Pith. sign in

REVIEW 3 major objections 6 minor 108 references

Steps towards an Ecology for the Internet

T0 review · 3 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The Internet has no built-in immune system; this paper argues it can evolve one by borrowing from ecology — local antibody-like hosts that contain malware, AI-driven mutation of protocol software to break monoculture, and a shift back to…

desk verdict A clear, honest position piece worth engaging with, but its one quantitative-sounding claim about local containment outrunning botnets does not survive an adaptive adversary. read the letter →

arxiv 2506.06469 v1 pith:UPGVXJWA submitted 2025-06-06 cs.NI cs.ET

Anil Madhavapeddy , Sam Reynolds , Alec P. Christie , David A. Coomes , Michael W. Dales , Patrick Ferris , Ryan Gibb , Hamed Haddadi
show 5 more authors
Sadiq Jaffer Josh Millar Cyrus Omar William J. Sutherland Jon Crowcroft
This is my paper
classification cs.NIcs.ET
keywords ecologyprotocolsevolutionbotnetsmalwarecontainmentcodemodelsre-decentralisation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper argues that the Internet, heading toward a trillion nodes, is dangerously fragile because it has no built-in immune system. It proposes three ecological mechanisms for repairing that: local 'antibotty' hosts that scan and patch neighboring devices before global botnets can; software-stack 'mutations' generated by AI code models under formal protocol constraints, to end the software monoculture; and a re-decentralisation toward mutualistic services. The authors present these as thought experiments and research directions rather than deployed systems, and they flag the risk that the antibody-like defenses themselves could be subverted. If the proposals are right, malware could be contained locally, a single vulnerability would no longer compromise every host at once, and a less centralized, more cooperative network becomes plausible.

What carries the argument

The carrying mechanism is the antibotty network, a digital analog of the adaptive immune system in which each host scans only nearby friendly hosts, isolates or patches infected ones, and is coordinated by a hierarchy of scoped controllers so the vigilantes do not themselves become a threat. The second mechanism is 'protocol mutatis mutandis,' directed mutation of protocol implementations by AI code models whose outputs are constrained by formal specifications distilled from the RFC corpus, with mutations inserted and evaluated at containment layers such as hardware protections, unikernels, WebAssembly sandboxes, and formal network verification. The paper's mapping table (DNA to RFCs, cells to endpoints, tissues to host clusters) frames both mechanisms as ecological engineering rather than metaphor.

What would settle it

A controlled testbed in which a fast-spreading worm can read the antibotty alert channel would settle the central claim: if the worm propagates faster through that channel than the local scanners can contain it, the containment mechanism fails and becomes a new infection vector.

Watch

Extended reading notes

Core claim

The paper's central claim is that the Internet's lack of a defensive immune system is a correctable design flaw, and that ecology provides the design patterns. The end-to-end principle kept the network core simple and pushed complexity to the edge, leaving hosts to defend themselves individually; combined with a software monoculture, this means a single vulnerability compromises millions of hosts within minutes. The authors propose an 'antibotty' network in which every host plays the role of an antibody, scanning only its local community — orders of magnitude fewer devices than a global botnet must scan — and taking active measures to isolate or deliberately patch infected neighbors before the botnet can act. They further propose 'protocol mutatis mutandis': directed mutation of protocol implementations by AI code models, constrained by formal specifications extracted from the RFC corpus and contained by layered hardware and OS sandboxing, to increase architectural diversity. The final claim is that these local defence mechanisms enable a re-decentralisation of the Internet toward mutualistic protocols such as ActivityPub and the AT Protocol, restoring the cooperation the early Internet exhibited.

Load-bearing premise

The argument stands on the premise that a vigilante host can scan, isolate, and patch its neighbors without itself being compromised, and that AI code models, even constrained by formal specifications, can produce protocol mutations that are safe rather than exploitable.

