Note
Companion appendix to the SGIR / PandemicSociety101 paper.
dv_ClaOp46Max_MMv1_sgir-appendix_2026m04d18Appendix: From Pandemic Modeling to Global Research Infrastructure#
This appendix presents material that extends the SGIR / PandemicSociety101 paper beyond its core scientific results. It addresses three questions: (1) Why did coordination fail during COVID-19? (2) What has the author been doing for six years instead of publishing this paper? (3) What infrastructure would be needed to deploy the Gap-of-Germs framework at scale?
A1. Why Coordination Failed: Work-Logic Cascades#
The Scenario 2 results in the main paper show that coordinated NPIs can produce a 60-fold reduction in pandemic infections. But coordination failed spectacularly during COVID-19. Why?
Figure 11 addresses this by presenting a work-logic cascade — a framework analogous to signal transduction cascades in molecular biology, showing how individual decisions about virus defense amplify through organizational levels to produce population-level effects.
The cascade operates across multiple scales: from personal decisions about mask-wearing and hygiene (the “last line of defense”), through household and workplace practices, to community-level coordination and national policy. At each level, the logic of virus defense interacts with pre-existing social structures — rule-making environments, trust levels, information quality, and resource availability. The cascade can amplify good decisions (when coordination succeeds) or dampen them (when trust breaks down or information is unreliable).
A key insight from the cascade analysis is that pandemic defense is a logistics problem, not primarily a virology problem. The biological knowledge for reducing Shed, Decay, and Catch rates existed early in the pandemic. What was missing was the organizational infrastructure to translate that knowledge into coordinated behavior change across diverse populations with different trust levels, information sources, and material resources.
A1.1 The MAPK Analogy#
The work-logic cascade design was directly informed by the author’s prior research on MAPK signal transduction cascades ((Loewe et al., 2009a), (Loewe et al., 2009b)). Figure 12 presents research notes from 2020 showing this mapping:
Panels A–C show stochastic simulations of a three-level MAPK cascade (from Ehlert and Loewe, 2014 (Ehlert and Loewe, 2014)): a weak input signal at Level 1 produces increasingly sharp, switch-like outputs at Levels 2 and 3. This signal amplification is the key property that makes cascades powerful.
Panel D shows how wiring between cascade levels determines whether signals are amplified or attenuated.
Panels E–G map this molecular framework onto the human decision landscape, identifying four behavioral states (Block, Trap, Rush, Best) at each cascade level and the probability conditions under which messages propagate through the cascade.
The analogy is precise: just as molecular signal transduction cascades can amplify a few signaling molecules into a cell-wide response, work-logic cascades can amplify individual decisions about virus defense into population-level behavioral change — if the “wiring” (care, hope, trust in work-logic cascades) is functional.
At the time (mid-2020), it was deemed unnecessary to translate this work-logic cascade into an actual Evolvix simulation, because the MAPK simulations had already demonstrated that corresponding molecular models exhibit switch-like behavior given the necessary parameters.
Molecular parameters can be measured in principle, even if difficult. But measuring human motivational parameters — e.g. sticking to the truth even if inconvenient, or persuading others to work for common goods instead of coveting selfish gains — would be a major undertaking. Current work-logic cascade model simulations would have to focus on exploring potentially interesting parameter ranges without direct measurements. Measuring these parameters is unnecessary for the practical decision to start organizing work-logic cascades: the COVID-19 pandemic itself demonstrated that coordination failures at many cascade levels had catastrophic consequences, not unlike the cascading organizational failures that led to the sinking of the Titanic (Butler, 2012).
A1.2 Rule-Making Environments and Decision Landscapes#
Figure 11 identifies four broad types of rule-making environments that shape work-logic cascades at every level:
A1 No Law: brute force wins, yet all lose — even the strong, since A1 is so inefficient.
A2 Bad Law: better than none (often, as it reflects good law; else like no law or worse).
A3 Ideal Law: is the fastest way to predict good decisions (e.g., pick a side of the road for driving); but Aristotle proved that law cannot cover all cases, so filling the gap needs…
A4 EpiEYEkeia: gentle kind reasonable equity justice; it adds posterior likelihood ideas from statistics to learning each situation before deciding with Aristotle’s epieikeia.
At each cascade level, four mental states shape decision-making: Block (fear-mind: stuck in analysis paralysis), Trap (fake-mind: speed over safety, seems OK until it isn’t), Rush (defy-mind: refuses to revisit or learn), and Best (edge-mind: narrow balance of courage and learning). The Titanic disaster (Section T in Figure 11) illustrates how Rush-state decisions cascaded through multiple organizational levels to produce catastrophe — a pattern that recurred during COVID-19.
