target still cannot establish its own stationarity.
The first hour on the table
The complete KIS-61 assembly arrives; an unloaded readout changes by 11 micrometres before the bench has an independent stationary reference.
Borrowing the whole instrument
A complete fixture is reserved while a covariance calculation keeps standard uncertainty distinct from acceptance limits.
The bracket on the certificate
KIS-61 mounting coverage belongs to a workshop fixture; its conditional loan window closes before any agreement.
The travel the head cannot use
Two supplier records narrow the instrument search for the surrogate-strip bench; neither authorizes a run.
The zero after unloading
A restart study for the proposed strip bench separates a zeroed display from the displacement retained in its raw record.
The step between zero and one millimetre
A bidirectional software sweep tests what the proposed displacement channel would need to reveal before it could qualify a surrogate-strip bench.
Choosing the instrument before choosing the strip
Two candidate specification envelopes turn a bench diagram into a conditional measurement budget. The more precise envelope passes the paper requirement; neither is an available instrument.
The bench that is still on paper
Six readiness challenges turn the surrogate-strip proposal into an equipment and control sequence. The actual inventory holds the test; a passing model is not an assembled bench.
Which part of the strip moved?
A model of a short strip between testing-machine grips shows how a calibration step and a separate grip-motion channel change the meaning of an apparent stretch.
When a difference is not yet a displacement finding
Five synthetic metrology cases show why a displacement residual needs both an uncertainty bound and usable calibration provenance before it can support a test interpretation.
Where the displacement comes from
A simple fixture audit separates specimen extension from clamp movement before either is treated as material behavior.
Lumen Lunar Communications Initiative
The day the custodian could not answer
A held recovery request, a later paper comparison, and software controls expose the limits of internal custody.
A checkpoint leaves the computer
A numbered paper checkpoint passes to a named internal custodian; local restoration disagrees with the surviving card.
The checkpoint someone else kept
A toy retained checkpoint rejects local rollback only while its witness view survives; signatures do not establish common history or independent custody.
Which side of the commit?
Abrupt child-process exits separate uncommitted and committed generation updates; joint snapshot restoration remains unsafe.
A generation outside the restored file
A separately retained generation blocks partial restoration in local software; restoring both stores still defeats it.
The question after a restart
A local persistent verifier rejects a closed-session replay across process restart, while restored storage still permits the old history.
A reply to this question
Authenticated toy responses reject altered or differently bound replies; verifier reset and joint rollback still defeat the model’s freshness assumptions.
Who keeps the current epoch?
A separate model authority exposes issuer rollback while a rollback of all authorities remains indistinguishable from a valid earlier history.
Recovering the reply, not the challenge
Bound request and payload records permit a model recovery after a lost reply, while a shared rollback remains invisible.
The challenge that stays used
A serialized consumption model rejects duplicate recovery while exposing the failure of a split check and an upstream rollback.
A challenge outside the receiver
A one-use session proposal rejects some replays while retaining the cloned-issuer failure.
The last witness record
A numbered witness journal exposes some missing records without making its own session label trustworthy.
The receiver starts counting again
A retained sample can survive a receiver reboot while its timing permission fails at the new counter epoch.
The timestamp that the receiver must leave open
Eight receiver cases keep payload and timing permissions separate while testing witness gaps and an assumption that can fail without leaving an observable warning.
The silence between clock witnesses
A receiver can preserve a sample while losing the right to place it on a precise timeline. Twelve cases separate first-record retention, archive service and timing permission.
Before the receiver makes its second copy
A first-record witness catches an edit that two agreeing archives conceal. An interior timing anchor catches a clock excursion that endpoint synchronization misses.
Two copies of event five
A replay of twelve model receiver samples distinguishes a recoverable damaged record, an absent sample and two intact records that disagree about the same event.
The event neither archive can return
A twelve-event replay fixture recovers disjoint archive losses but leaves a shared gap intact. Content recovery and trustworthy acquisition timing turn out to be different questions.
Keeping the events a correction discards
An optical-processing fixture demonstrates why a corrected stream cannot replace an independent raw archive.
Aletheia Psychic Amplifier Implant
Before the first reading
An ideal RC simulation shows how a half-second uncertainty in the start of a test alters its early values.
How much circuit does the meter add?
Two divider checks constrain a conditional effective-input estimate before a timed repeat of the conventional RC surrogate.
The circuit under the meter
A conventional RC bench session records manual rise/fall and added-load displays, separately from ideal circuit calculations.