Editorial extensions

If this is right

  • Antibotty containment would outpace global botnets because a local host scans orders of magnitude fewer devices than a netblock-wide scanner must.
  • Software-stack mutations constrained by formal protocol specifications would break the monoculture, so a single vulnerability would no longer expose every host at once.
  • Layered containment (hardware protection, OS sandboxing, capability-based access, formal verification) can keep AI-generated mutations from becoming new attack surfaces.
  • A hierarchy of controllers, analogous to the blood-brain barrier, would stop vigilante hosts from inflicting unintended harm on mission-critical devices.
  • Redecentralized social protocols (ActivityPub, ATProto) plus mutualistic defense protocols could restore the cooperative structure the early Internet had.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the antibotty argument is right, the speed advantage of local scanning is testable today: measure how many devices a local scanner covers versus a netblock-wide botnet scan in the same time window.
  • The mutation proposal would invert into a liability if an attacker can steer code-model outputs; the paper's containment layers are then the entire security story, and hardening them becomes the real research problem.
  • The biological analogy implies that hosts that stop contributing ('cheaters') must be detectable and sanctionable; the paper does not propose such a protocol, so designing an incentive-compatible cooperation metric is a concrete next step.
  • Read sympathetically, the end-to-end principle is both the cause of the immune deficit and the reason new defenses can be deployed edge-first; the natural synthesis is a protocol that preserves edge flexibility while adding local, scoped defense.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. The paper is a vision/position piece arguing that the Internet has become a software monoculture with no built-in immune system, and that concepts from ecology—local 'antibotty' defenses, AI-generated protocol-compatible mutations, and mutualistic re-decentralization—could make it more resilient as it grows toward a trillion nodes. It is explicitly framed as an inspiration for future research rather than a comprehensive solution, and it contains no implementations, measurements, or formal models. The core thought experiments are presented through an analogy table (Table 1) and three main sections, with a caveats section that acknowledges several open risks.

Significance. If the broad claims held, the paper would open a productive interdisciplinary research agenda connecting ecological theory, network security, and programming languages. Its strengths are its wide-ranging and mostly well-chosen references, its explicit statement that the proposals are thought experiments, and its honest caveats in Section 5 about mutualism not being guaranteed and about local defenses being unable to handle global rapid spread. However, the specific feasibility claims that carry the argument—notably the claim that antibotties 'outrace' botnets and that AI code models can be safely constrained by formal specifications—are not yet supported, and the manuscript's own caveats concede the central open problems. There are no machine-checked proofs or reproducible artifacts to verify, so the contribution is conceptual.

major comments (3)
  1. [§2.1] The sentence 'The antibotty network approach outraces global botnets because local hosts only need to scan orders of magnitude fewer devices in their vicinity' is load-bearing for the proposed defense, but it is not established and is unlikely to hold against an adaptive adversary. The paper grants every vigilante host the capability to scan neighbours, exploit a vulnerability, and patch it; that is precisely the primitive a botnet needs, and a botnet that has compromised one host can use the same local-scanning strategy, so the claimed temporal advantage disappears. 'Advertising its intent' and 'not incentivised to exploit' do not constitute security mechanisms. The paper should either weaken this to an explicitly open conjecture with a threat model, or provide a concrete mechanism (e.g., hardware attestation or authenticated patch signing) that gives vigilantes an asymmetric capability.
  2. [§3.1] The feasibility claim that AI code models, 'with sufficient guardrails,' can generate protocol-compatible mutations, and that formal specifications can guide LLMs 'towards generating only permissible protocol messages,' is not supported. The cited systematic review [57] and adversarial-attacks work [101] point to known failure modes, and Section 3.2 itself concedes that malware could guide mutations into new exploits. Because the mutation pipeline is one of the paper's three central defenses, this needs to be reframed as a research challenge with an explicit threat model and a discussion of what selection pressure or fitness function would prevent harmful mutations from spreading; otherwise the claim should be removed.
  3. [§2.1, §3.2] The paper's own risk acknowledgment, 'great care needs to be taken to ensure that the antibotties do not themselves become a threat' (§2.1), is not backed by any design argument; the proposed 'hierarchy of controllers' (§2.1) and 'capability-based programming' (§3.2) are mentioned as directions, not mechanisms. Because the antibotty concept is the one place where the paper makes a concrete quantitative-sounding claim, the absence of even a sketch of the trust or attestation architecture is a load-bearing gap. The authors should either supply a minimal adversary model showing what property gives vigilantes an advantage, or explicitly mark the scheme as an open problem.
minor comments (6)
  1. [Abstract] The abstract contains a duplicated phrase: 'We take lessons from from biological systems' should be 'We take lessons from biological systems.'
  2. [§1.1] The word 'openess' should be 'openness,' and the phrase 'The Internet now currently lacks' is redundant; 'currently' can be removed.
  3. [§3] The section heading 'SCALING THE INTERNET TOW ARDS A TRILLION NODES' contains a stray space; it should read 'TOWARDS.'
  4. [§2.1] The name 'Ribiero' in the discussion of reference [74] should be spelled 'Ribeiro' to match the cited author.
  5. [§3.2] In the paragraph on sexual reproduction, the sentence 'These combinations may result in produced diverse lifeforms that are not only exciting new options but are also are different from each other and the parent' contains the awkward 'produced' and the duplicated 'are also are'; the sentence should be rewritten.
  6. [Table 1] The row mapping 'Chromosome' to 'Software interfaces, formal specifications' is not used consistently in the text: Section 2, which the table cites, does not return to the chromosome analogy after the introductory mapping.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the paper is a position/vision essay with no equations, fitted parameters, or derived predictions, and its self-citations are background references rather than load-bearing premises.