A1.3 Information Flow: Places of Reasoning#
Figure 13 presents a framework for organizing information flow to reduce the chaos that currently hampers pandemic response.
The key insight is that unstructured information flows (“info-jungle”) are slow, costly, and error-prone, while typed Places of Reasoning (PoR) that declare their reasoning standards and link to evidence can dramatically cut the cost of fact-checking and quality control. This is essential for any coordinated pandemic response where decisions at many cascade levels must be made rapidly and reliably.
A2. From Pandemic to All Existential Challenges (2020–2026)#
The six-year gap between simulations (2020) and publication (2026) was not idle time. It was consumed by a discovery that changed the scope of the work entirely.
The generalization. While analyzing why pandemic coordination failed, it became clear that the same cascading work-logic failures that prevented effective COVID-19 response also undermine responses to every other existential challenge: climate change, biodiversity loss, AI safety, nuclear risk, antibiotic resistance, soil erosion, water security. In each case, the biological or physical knowledge for addressing the problem exists, but the organizational infrastructure for translating that knowledge into coordinated action does not.
This generalization is documented in detail in OL5b — a 28-page letter to the UN Secretary-General (2025m01d27, revised through 2025m11d26; available at Balospe.com) — which identifies seven categories of existential risk (“7 Death Urn Incinerators”) that share the same structural failure pattern: cascading work-logic breakdowns across organizational levels, compounded by information overload that prevents warnings from reaching decision-makers (the “datageddon” problem, analogous to the Titanic’s failure to escalate iceberg telegrams to the bridge).
The governance problem. Extending work-logic cascades to multiple existential challenges immediately revealed a deeper problem: who governs the infrastructure? Any global research institution powerful enough to coordinate existential-risk responses is also powerful enough to be captured by special interests, corrupted by internal politics, or paralyzed by its own bureaucracy. The author spent much of the intervening years working on the fundamental governance challenges that any such institution must solve before it can function reliably.
This work led to the development of a mathematical framework for governance — presented in the Matheo paper series at Balospe.com — that addresses the structural conditions under which coordinated action remains gentle, kind, and reasonable over the long term for all affected sides, including the weakest. The framework draws on mathematical logic, game theory, information theory, and a careful analysis of historical governance failures.
The Jonah problem. The irony is not lost on the author: like the biblical Jonah, who slept in the hold of a ship while a storm raged, the author had the tools to address the pandemic crisis but did not deploy them in time. The delay was not caused by laziness but by the conviction that releasing alarming pandemic numbers without offering a constructive path forward would amount to fear-mongering — and the constructive path (ResearchCity, work-logic cascades, governance framework) required years of development that could not be shortcut.
Whether this conviction was wise or whether the pandemic paper should have been published immediately in 2020 — accepting the risk of fear-mongering in exchange for earlier public benefit — is a question the author cannot answer with certainty. The work-logic cascade framework itself predicts that this type of delay (Block state: too many details to test, unable to act) is one of the four failure modes at every cascade level. Acknowledging this failure mode in oneself is part of the framework’s self-testing design.
A3. Virodefense Olympics and ResearchCity#
The pandemic modeling and work-logic cascade analysis motivated two institutional concepts:
A3.1 Virodefense Olympics#
Annual, competitive, gamified training exercises designed to build and maintain the organizational muscle for coordinated virus defense before the next pandemic arrives. Just as fire drills prepare buildings for emergencies and military exercises maintain readiness, Virodefense Olympics would maintain society’s capacity for rapid, coordinated NPI adoption. The competitive element (between teams, cities, or nations) provides motivation, while the annual recurrence prevents the institutional amnesia that leaves societies unprepared when decades pass between major pandemics.
A3.2 ResearchCity#
A distributed research infrastructure organized around work-logic cascades, where dedicated units (stadia) each address a specific existential challenge as a global public service. The virodefense stadion (STa2-WWV) would organize Virodefense Olympics; an Evolvix stadion (STa1-EVX) would maintain and develop the modeling infrastructure; other stadia would address other existential challenges identified by the 7DUI analysis.
The organizational challenge of scaling such infrastructure requires careful attention to information flow. Figure 13 presents a framework for how different Places of Reasoning can organize information flow to reduce chaos and improve the quality and speed of decision-making across cascade levels.
The detailed institutional design for ResearchCity — including governance structures (epiocracy), funding mechanisms, accountability systems, and a proposed UN Mandate — is presented in OL5b and subsequent documents at Balospe.com.