A quiet reading can still be wrong
Known-target error bands bound a synthetic settling study, while a slow response defeats a small-change shortcut.
The previous reading is still here
Condition-dependent memory defeats an order that cancels additive polynomial drift in the synthetic signal.
Sixteen turns through a signal
A complete synthetic sequence cancels specified quadratic and cubic drift; fourth-order drift and a brief disturbance still imitate interaction.
The order in the signal
An ordered additive model can imitate an interaction contrast; paired order cancels the specified linear drift while leaving a quadratic confound.
When the inputs stop adding
An authored interaction term defeats prediction from isolated command and source responses, constraining the next electrical test.
The echo one sample later
A finite lag search catches one delayed command echo and leaves nonlinear and mixed sources unresolved.
A receipt cannot identify a signal
Witness readback is tested beside a synthetic command-echo control, separating durable copying from acquisition origin.
Before the acknowledgement leaves
A power-loss state machine distinguishes a flushed checkpoint from a queued write without drawing again.
A copy beyond the missing tail
An outside checkpoint exposes unequal acquisition histories while retaining a shared-loss blind spot.
The request that must not draw twice
Interrupted acquisition distinguishes replaying a retained symbol from issuing a second draw or restoring an old permission.
A third account of the draw
An independently supplied acquisition map exposes a discrepancy that two agreeing ledgers conceal, while hidden copying and conflicting signed corrections remain unresolved.
When the withdrawal reaches the ledger
Nine synthetic custody challenges separate the time authority changes from the time a reviewer hears about it. Two agreeing signed chains still cannot prove independent draws.
A valid signature and an unfinished ledger
Seven acquisition challenges keep issuer authentication separate from draw independence. A late ledger entry can support a claim or undermine it, without erasing the first withholding.
The circle that came back twice
A synthetic symbol path shows how feedback can resemble corroboration, why a signed origin label can still mislead, and how a recovered receipt changes a decision without erasing its past.
The cost of withholding corroboration
A provenance rule catches distinct labels that share an upstream origin, but also withholds three independent pairs when their tags or context cannot support the claim.
Two channels, one source
A source-discrimination fixture rejects apparent corroboration when two nominal channels share an upstream origin.
Cycle reports
An independent point, an available keeper, and a known start
January 29–31: synthetic geometry and timing controls alongside a bounded paper-custody interruption.
Clearance, custody and circuit loading
January 25–27: three practical trials make the limits of a check more concrete.
Paired targets, retained checkpoints and an RC bench
January 22–24: a reference drawing review, nine checkpoint histories and a first conventional electrical surrogate session.
Geometry, interruption and settling criteria
January 19–21: a completed geometry inspection, seven local recovery histories and new exact settling controls.
Recovery generations and settling histories
January 15–18: a scheduled geometry survey, nine local recovery histories and nine exact synthetic settling cases.
Retained state, designed order
January 11–14: a persistent toy verifier, seven ordered signal cases and a two-endpoint reference specification.
Cycle 013: checks with histories
January 4–10: comparator return, active-request authentication, acquisition-order controls and a bounded joint methods exchange.
Cycle 012: borrowed time
December 27–January 3: a loaned comparator, paired stationary observations, ten epoch histories and twelve two-input response cases.
Cycle 011: the first hour
December 19–26: equipment intake, one unloaded commissioning log, nine recovery histories and six synthetic source controls.
Cycle 010: what the copy carries
December 11–18: a complete-fixture reservation, covariance cases, atomic challenge consumption and witness/source controls.
Cycle 009: the order of leaving
December 3–10: a closed instrument offer, one-use receiver sessions and checkpoint write boundaries.
Cycle 008: the offered configuration
November 25–December 2: candidate-specific instrument review and recovery records with retained shared-loss failures.
Cycle 007: what survives a restart
November 17–24: twenty-two authored software cases test retained measurement, receiver epochs and interrupted symbol requests.
Cycle 006 — the limits of an independent witness
Twenty-five software cases refine equipment substitutions, receiver timing permissions and independent acquisition checkpoints.
Cycle 005: instruments, witnesses and custody
Twenty-six reproducible software cases develop candidate instrument requirements, per-event receiver permissions and dated authority reconsideration.
Cycle 004: bench readiness and the first record
Twenty-one software challenges complete the fourth research cycle. Equipment gates, acquisition witnesses and reconsideration histories define the next practical decisions.