full rationale

The paper makes no quantitative derivation that could reduce to its own inputs. Its two central proposals, the antibotty local-defense scheme (§2.1) and protocol mutation guided by formal specifications (§3.1), are presented as thought experiments and research directions, explicitly framed as "our primary goal is to inspire further research and discussion, rather than to provide a comprehensive solution." The antibotty timing claim that "local hosts only need to scan orders of magnitude fewer devices in their vicinity" is an argumentative assertion, not a fitted parameter or a prediction computed from a model, so it cannot be circular in the sense of re-deriving an input. Table 1 maps biological concepts to Internet concepts by fiat, but that mapping is expository framing and does not define any outcome in terms of another claimed result. The paper cites several works by its own authors (e.g., databoxes [14], unikernels [49, 50], and [61]), but these are contextual background references and are not used as the justification for the paper's central claims; no uniqueness theorem or machine-checked result is imported to force a conclusion. The authors' own caveat that "great care needs to be taken to ensure that the antibotties do not themselves become a threat" flags a genuine open security concern, but an unresolved risk is a correctness or feasibility issue, not a circularity issue. There are no fitted inputs renamed as predictions, no self-definitional equations, and no self-citation chain that makes the argument equivalent to its premises. The derivation chain is therefore self-contained in the only sense applicable here: the paper proposes analogies and directions rather than deriving results, and none of those proposals is forced by construction.

Assumptions & free parameters 0 free parameters · 5 assumptions · 1 invented entities

The paper introduces no fitted numeric parameters. Its central reasoning rests on domain assumptions about the transferability of biological mechanisms to software, each invoked without empirical validation and in some cases explicitly flagged by the authors as unproven.

assumptions (5)
  • domain assumption Biological immune and ecological mechanisms (local surveillance, somatic hypermutation, horizontal gene transfer, mutualism) are functionally transferable to Internet protocols and software stacks.
    Invoked in Sections 2.1, 3.1, and 4; no evidence or quantitative model is given to show the mapping preserves the properties that make these mechanisms adaptive in nature.
  • domain assumption AI-driven code models, given formal specifications as guardrails, can generate protocol-compatible patches that are safe to deploy.
    Section 3.1 assumes 'with sufficient guardrails, it is possible to create patches for existing code'; cited refs [46, 57] show code generation is improving but not yet at human level.
  • domain assumption A local host can reliably detect malware on neighboring hosts and take containment actions without being co-opted by an adversary.
    Section 2.1 makes this the basis of antibotty networks and immediately flags it as an unsolved risk; no design or evaluation is supplied.
  • domain assumption Increasing software diversity will increase Internet resilience, analogous to biodiversity in ecosystems.
    Section 2 and 3.1 draw this conclusion from ecology; while plausible, the paper offers no network-level experiments or data.
  • domain assumption After the current parasitic phase, mutualism is a likely stable outcome for Internet stakeholders.
    Section 4 relies on ecological theory (refs [8, 77, 80]) and explicitly hedges that mutualism is 'not guaranteed'.
invented entities (1)
  • Antibotty (vigilante endpoint acting as an immune cell)
    purpose: Acts locally to detect, isolate, and patch malware on neighboring hosts without central authority.
    Proposed in Section 2.1 as a thought experiment; no implementation, prototype, or measurement exists, and the authors note the mechanism could itself become a threat.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Steps towards an Ecology for the Internet." pith.science (2026). https://pith.science/paper/UPGVXJWA

@misc{pith2026250606469,
  author       = {Pith},
  title        = {Pith review of: Steps towards an Ecology for the Internet},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/UPGVXJWA}},
  note         = {Machine review of arXiv:2506.06469}
}
read the original abstract

The Internet has grown from a humble set of protocols for end-to-end connectivity into a critical global system with no builtin "immune system". In the next decade the Internet will likely grow to a trillion nodes and need protection from threats ranging from floods of fake generative data to AI-driven malware. Unfortunately, growing centralisation has lead to the breakdown of mutualism across the network, with surveillance capitalism now the dominant business model. We take lessons from from biological systems towards evolving a more resilient Internet that can integrate adaptation mechanisms into its fabric. We also contribute ideas for how the Internet might incorporate digital immune systems, including how software stacks might mutate to encourage more architectural diversity. We strongly advocate for the Internet to "re-decentralise" towards incentivising more mutualistic forms of communication.