A4. Funding: The $8 Design and Its Rationale#
Any institution organizing Virodefense Olympics must confront an uncomfortable reality: enormous levels of mistrust in organized medicine — and especially medical for-profit companies — mean that conventional funding would undermine the very trust that such Olympics are designed to build.
This led to the concept of independent crowd-funded research stadia, each running its own public campaign inviting contributions of at most approximately $8 per person per year — roughly two cents a day.
Why $8? This cap is not arbitrary. It is deliberately calibrated to be accessible even at the median income of the world’s poorest countries. The design intent is that everyone — including the poorest of the poor — can contribute their share toward an institution that is audited to work for everybody, including the weakest. The cap simultaneously keeps large corporate donors at arm’s length, ensuring fiduciary responsibility toward the global public rather than toward special-interest shareholders.
This echoes a principle as old as recorded ethics: a small contribution freely given by someone who has almost nothing can matter more than a large contribution from someone who will never miss it — because the former represents genuine commitment while the latter may represent mere convenience. The design ensures that the institution’s accountability runs toward the many, not the few.
Those with greater means are invited to sponsor access for others who cannot yet participate — effectively buying in on behalf of the two-thirds of the world’s population who lack the infrastructure (credit cards, bank accounts, internet access) to contribute directly. The transparency mechanisms needed to make this work reliably are themselves a research challenge, addressed by the governance framework described in the Matheo series.
A5. Lessons for Evolvix Modeling Language Design#
The declarative Evolvix syntax used in this work demonstrates the potential of domain-specific languages for biological modeling: without the simplifications introduced by this Evolvix prototype, the PandemicSociety101 model would have been far more difficult to construct and modify. The Evolvix mission — simplify accurate modeling with a long-term stable, extensible, humane computer language for biologists — and its vision — improve responsible decision-making worldwide by modeling uncertainties, values, and logics — are both directly relevant to pandemic preparedness.
However, using Evolvix during an active pandemic revealed important design limitations. As any careful reader of the supplementary code will observe, there are numerous instances where general-purpose multi-paradigm programming constructs would have made the code substantially more readable. The key challenge is that in biology, almost everything is uncertain to some degree. Making this biouncertainty — inherently foreign to digital systems — a first-class citizen in the language is essential but architecturally demanding.
Working on Evolvix design while an active pandemic was unfolding revealed a series of architectural flaws that must be addressed to produce a version deserving the label “pandemic-grade.” It proved impossible to anticipate the needs of pandemic-stress modeling without experiencing that stress. This is itself a lesson for pandemic preparedness: the tools needed for the next pandemic must be developed and stress-tested before the crisis, not during it.
Developing such common-goods infrastructure requires sustained focus over many years — far longer than typical university research material retention periods (~7 years) or grant cycles. This observation further motivates the ResearchCity concept: a dedicated Evolvix stadion (STa1-EVX) working alongside the virodefense stadion (STa2-WWV) would provide the institutional continuity needed to evolve a pandemic-grade modeling language over the decades required.
References#
See the main paper for the full reference list. Additional references for this appendix:
Loewe, L., et al. (2009). “Quantifying the implicit process flow abstraction in SBGN-PD diagrams with Bio-PEPA.” EPTCS 6: 93–107. http://arxiv.org/abs/0910.1410
Loewe, L., et al. (2009). “Defining a textual representation for SBGN Process Diagrams and translating it to Bio-PEPA for quantitative analysis of the MAPK signal transduction cascade.” Technical Report EDI-INF-RR-1334, School of Informatics, University of Edinburgh.
Butler, D. A. (2012). “Unsinkable”: The Full Story of RMS Titanic. Boston, MA: Da Capo Press. [The Titanic disaster is a paradigmatic example of cascading work-logic failures: ignored warnings, overconfidence in technology, inadequate safety provisions, and organizational failures at every level. The 2023 OceanGate Titan implosion while searching for the Titanic wreck provided a tragic reminder that these cascade failure patterns recur when organizational safeguards are bypassed.]
Ehlert, K. and L. Loewe (2014). “Lazy Updating of hubs can enable more realistic models by speeding up stochastic simulations.” Journal of Chemical Physics 141(20): 204109. doi: https://doi.org/10.1063/1.4902225
OL5b: Loewe, L. (2025). “Detailed Letter to UN Secretary-General: Stop Accidental Nuclear Winter with ResearchCity.” 28 pages. Available at Balospe.com.
Matheo series: Loewe, L. (2026). Papers on mathematical theology and governance frameworks. In preparation. Balospe.com.