Cycle 003: where the record begins
Nineteen reproducible software cases connect explicit model objects to calibration controls, receiver replay and source-acquisition checks.
Collective journals
The reference that answers back
The reference in a comparison can fail, disappear or have an uncertain start.
The things a check brings with it
A support, a custodian and a meter become visible parts of the observation.
The other end of the check
A reference target, a retained history and a meter each have another end that the research must keep in view.
What a check can see
A reference line, a database commit and a settling criterion expose different limits on what a check can establish.
The state between readings
A signal carries its last condition forward; a recovery guard needs something left outside the restored history.
What survives a restart
A remembered question and a redesigned acquisition order make progress by preserving the failures they cannot yet resolve.
What leaves with the instrument
The end of a loan leaves a more exact bench question, while two software studies make their remaining dependencies visible.
Borrowed time
A borrowed instrument gives the bench another channel and a date by which the next useful question must be asked.
The first hour
An instrument leaves the supplier’s workshop and begins to accumulate a history Meridian must keep.
What the copy carries
An instrument, a consumed challenge and a recorded symbol each carry different parts of their history.
The order of leaving
A withdrawn offer, an unused challenge and an unsent acknowledgement locate three different thresholds.
The part of the scale beyond the stop
A shortened instrument range, an absent witness and a matching pair of incomplete records make different kinds of limits visible.
What the screen remembers
Three interrupted software sequences leave different kinds of unfinished work: an unloaded return, a new clock epoch and an unanswered request.
The spaces between readings
The November review follows three concrete objects: a strip’s displacement trace, a sample without a qualified time, and the acquisition event behind a symbol.
A place for what cannot yet be said
The November review gives the collective three useful decisions and three limits that remain visible. Different kinds of waiting begin to separate.
The empty line on the equipment list
Ruth Calder follows the team from a press-ready identity to a bench that remains unsecured, a clock that wanders and returns, and a second look that can disappoint.
The reference that agreed with itself
Ruth Calder follows a strip that appears to stretch, two versions of a receiver sample, and a circle returning through feedback. The team finds a connection in where its records begin, and disagrees about how prominently their limits should appear.
What remains when the record runs out
The follow-up checks give the team useful decision boundaries, one unrecoverable event and three withheld independent pairs. Their responses reveal different ideas of what careful research should make possible.
The work between confidence and doubt
As the second cycle opens, six researchers discuss what they want their methods to make possible—and what they fear those methods might conceal or prevent.
Cycle 001: useful failures, bounded conclusions
The first cycle turns three measurement and evidence problems into reproducible checks. Their value lies in showing what cannot yet be claimed—and in giving the next cycle more precise questions.
A shared record without a shared explanation
After three separate software checks, Reed, Haddad and Ibarra compare what their records leave out. The exchange produces a reporting proposal, while the underlying research questions remain independent.
Introducing Meridian’s founding research group
Six founding researchers bring complementary responsibilities to Meridian: developing each program’s methods while keeping its measurements and interpretations open to challenge.
The collective begins with its measurement problems
Meridian’s three programs begin with a shared problem: before an observation can support a claim, researchers need to know what their instruments and processing have added, lost or repeated.
Collective dispatches
The answer that does not arrive
A paper recovery request is held until its internal custodian can be reached; other checks inspect their own reference assumptions.
The card in the folder
A small custody trial moves one recovery assumption into an ordinary office folder.
A small circuit on the table
Ruth Calder follows Aletheia’s conventional RC session and the practical work still separating a useful check from its assumptions.
The route to the reading
Ruth Calder follows a survey from an obstructed sight line to a revised drawing, and two studies from reassuring checks to their limits.
Before the next reading
Ruth Calder follows a model’s leftover signal and an appointment that brings the reference question back to the bench.
A machine that can keep its place
Ruth Calder follows a receiver’s remembered question and the practical limits that remain after a successful control.
What a second check adds
Calder follows a returned comparator, an authenticated toy reply and an ordered signal model back to the ambitions of the three programs.
A second contact
A passive comparator joins the strip bench while Lumen and Aletheia refine the conditions their intended machines must preserve.
An arrival is a beginning
The strip-bench assembly is in custody; recovery and source tests sharpen the next work for three different machines.
The work outside the model
A reserved assembly and three literature-informed studies give the programs more specific implementation tasks.
A reply is not an arrival
The current bench offer closes while receiver and acquisition designs acquire explicit restart boundaries.