Figures

Figures reproduced from arXiv: 2506.06469 by the authors.

Figure 1
Figure 1. The Internet needs to urgently evolve active de [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. The modern software stack has a number of points [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

108 extracted references · 54 canonical work pages

  1. [57]

    A systematic literature review on the impact of AI models on the security of code genera- tion

    Claudia Negri-Ribalta et al. “A systematic literature review on the impact of AI models on the security of code genera- tion”. In: Frontiers in Big Data 7 (May 2024). doi: 10.3389/ fdata.2024.1386720

    arXiv 2024

  2. [101]

    CARL: Unsupervised Code-Based Adver- sarial Attacks for Programming Language Models via Rein- forcement Learning

    Kaichun Yao et al. “CARL: Unsupervised Code-Based Adver- sarial Attacks for Programming Language Models via Rein- forcement Learning”. In: ACM Trans. Softw. Eng. Methodol. 34.1 (Dec. 2024). doi: 10.1145/3688839

    work page doi:10.1145/3688839 2024

  3. [1]

    Structure and function of the blood– brain barrier

    N Joan Abbott et al. “Structure and function of the blood– brain barrier”. In:Neurobiology of disease 37.1 (2010), pp. 13– 25

    2010

  4. [2]

    A call to reconcep- tualize lichen symbioses

    Jessica L Allen and James C Lendemer. “A call to reconcep- tualize lichen symbioses”. In: Trends in Ecology & Evolution 37.7 (2022), pp. 582–589

    2022

  5. [3]

    NetKAT: semantic founda- tions for networks

    Carolyn Jane Anderson et al. “NetKAT: semantic founda- tions for networks”. In:Proceedings of the 41st ACM SIGPLAN- SIGACT Symposium on Principles of Programming Languages. POPL ’14. San Diego, California, USA: Association for Com- puting Machinery, 2014, pp. 113–126. doi: 10.1145/2535838. 2535862

    doi:10.1145/2535838 2014

  6. [4]

    Reli- able Recon in Adversarial Peer-to-Peer Botnets

    Dennis Andriesse, Christian Rossow, and Herbert Bos. “Reli- able Recon in Adversarial Peer-to-Peer Botnets”. In:Proceed- ings of the 2015 Internet Measurement Conference . IMC ’15. ACM, Oct. 2015, pp. 129–140. doi: 10.1145/2815675.2815682

    arXiv 2015

  7. [5]

    Understanding the mirai botnet

    Manos Antonakakis et al. “Understanding the mirai botnet”. In: 26th USENIX security symposium (USENIX Security 17) . 2017, pp. 1093–1110

    2017

  8. [6]

    Hor- izontal gene transfer and adaptive evolution in bacteria

    Brian J. Arnold, I-Ting Huang, and William P. Hanage. “Hor- izontal gene transfer and adaptive evolution in bacteria”. In: Nature Reviews Microbiology 20.4 (Nov. 2021), pp. 206–218. doi: 10.1038/s41579-021-00650-4

    work page doi:10.1038/s41579-021-00650-4 2021

Show all 108 references
  1. [7]

    Biomimetics: biologically inspired tech- nologies

    Yoseph Bar-Cohen. Biomimetics: biologically inspired tech- nologies. CRC press, 2005

    2005

  2. [8]

    Mutualistic networks

    Jordi Bascompte. “Mutualistic networks”. In: Frontiers in Ecology and the Environment 7.8 (Sept. 2008), pp. 429–436. doi: 10.1890/080026

    2008 doi

  3. [9]

    Steps to an ecology of mind

    Gregory Bateson. Steps to an ecology of mind . en. Chicago, IL: University of Chicago Press, May 2000

    2000

  4. [10]

    The Red Queen and the Court Jester: Species Diversity and the Role of Biotic and Abiotic Factors Through Time

    Michael J. Benton. “The Red Queen and the Court Jester: Species Diversity and the Role of Biotic and Abiotic Factors Through Time”. In: Science 323.5915 (Feb. 2009), pp. 728–

    2009

  5. [11]

