Skip to content

GCRI · Foundations

Stand-down, archive and renewal: making a temporary build cumulative

The annual programme assembles temporary technical infrastructure for a week of controlled exercises. This page explains how the design ends it cleanly, keeps its evidence, and turns what was learned into the next cycle.

As of
21 October 2026
Reading
13 minutes
Evidence
16 sources, cited in order
NEXUS TWIN · WF-1 · NETWORKSEED c1e075d1GCRI-MAIN.POST.1076870xyzSCHEMATIC · NETWORKST-01 · 0.215 0.489ST-02 · 0.435 0.307ST-03 · 0.299 0.314
WF-1 · network Schematic wireframe twin (network, information diffusion). An illustration, not data.

Methods and evidence

Evidence
16 sources, cited in order and listed under References
Figures
5 figures
Data availability
Every chart's data is published as a table on this page.
Stand-down
1 day
proposed to stop work, revoke access, seal evidence and freeze the manifest1
Decay
38%
of webpages that existed in 2013 were no longer accessible a decade later4
Gates
G5 to G7
evidence sealed, publication authorised, renewal reviewed1
Status
R0
information and enquiries; no cycle has run5

01What a stand-down is

What a stand-down is

The stand-down is the designed end of the temporary technical environment that the annual programme assembles for its week of controlled exercises. In the programme year of the Nexus Consortium, the public-good programme through which The Global Centre for Risk and Innovation (GCRI), The Global Risks Forum (GRF) and The Global Risks Alliance (GRA) help countries examine interconnected risk with shared methods, the rule is short: prepared for a year, built in a month, tested in a week and stood down in a day, then reviewed and renewed 1. That week is Nexus Universe, in the third quarter, and the annual programme cycle and the Nexus cycle set out where it sits.

It is the opposite of a switch-off. Each pillar has a defined task on the day: Nexus Universe stops new work and checkpoints authorised jobs; the Foundry pillar revokes temporary access and releases temporary capacity; Nexus Standards seals transferable evidence with its provenance, recorded in a lineage graph designed to follow the W3C PROV-O ontology for representing and interchanging provenance 6,7; Reports freezes the evidence manifest it will draw on; and Risk Academy closes exercise assessments and preserves their evidence 1. Its outputs are a stand-down receipt and an evidence manifest 1.

The earlier version of this page called this "teardown" and described it as an operating discipline already in use. It is a design. The one-day stand-down is still marked as proposed in the programme spine, and no cycle has yet run 1.

02Why temporary infrastructure must end on purpose

Why temporary infrastructure must end on purpose

Temporary systems are not safe because they are temporary. An account created for an exercise keeps its permissions until someone removes them; a workload left running keeps consuming capacity and keeps its access to data; a display left online keeps being read long after its context has gone. The Network and Core page describes the stand-down of the temporary Core in exactly these terms: it closes new submissions, checkpoints authorised jobs, freezes the evidence manifest, revokes temporary credentials and records open exceptions 2.

The opposite failure is just as costly: evidence that disappears with the infrastructure that held it. The Pew Research Center found that a quarter of all webpages that existed at some point between 2013 and 2023 were no longer accessible by October 2023, and that 38% of pages that existed in 2013 had gone, against 8% of pages from 2023 4. Links decay too: 23% of news webpages and 21% of government webpages contained at least one broken link, and 54% of Wikipedia pages had at least one dead link in their references 4.

A cycle that ends badly therefore loses twice: it leaves exposure behind and takes its lessons away. The design aims at the reverse, so that temporary systems end cleanly and their evidence does not end with them.