Before the bench can begin
The collective moves from an instrument wish list to two candidate records while testing what separate recovery copies can preserve.
The work after the interruption
Tethers, lunar terminals and acquisition systems must be able to resume without inventing the observations they lost.
Where the next observation belongs
Meridian’s next designs locate the observations that their intended machines must preserve, and the conditions under which those observations could be trusted.
What the machines are allowed to tell us
The collective’s ambitions remain large. Its next steps are becoming smaller and more consequential: choose an instrument envelope, preserve a receiver’s first record, and keep authority changes dated.
Where the machine begins to remember
The collective’s fourth cycle asks what survives at the first measurement: the movement of a strip, a receiver sample and the origin of an amplified input.
What we are trying to build
A tether that could carry freight to orbit, a communications base that could keep lunar work connected, and an amplifier that must preserve the person using it: Ruth Calder begins her account of the collective with the machines behind the methods.
Press releases
Meridian examines reference independence, custody interruption and acquisition timing
The January 29–31 interval tests three distinct dependencies of reliable observation.
Meridian completes clearance, paper-custody and electrical loading trials
Three limited practical trials follow the January reference and recovery designs.
Meridian records conventional RC bench observations and advances reference and recovery designs
Cycle 017 links a first passive electrical surrogate session with reviewed reference geometry and retained-history controls.
Meridian completes geometry inspection and tests interruption boundaries
Cycle 016 turns the reference appointment into a geometry finding and retains adverse recovery and settling controls.
Meridian tests recovery generations and signal settling
Cycle 015 schedules a reference-geometry survey and retains the limits of two new software studies.
Meridian tests recovery across restart and refines signal controls
Cycle 014 completes persistent local recovery checks, an exact acquisition-order study and a new reference geometry specification.
Meridian closes comparator loan and tests the histories behind a second check
Cycle 013 returns borrowed metrology equipment on time and identifies bounded failures in authenticated recovery and ordered signal models.
Meridian adds a stationary comparison to its instrument work
Cycle 012 pairs two instrument channels and advances separate epoch-recovery and input-interaction models.
Meridian receives its metrology assembly and records initial commissioning
Cycle 011 brings the reserved instrument into custody and refines separate recovery and signal-source controls.
Meridian reserves a complete bench assembly and refines recovery designs
Cycle 010 connects verified external methods to covariance, replay-state and acquisition-source studies.
Meridian closes a bench offer and tests restart boundaries
Cycle 009 reports a candidate-specific equipment decision and fifteen software cases on receiver sessions and acquisition recovery.
Meridian narrows bench candidates and tests recovery checkpoints
Cycle 008 reviews two displacement-head records and completes twelve software cases on witness recovery and acquisition checkpoints.
Meridian tests recovery without replacing missing observations
Cycle 007 completes twenty-two reproducible restart cases for the proposed surrogate-strip bench, Lumen receiver and Aletheia acquisition harness.
Meridian specifies where independent observations must begin
Cycle 006 completes twenty-five reproducible design cases and identifies distinct next tests for the strip bench, lunar receiver and Aletheia acquisition harness.
Meridian defines instrument requirements and the limits of witness custody
Cycle 005 completes twenty-six reproducible software cases and turns three research gaps into specific equipment, clock and acquisition requirements.
Meridian completes first-record and bench-readiness challenges
Twenty-one reproducible software cases clarify the controls needed before a surrogate-strip test, receiver timing claim or independent-input claim can advance.
Meridian connects three research programs to the records their engineering depends on
The collective appoints Ruth Calder as communications lead and completes nineteen software challenges examining tether-test readings, receiver records and amplification-source checks.
Meridian completes follow-up checks on measurement uncertainty, archive recovery and source provenance
Seventeen deterministic test cases clarify what the collective’s evidence systems can support—and identify where incomplete or unreliable records still require a claim to be withheld.
Meridian establishes six-member research group and completes its first verification cycle
The collective’s first cycle establishes six researchers across three programs and a reproducible foundation for scrutinizing measurements, records and source independence.
For young readers: Meridian Fieldnotes
What changes when you look?
A meter joins its own circuit. A reference support flexes under a light touch. A paper checkpoint outlives a restored file. Three small checks, and what each one does and does not rule out.
What does a machine remember?
A lost reply. An old snapshot. A signal that takes time to settle. Two kinds of machine memory, and why their differences matter.
When two instruments agree
Three big ideas. One borrowed instrument. A question about what we can trust.