    Mechanisms under- lying beneficial plant–fungus interactions in mycorrhizal symbiosis

    Paola Bonfante and Andrea Genre. “Mechanisms under- lying beneficial plant–fungus interactions in mycorrhizal symbiosis”. In: Nature Communications 1.1 (July 2010). doi: 10.1038/ncomms1046

    2010 doi

  6. [12]

    Connecting and inte- grating cooperation within and between species

    Judith L. Bronstein and Hari Sridhar. “Connecting and inte- grating cooperation within and between species”. In: Philo- sophical Transactions of the Royal Society B: Biological Sci- ences 379.1909 (July 2024). doi: 10.1098/rstb.2023.0203

    1909

  7. [13]

    A Survey on Web Tracking: Mecha- nisms, Implications, and Defenses

    Tomasz Bujlow et al. “A Survey on Web Tracking: Mecha- nisms, Implications, and Defenses”. In: Proceedings of the IEEE 105.8 (Aug. 2017), pp. 1476–1510. doi: 10.1109/jproc. 2016.2637878

    2017

  8. [14]

    Personal Data: Thinking Inside the Box

    Amir Chaudhry et al. “Personal Data: Thinking Inside the Box”. In: Aarhus Series on Human Centered Computing 1.1 (Oct. 2015), pp. 4–4. doi: 10.7146/aahcc.v1i1.21312

    2015 doi

  9. [15]

    Editorial: Symbiotic Relation- ships as Shapers of Biodiversity

    Guillaume Chomicki et al. “Editorial: Symbiotic Relation- ships as Shapers of Biodiversity”. In: Frontiers in Ecology and Evolution 10 (Feb. 2022). doi: 10.3389/fevo.2022.850572

    2022

  10. [16]

    2024 Global Networking Trends Report

    Cisco. 2024 Global Networking Trends Report . 2024

    2024

  11. [17]

    The size and growth rate of the Internet

    Kerry Coffman and Andrew Odlyzko. “The size and growth rate of the Internet”. In: (1998)

    1998

  12. [18]

    Vigilante: end-to-end containment of internet worms

    Manuel Costa et al. “Vigilante: end-to-end containment of internet worms”. In: SIGOPS Oper. Syst. Rev. 39.5 (Oct. 2005), pp. 133–147. doi: 10.1145/1095809.1095824

    2005

  13. [19]

    Measuring and fingerprinting click-spam in ad networks

    Vacha Dave, Saikat Guha, and Yin Zhang. “Measuring and fingerprinting click-spam in ad networks”. In: Proceedings of the ACM SIGCOMM 2012 conference . 2012, pp. 175–186

    2012

  14. [20]

    Perspectives on Computation in Plants

    Emanuela Del Dottore and Barbara Mazzolai. “Perspectives on Computation in Plants”. In: Artificial Life 29.3 (2023), pp. 336–350. doi: 10.1162/artl_a_00396

    2023 doi

  15. [21]

    Large mammal declines and the incipient loss of mammal-bird mutualisms in an African savanna ecosystem

    Nathan Diplock et al. “Large mammal declines and the incipient loss of mammal-bird mutualisms in an African savanna ecosystem”. In: PLOS ONE 13.8 (Aug. 2018). Ed. by Willem F. de Boer, e0202536. doi: 10.1371/journal.pone. 0202536

    2018 doi

  16. [22]

    Architecture, constraints, and behavior

    John C. Doyle and Marie Csete. “Architecture, constraints, and behavior”. In: Proceedings of the National Academy of Sciences 108.supplement_3 (2011), pp. 15624–15630. doi: 10.1073/pnas.1103557108

    2011 doi

  17. [23]

    Confidential carbon commuting: ex- ploring a privacy-sensitive architecture for incentivising ’greener’ commuting

    Chris Elsmore et al. “Confidential carbon commuting: ex- ploring a privacy-sensitive architecture for incentivising ’greener’ commuting”. In: Proceedings of the First Workshop on Measurement, Privacy, and Mobility. MPM ’12. ACM, Apr. 2012, pp. 1–6. doi: 10.1145/2181196.2181201

    2012

  18. [24]

    LLM Agents can Autonomously Exploit One-day Vulnerabilities

    Richard Fang et al. LLM Agents can Autonomously Exploit One-day Vulnerabilities. 2024. doi: 10.48550/ARXIV.2404. 08144

    2024 doi

  19. [25]