Fig. 01One day to stop, revoke, seal and freeze

One day to stop, revoke, seal and freeze Five cards for the stand-down by pillar: Universe stops new work and checkpoints jobs; Foundry, highlighted, revokes temporary access and releases capacity; Standards seals transferable evidence; Reports freezes the evidence manifest; Academy closes assessments. Below: no promise that data is deleted within a day. 01Stop new work02Revoke and release03Seal04Freeze manifest05Close assessments
A designed day; the one-day stand-down is marked as proposed and no cycle has run.1,2Sources: The programme year; GCRI, Network and Core. As of 21 October 2026.
Figure 1 in words

Five cards joined by arrows. Universe, stop new work: new submissions close and authorised jobs are checkpointed. Foundry, highlighted, revoke and release: temporary access revoked and temporary capacity released. Standards, seal what transfers: transferable evidence sealed with its provenance. Reports, freeze the manifest: the evidence manifest the editions will draw on is frozen. Academy, close assessments: exercise assessments closed and their evidence preserved. A panel below: no promise is made that data is deleted within a day; media are purged or retained under approved, validated rules; open exceptions are recorded with accountable owners, and review continues after stand-down.

03Revoke, release, seal, freeze

Revoke, release, seal, freeze

The order of the day matters. Work stops before access is revoked, so that authorised jobs can be checkpointed rather than killed. Access is revoked before evidence is sealed, so that nothing changes the evidence after it is fixed. And the manifest is frozen before anyone writes about the exercise, so that every later edition draws on the same record 1,2.

Ending is not the same as deleting. No promise is made that data is deleted within a day: media are purged or retained under approved, validated rules, and review continues after stand-down 2. That is the posture of NIST SP 800-88 Revision 2, published in September 2025, which supersedes the 2014 revision and helps organisations set up a media sanitization programme with techniques and controls matched to the sensitivity of their information 3.

Open problems are recorded, not hidden. At closeout, Nexus Universe records every unresolved exception with an accountable owner, and the Foundry pillar reconciles equipment, reservations and costs and verifies the services that persist 1. The evidence is sealed at gate G5, "evidence sealed and stand-down recorded", which closes the closeout phase 1.

Fig. 04Temporary systems end; their evidence must not

Temporary systems end; their evidence must not Two columns. Ends at stand-down: temporary credentials, temporary capacity, new submissions, open assessments, and equipment and reservations reconciled at closeout. Persists, highlighted: the evidence manifest and seal, the exception register, identifiers and versions, completion records, and change proposals and gaps.
Designed; no cycle has run.1,8,2,9Sources: The programme year; Identifiers and records; Network and Core; NXS-POR (public, client and partner delivery). As of 21 October 2026.
Figure 4 in words

Left column, ends at stand-down: temporary credentials, revoked; temporary capacity, released; new submissions, closed with authorised jobs checkpointed; open assessments, closed with their evidence preserved; equipment and reservations, reconciled at closeout with their costs. Right column, highlighted, persists after it: the evidence manifest and seal, frozen at stand-down and sealed at G5 with provenance; the exception register, with an accountable owner for every unresolved exception; identifiers and versions, with each UNI still resolving and corrections as new versions; completion records for assessed participants, separate from any credential; change proposals and gaps, carried into the docket and the next mandate.

PillarStand-downCloseout (G5)
Nexus UniverseStops new work; checkpoints authorised jobsRecords unresolved exceptions with accountable owners 1
FoundryRevokes temporary access; releases temporary capacityReconciles equipment, reservations and costs; verifies persistent services 1
Nexus StandardsSeals transferable evidence with its provenanceRecords sealed evidence against the standards it was tested on 1
ReportsFreezes the evidence manifest it will draw onReceives the sealed evidence for the Q3 edition 1
Risk AcademyCloses exercise assessments; preserves the evidenceWrites completion records for assessed participants 1
The stand-down and closeout by pillar, as designed in the programme year on 21 October 2026. The one-day stand-down is marked as proposed.

04An archive designed to outlast the build

An archive designed to outlast the build

What persists must be findable and readable years later, by people who were not there. Every object of the programme year, a cycle, a mission, an edition, a course, a participation, carries one identifier, a UNI under the proposed Nexus Standard NS-1, so that it can be cited, versioned and resolved; a correction to an edition is a new version with a recorded reason 8. A record moves between pillars only through a recorded handoff, and nothing crosses unrecorded 8.