    How to lease the internet in your spare time

    Nick Feamster, Lixin Gao, and Jennifer Rexford. “How to lease the internet in your spare time”. In:SIGCOMM Comput. Commun. Rev. 37.1 (Jan. 2007), pp. 61–64. doi: 10.1145/ 1198255.1198265

    2007

  20. [26]

    CODE reuse in practice: Benefiting or harming technical debt

    Daniel Feitosa et al. “CODE reuse in practice: Benefiting or harming technical debt”. In:Journal of Systems and Software 167 (Sept. 2020), p. 110618. doi: 10.1016/j.jss.2020.110618

    2020

  21. [27]

    Why the Internet only just works

    Mark Handley. “Why the Internet only just works”. In: BT Technology Journal 24.3 (2006), pp. 119–129

    2006

  22. [28]

    Mycorrhizal fungal di- versity determines plant biodiversity, ecosystem variability and productivity

    Marcel G. A. van der Heijden et al. “Mycorrhizal fungal di- versity determines plant biodiversity, ecosystem variability and productivity”. In: Nature 396.6706 (Nov. 1998), pp. 69–

    1998

  23. [29]

    Efficient evolution of human antibodies from general protein language models

    Brian L. Hie et al. “Efficient evolution of human antibodies from general protein language models”. In: Nature Biotech- nology 42.2 (Apr. 2023), pp. 275–283. doi: 10.1038/s41587- 023-01763-2. Steps towards an Ecology for the Internet AAR 2025, August 18–22, 2025, Aarhus N, Denmark

    2023 doi

  24. [30]

    A new vision for how evolution works is long overdue

    Eva Jablonka. “A new vision for how evolution works is long overdue”. In: Nature 637.8046 (Jan. 2025), pp. 539–541. doi: 10.1038/d41586-025-00054-x

    2025 doi

  25. [31]

    Solar superstorms: planning for an internet apocalypse

    Sangeetha Abdu Jyothi. “Solar superstorms: planning for an internet apocalypse”. In: Proceedings of the 2021 ACM SIG- COMM 2021 Conference. SIGCOMM ’21. Virtual Event, USA: ACM, 2021, pp. 692–704. doi: 10.1145/3452296.3472916

    2021

  26. [32]

    J. Kempf. The Rise of the Middle and the Future of End-to- End: Reflections on the Evolution of the Internet Architecture . Ed. by R. Austein. Mar. 2004. doi: 10.17487/rfc3724

    2004 doi

  27. [33]

    The Web We Weave: Untangling the Social Graph of the IETF

    Prashant Khare et al. “The Web We Weave: Untangling the Social Graph of the IETF”. In:Proceedings of the International AAAI Conference on Web and Social Media 16 (May 2022), pp. 500–511. doi: 10.1609/icwsm.v16i1.19310

    2022 doi

  28. [34]

    Bluesky and the AT Protocol: Usable Decentralized Social Media

    Martin Kleppmann et al. “Bluesky and the AT Protocol: Usable Decentralized Social Media”. In: Proceedings of the ACM Conext-2024 Workshop on the Decentralization of the Internet. CoNEXT ’24. ACM, Dec. 2024, pp. 1–7. doi: 10. 1145/3694809.3700740

    2024

  29. [35]

    Intentional Ecol- ogy: Integrating environmental expertise through a focus on values, care and advocacy

    Alexandra R. Knight and Catherine Allan. “Intentional Ecol- ogy: Integrating environmental expertise through a focus on values, care and advocacy”. In:Humanities and Social Sci- ences Communications 8.1 (Nov. 2021). doi: 10.1057/s41599- 021-00960-1

    2021 doi

  30. [36]

    FinchGPT: a Transformer based lan- guage model for birdsong analysis

    Kosei Kobayashi et al. FinchGPT: a Transformer based lan- guage model for birdsong analysis . 2025. doi: 10 . 48550 / ARXIV.2502.00344

    2025 doi arXiv

  31. [37]

    Metabolic exchanges are ubiquitous in natural microbial communities

    Christian Kost et al. “Metabolic exchanges are ubiquitous in natural microbial communities”. In: Nature Microbiology 8.12 (Nov. 2023), pp. 2244–2252. doi: 10.1038/s41564-023- 01511-x

    2023 doi

  32. [38]

    Evolution evolving

    Kevin N Lala et al. Evolution evolving. en. Princeton, NJ: Princeton University Press, Oct. 2024

    2024

  33. [39]

    Does evolutionary theory need a re- think?

    Kevin Laland et al. “Does evolutionary theory need a re- think?” In: Nature 514.7521 (Oct. 2014), pp. 161–164. doi: 10.1038/514161a

    2014 doi

  34. [40]

    Cooperative diversity in wireless networks: Efficient protocols and out- age behavior

    J.N. Laneman, D.N.C. Tse, and G.W. Wornell. “Cooperative diversity in wireless networks: Efficient protocols and out- age behavior”. In: IEEE Transactions on Information Theory 50.12 (2004), pp. 3062–3080. doi: 10.1109/TIT.2004.838089

    2004

  35. [41]

    SoK: Automated Software Diversity

    Per Larsen et al. “SoK: Automated Software Diversity”. In: 2014 IEEE Symposium on Security and Privacy . IEEE, May 2014, pp. 276–291. doi: 10.1109/sp.2014.25

    2014 doi

  36. [42]

    The science of fake news

    David MJ Lazer et al. “The science of fake news”. In:Science 359.6380 (2018), pp. 1094–1096

    2018

  37. [43]

    Open questions in the social lives of viruses

    Asher Leeks et al. “Open questions in the social lives of viruses”. In: Journal of Evolutionary Biology 36.11 (Nov. 2023), pp. 1551–1567. doi: 10.1111/jeb.14203

    2023 doi

  38. [44]

    Gaia as a com- plex adaptive system

    Timothy M. Lenton and Marcel van Oijen. “Gaia as a com- plex adaptive system”. In: Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 357.1421 (May 2002). Ed. by R. V. Solé and S. A. Levin, pp. 683–695. doi: 10.1098/rstb.2001.1014

    2002

  39. [45]

    Relations with Animals of Other Species

    Walter Leuthold. “Relations with Animals of Other Species”. In: African Ungulates . Springer Berlin Heidelberg, 1977, pp. 71–83. doi: 10.1007/978-3-642-81073-2_8

    1977 doi

  40. [46]

    Competition-level code generation with Al- phaCode

    Yujia Li et al. “Competition-level code generation with Al- phaCode”. In: Science 378.6624 (Dec. 2022), pp. 1092–1097. doi: 10.1126/science.abq1158

    2022 doi

  41. [47]

    Good News for People Who Love Bad News: Centralization, Privacy, and Trans- parency on US News Sites

    Timothy Libert and Reuben Binns. “Good News for People Who Love Bad News: Centralization, Privacy, and Trans- parency on US News Sites”. In: Proceedings of the 10th ACM Conference on Web Science. WebSci ’19. New York, NY, USA: Association for Computing Machinery, 2019, pp. 155...

    2019

  42. [48]

    Environmental quality mediates the ecological dominance of cooperatively breeding birds

    Yu-Heng Lin et al. “Environmental quality mediates the ecological dominance of cooperatively breeding birds”. In: Ecology Letters 26.7 (May 2023), pp. 1145–1156.doi: 10.1111/ ele.14226

    2023

  43. [49]

    Unikernels: Rise of the Virtual Library Operating System

    Anil Madhavapeddy and Dave Scott. “Unikernels: Rise of the Virtual Library Operating System”. In:ACM Queue 11.11 (Nov. 2013), pp. 30–44. doi: 10.1145/2557963.2566628

    2013

  44. [50]

    Unikernels: library operating systems for the cloud

    Anil Madhavapeddy et al. “Unikernels: library operating systems for the cloud”. In: Proceedings of the eighteenth in- ternational conference on Architectural support for program- ming languages and operating systems . Houston Texas USA: ACM, Mar. 2013, pp. 461–472.doi: 10.1145...

    2013

  45. [51]

    Characterising the IETF through the lens of RFC deployment

    Stephen McQuistin et al. “Characterising the IETF through the lens of RFC deployment”. In: Proceedings of the 21st ACM Internet Measurement Conference. IMC ’21. ACM, Nov

  46. [52]

    The attention econ- omy

    Filippo Menczer and Thomas Hills. “The attention econ- omy”. In: Scientific American 323.6 (2020), pp. 54–61

    2020

  47. [53]

    Inside the slammer worm

    David Moore et al. “Inside the slammer worm”. In: IEEE Security & Privacy 1.4 (2003), pp. 33–39

    2003

  48. [54]

    Using Dust Clouds to Enhance Anony- mous Communication

    Richard Mortier et al. “Using Dust Clouds to Enhance Anony- mous Communication”. In: Security Protocols XVIII . Ed. by Bruce Christianson and James Malcolm. Vol. 7061. Berlin, Heidelberg: Springer, Mar. 2014, pp. 54–59.doi: 10.1007/978- 3-662-45921-8\_10