The archive is designed against an established reference model. The Reference Model for an Open Archival Information System, CCSDS 650.0-M-3 of December 2024, addresses the full range of preservation functions, ingest, archival storage, data management, access and dissemination, as well as migration of digital information to new media and forms; it requires cooperating archives to agree at least one common submission and dissemination package 10. The programme cites it as a reference for its evidence archive, not as a claim of conformance.

Corrections are part of the archive, not a threat to it. NXS-POR (public, client and partner delivery) preserves the withdrawal, replacement and recipient notification history of each publication, and NXS-GRX (canonical data, provenance and risk) propagates a source correction to dependent products without rewriting old results 9,11. The companion page on evidence records and archive sets out the record model in detail.

Fig. 03Evidence that is only online does not stay online

Evidence that is only online does not stay online Left, share of webpages no longer accessible in October 2023: 38% of pages from 2013, about 25% of all pages from 2013 to 2023, 8% of pages from 2023. Right, pages with at least one broken link: Wikipedia references 54%, news websites 23%, government websites 21%. Pages from 201338 per centAll pages 2013 to 202325 per centPages from 20238 per cent
Two panels on separate bases: pages by year, and pages containing at least one broken link.4Sources: Pew Research Center (2024). As of 21 October 2026.
Figure 3 in words, and its data

Left panel, share of pages that existed in a given year and were no longer accessible as of October 2023: pages from 2013, 38 per cent, highlighted; all pages from 2013 to 2023, about 25 per cent; pages from 2023, 8 per cent. Right panel, share of pages in each collection with at least one broken link, spring 2023: Wikipedia references section, 54 per cent, highlighted; news websites, 23 per cent; government websites, 21 per cent.

MeasureValueScope
Pages from 2013 no longer accessible38% 4Checked October 2023 4
Pages from 2023 no longer accessible8% 4Checked October 2023 4
All pages from 2013 to 2023 no longer accessibleAbout a quarter 4Checked October 2023 4
News webpages with at least one broken link23% 4Spring 2023 4
Government webpages with at least one broken link21% 4Spring 2023 4
Wikipedia pages with a dead reference link54% 4Spring 2023 4
Data of Figure 3, from Pew Research Center, When Online Content Disappears (17 May 2024).

05Closing out what others contributed

Closing out what others contributed

Much of a temporary build belongs to someone else: computing time offered by a centre, network capacity from a research network, a venue, an engineer’s week. The ending has to be clear for them too. Offering O09, Annual Universe Missions and Contributed Infrastructure, lists a technical closeout and an improvement backlog among its deliverables, beside the mission contract, resource allocation, Core Build evidence and exercise evidence, and its boundary states that neither membership nor sponsorship guarantees a compute allocation 5.

The routes for contributors set the same expectations. For HPC centres, research and education networks, telecoms, cloud and equipment contributors, the first outcome is a resource-offer disposition, a qualification record and separately approved allocations, and offered capacity is never counted as reserved, consumed or funded 5. For hosts, venues and campuses, the first outcome is a host readiness decision, a responsibility schedule and a closeout record 5. A contributor should be able to see, after stand-down, what was used, what was released and what remains open.

The same applies to people. Risk Academy writes completion records for assessed participants at closeout, and a learning record is not a professional licence or a promise of employment 1,5. What a participant did in the cycle stays on record for the next one, separate from any credential.

06From Action Week to renewal

From Action Week to renewal

The stand-down is one day in a longer ending. After Action Week, which closes at gate G4, come the stand-down, closeout at G5, the after-action review, reporting with publication authorised at G6, correction, docket review, network review and renewal at G7 1. Each phase has its outputs: the after-action report and change proposals; the annual synthesis with country and regional profiles; correction records and new edition versions; docket routing and a capability gap register; measured capacity results; and the renewal decision with the next-cycle mandate 1.

GRF records every gate. A gate is a programme decision about the cycle, never a public-authority approval, a certification or a deployment authorisation 1. The competent authority sits outside the programme throughout and decides and acts under its own rules 1.

Fig. 02Ending a cycle takes longer than the day it stops

Ending a cycle takes longer than the day it stops Timeline of eight phases from Action Week at gate G4 through stand-down, closeout at G5 highlighted, after-action review, reporting at G6, correction, docket review and renewal at G7, each with its outputs, grouped under the third and fourth quarters.
Gates G4 to G7 are programme decisions recorded by GRF; none is an approval or a certification.1Sources: The programme year. As of 21 October 2026.
Figure 2 in words

A horizontal timeline grouped under Q3, stress test and readiness, and Q4, annual synthesis and renewal. Action Week, gate G4: run, decision and observation records. Stand-down: stand-down receipt and evidence manifest. Closeout, gate G5, highlighted: evidence seal, exception register and completion records. After-action review: after-action report and change proposals. Reporting, gate G6: annual synthesis with country and regional profiles. Correction: correction records and new edition versions. Docket review: capability gap register. Renewal, gate G7: renewal decision and next-cycle mandate.

07Learning that changes the next cycle

Learning that changes the next cycle

An after-action review that changes nothing is a ceremony. The WHO guidance for after action review, published on 1 April 2019, presents AAR both as a review of actions taken in response to an event and as a routine management tool for continuous learning and improvement, and describes four formats: the debrief, the working group, the key informant interview and the mixed-method review 12.

The programme year makes the learning travel through records. In the after-action review, Nexus Universe reviews what worked, what failed and what was not tested, preserving dissent; Nexus Standards turns findings about methods and gaps into change proposals; and Risk Academy turns findings into course material 1. In correction, the Foundry pillar re-runs affected analysis with corrected inputs and Reports publishes a new version with the correction and its reason. In docket review, capability gaps are registered as work for the next build and open questions are listed for the next first-quarter edition 1.

Renewal closes the loop. At G7 the next cycle’s mandate is set, open projects and each country’s National Portfolio are carried forward, and the next first-quarter edition starts from the fourth-quarter synthesis 1. Measured results, not device counts, inform the network review, so that the next build is sized on what was observed 1.

Fig. 05How one cycle becomes the start of the next

How one cycle becomes the start of the next Circular diagram of seven steps: after-action review with dissent preserved, change proposals, correction as a new version, docket review registering capability gaps, network review on measured results, renewal at gate G7 highlighted, and the next first-quarter edition starting from the fourth-quarter synthesis. 01After-action review02Change proposals03Correction04Docket review05Network review06Renewal (G7)07Next Q1Next cycle
A designed loop; no cycle has run.1Sources: The programme year. As of 21 October 2026.
Figure 5 in words

A dial of seven arcs, clockwise from the top. After-action review: what worked, failed or was not tested, with dissent preserved. Change proposals: findings about methods and gaps become proposals. Correction: affected analysis re-run and a new version with its reason. Docket review: capability gaps registered as work for the next build. Network review: measured capacity results, since device counts are not results. Renewal, gate G7, highlighted: next-cycle mandate and open projects carried forward. Next Q1: the new edition series starts from the Q4 synthesis.

08How the evidence on this page was read

How the evidence on this page was read

The digital decay figures are from a Pew Research Center study of a representative sample of webpages from the Common Crawl repository for each year from 2013 to 2023, checked for accessibility in October 2023, and of the links on government, news and Wikipedia pages collected in spring 2023 4. Figure 3 keeps the two measures in separate panels because they have different bases: pages by year, and pages containing at least one broken link.

The instruments, NIST SP 800-88 Revision 2, CCSDS 650.0-M-3 and the WHO AAR guidance, are cited from their issuers’ own pages 3,10,12. The programme phases and outputs are designs taken from the canonical programme year; no cycle has run, and no stand-down has taken place 1,5.

09Instruments and proposed standards

Instruments and proposed standards

The ending of a cycle follows instruments that already govern each part of it, cited as references, not as claims of conformance: NIST SP 800-88 Revision 2 for sanitization, the OAIS reference model for archives, and the WHO guidance for after action review 3,10,12. The WHO Simulation Exercise Manual describes simulation exercises and after action reviews together as the functional assessment of capacities 13.

The programme’s own standards are proposed. NS-1 identifies every record of the cycle, and Nexus Standards records sealed evidence against the standards it was tested on, turns findings into change proposals and publishes the year’s change proposals at renewal 8,1,14. Statuses such as proposed or open describe drafting stages only; none records adoption or certification.

10What is changing

What is changing

  • Sanitization as a programme. NIST’s 2025 revision treats media sanitization as a programme matched to data sensitivity rather than a single wipe, which suits temporary environments that mix public and restricted work 3.
  • Preservation against link rot. With a quarter of pages from the last decade gone, evidence records rely on resolvable identifiers and archived copies rather than on live links alone 4,8.
  • Archives that cooperate. The 2024 edition of the OAIS reference model keeps the requirement that cooperating archives share common submission and dissemination packages, a basis for exchanging evidence between national programmes 10.
  • After action review as routine. WHO presents AAR as a routine management tool, not only a post-crisis exercise, which matches a programme that ends every year 12.

11Three institutions at the end of a cycle

Three institutions at the end of a cycle

GRF records the gates, including G5 at closeout, G6 for publication and G7 for renewal, in the manner of exercise direction, apart from controllers and evaluators 1,15. GCRI runs the technical office, including closeout, and the Foundry pillar, where the build is stood down and its evidence sealed; it leads on science, methods and evidence 1. GRA reads the year’s findings for the finance and insurance chapter of the flagship and runs the finance-readiness track; it does not lend, underwrite, rate or invest 1,16. Each institution sets out its own role on its site: The Global Centre for Risk and Innovation, The Global Risks Forum and The Global Risks Alliance.

The pages on Nexus Core, live operations and GRF’s evidence and outcomes cover the build, the week and what is published from it.

What this page does not claim: no cycle has run, no stand-down has taken place, the one-day stand-down is marked as proposed, and the programme is at service level R0, information and enquiries. No promise is made that data is deleted within a day. Gates are programme decisions, never public-authority approvals, certifications or deployment authorisations. GCRI does not regulate or certify, GRF does not regulate, finance or procure, and GRA does not lend, underwrite, rate or invest. Naming an agency, instrument or standard is not a claim of partnership, conformance or endorsement.

Questions

Questions and answers

Does the stand-down delete all data at the end of the exercise?

No. No promise is made that data is deleted within a day. Temporary credentials are revoked and the evidence manifest is frozen; media are then purged or retained under approved, validated rules, and review continues after stand-down.

What happens to problems found during the exercise?

They are recorded, not dropped. At closeout every unresolved exception is recorded with an accountable owner; after-action findings become change proposals, corrections become new versions, and capability gaps are registered as work for the next build.

How long does the end of a cycle take?

The stand-down itself is designed to take one day, and is still marked as proposed. Closeout, after-action review, reporting, correction, docket review, network review and renewal follow over the third and fourth quarters.

Will links to this year’s evidence still work in ten years?

That is the aim of the design. Every object carries a UNI that resolves to its record, a withdrawn record keeps its identifier and says what happened, and the archive is designed against the OAIS reference model.

Has a cycle been stood down yet?

No. No cycle has run, and the programme is at service level R0, information and enquiries. This page describes the designed end of a cycle, not an operating record.

Terms

Glossary

Stand-down
The proposed one-day end of the temporary build: new work stops, temporary access is revoked, capacity released, evidence sealed and the manifest frozen.
Evidence manifest
The frozen list of evidence that the year’s editions will draw on, fixed at stand-down.
Closeout (G5)
The phase that seals the evidence and records the stand-down, with an exception register and completion records.
After action review
A structured review of what worked, what failed and what was not tested, used for continuous learning.
Media sanitization
A process that renders access to target data on media infeasible for a given level of effort.

Reuse and cite

Text CC BY 4.0. Cite the page with its date; data tables are linked under each figure.

Sources

References

16 sources, numbered in the order they are first cited
  1. Nexus documentation (docs.therisk.global) (2026-10). The programme year. docs.therisk.global. Accessed 2026-10-21.
  2. The Global Centre for Risk and Innovation (2026-10). Network and Core. therisk.global. Accessed 2026-10-21.
  3. US National Institute of Standards and Technology (NIST) (2025-09). SP 800-88 Rev. 2, Guidelines for Media Sanitization. csrc.nist.gov. Accessed 2026-10-21.
  4. Pew Research Center (2024-05-17). When Online Content Disappears: 38% of webpages that existed in 2013 are no longer accessible a decade later. www.pewresearch.org. Accessed 2026-10-21.
  5. Nexus documentation (docs.therisk.global) (2026-10). Status and service levels. docs.therisk.global. Accessed 2026-10-21.
  6. World Wide Web Consortium (W3C) (2013-04-30). PROV-O: The PROV Ontology, W3C Recommendation. www.w3.org. Accessed 2026-10-21.
  7. The Global Centre for Risk and Innovation (2026-10). Provenance and attribution. therisk.global. Accessed 2026-10-21.
  8. Nexus documentation (docs.therisk.global) (2026-10). Identifiers and records. docs.therisk.global. Accessed 2026-10-21.
  9. Nexus documentation (docs.therisk.global) (2026-10). NXS-POR: public, client and partner delivery. docs.therisk.global. Accessed 2026-10-21.
  10. Consultative Committee for Space Data Systems (CCSDS) (2024-12). CCSDS 650.0-M-3, Reference Model for an Open Archival Information System (OAIS), Magenta Book. public.ccsds.org. Accessed 2026-10-21.
  11. Nexus documentation (docs.therisk.global) (2026-10). NXS-GRX: canonical data, provenance and risk. docs.therisk.global. Accessed 2026-10-21.
  12. World Health Organization (WHO) (2019-04-01). Guidance for after action review (AAR) (WHO/WHE/CPI/2019.4). www.who.int. Accessed 2026-10-21.
  13. World Health Organization (WHO) (2017-02-01). WHO Simulation Exercise Manual (WHO/WHE/CPI/2017.10). www.who.int. Accessed 2026-10-21.
  14. Nexus Standards (nexustandards.com) (2026-10). Nexus Standards: foundation standards NS-0 to NS-16 and the Nexus Risk Standards. nexustandards.com. Accessed 2026-10-21.
  15. Nexus documentation (docs.therisk.global) (2026-10). Nexus Universe, the programme pillar. docs.therisk.global. Accessed 2026-10-21.
  16. The Global Risks Alliance (2026-10). Consortium finance interface. globalriskalliance.com. Accessed 2026-10-21.

Record

Editorial record

Facts and programme states are given as of 21 October 2026. This page is reviewed at least every 180 days and whenever a programme state, a service level or a cited figure changes. It replaces an earlier version of the same post, titled "Teardown, Archive, and Next-Cycle Improvement", which described teardown protocols and platform names that are now retired as if in use; no cycle is on record, and the end of a cycle is described here as the programme year designs it.

Published by the GCRI editorial desk under the scientific and editorial integrity policy. To report an error or request a correction, use the corrections and concerns route; every accepted correction is recorded as a new version, never a silent edit.

Cite this article

GCRI editorial desk (2026, June 8). Stand-down, archive and renewal: making temporary builds last. The Global Centre for Risk and Innovation. https://therisk.global/teardown-archive-and-next-cycle-improvement-making-temporary-technical-infrastructure-cumulative/

Republish

Republish or translate this article free of charge, in print or online, under CC BY 4.0: keep the byline and The Global Centre for Risk and Innovation, link to the original and note any change. Figures, maps and twins made by the institutions carry the same licence; third-party data keeps its own credit line.

Attribution to paste
<p>Stand-down, archive and renewal: making a temporary build cumulative — GCRI editorial desk, The Global Centre for Risk and Innovation. First published 8 June 2026 at <a href="https://therisk.global/teardown-archive-and-next-cycle-improvement-making-temporary-technical-infrastructure-cumulative/">https://therisk.global/teardown-archive-and-next-cycle-improvement-making-temporary-technical-infrastructure-cumulative/</a>. Republished under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/).</p>

Topics

Related across the Consortium

Share

EmailLinkedInBlueskyMastodon

The Global Centre for Risk and Innovation (GCRI Canada) is a Canadian not-for-profit organisation, not a registered charity, with its registered office in Ontario.