    2014 doi

  49. [55]

    Microbial markets: socio-economic perspective in studying microbial communities

    Fariha Mostafa et al. “Microbial markets: socio-economic perspective in studying microbial communities”. In: micro- Life 5 (2024). doi: 10.1093/femsml/uqae016

    2024 doi

  50. [56]

    Spreading like wildfire

    “Spreading like wildfire”. In: Nature Climate Change 7.11 (Nov. 2017), pp. 755–755. doi: 10.1038/nclimate3432

    2017 doi

  51. [58]

    Mycorrhizal Markets, Firms, and Co-ops

    Ronald Noë and E. Toby Kiers. “Mycorrhizal Markets, Firms, and Co-ops”. In: Trends in Ecology and Evolution 33.10 (Oct. 2018), pp. 777–789. doi: 10.1016/j.tree.2018.07.007

    2018 doi

  52. [59]

    The Internet peering playbook: connecting to the core of the Internet

    William B Norton. The Internet peering playbook: connecting to the core of the Internet . DrPeering Press, 2011

    2011

  53. [60]

    Declining resilience of ecosystem func- tions under biodiversity loss

    Tom H. Oliver et al. “Declining resilience of ecosystem func- tions under biodiversity loss”. In: Nature Communications 6.1 (Dec. 2015). doi: 10.1038/ncomms10122

    2015 doi

  54. [61]

    Mod- ularizing Reasoning about AI Capabilities via Abstract Di- jkstra Monads

    Cyrus Omar, Patrick Ferris, and Anil Madhavapeddy. “Mod- ularizing Reasoning about AI Capabilities via Abstract Di- jkstra Monads”. In: the 12th ACM SIGPLAN Workshop on Higher-Order Programming with Effects . Sept. 2024. AAR 2025, August 18–22, 2025, Aarhus N, Denmark Madhavap...

    2024

  55. [62]

    Governing the Commons: The Evolution of In- stitutions for Collective Action

    Elinor Ostrom. Governing the Commons: The Evolution of In- stitutions for Collective Action . Cambridge University Press, Sept. 2015. doi: 10.1017/cbo9781316423936

    2015 doi

  56. [63]

    The Evolutionary Enigma of Sex

    Sarah P. Otto. “The Evolutionary Enigma of Sex”. In: The American Naturalist 174.S1 (July 2009), S1–S14. doi: 10 . 1086/599084

    2009

  57. [64]

    Arpanet (1969– 2019)

    Camille Paloque-Bergès and Valérie Schafer. “Arpanet (1969– 2019)”. In: Internet Histories 3.1 (2019), pp. 1–14

    2019

  58. [65]

    An overview of the immune system

    Jacqueline Parkin and Bryony Cohen. “An overview of the immune system”. In: The Lancet 357.9270 (2001), pp. 1777– 1789

    2001

  59. [66]

    European ash (Fraxinus excelsior) dieback–A conservation biology challenge

    Marco Pautasso et al. “European ash (Fraxinus excelsior) dieback–A conservation biology challenge”. In: Biological conservation 158 (2013), pp. 37–49

    2013

  60. [67]

    Defence against the dark artefacts: Smart home cybercrimes and cybersecurity standards

    Stanislaw Piasecki, Lachlan Urquhart, and Professor Derek McAuley. “Defence against the dark artefacts: Smart home cybercrimes and cybersecurity standards”. In: Computer Law and Security Review 42 (Sept. 2021), p. 105542. doi: 10.1016/j.clsr.2021.105542

    2021

  61. [68]

    Getting the measure of biodiversity

    Andy Purvis and Andy Hector. “Getting the measure of biodiversity”. In: Nature 405.6783 (May 2000), pp. 212–219. doi: 10.1038/35012221

    2000 doi

  62. [69]

    How hard can it be? Designing and implementing a deployable multipath TCP

    Costin Raiciu et al. “How hard can it be? Designing and implementing a deployable multipath TCP”. In:9th USENIX symposium on Networked Systems Design and Implementa- tion. 2012, pp. 399–412

    2012

  63. [70]

    Why We Die: The New Science of Age- ing and Longevity

    Venki Ramakrishnan. Why We Die: The New Science of Age- ing and Longevity. Hachette UK, 2024

    2024

  64. [71]

    Unikernel Linux (UKL)

    Ali Raza et al. “Unikernel Linux (UKL)”. In: Proceedings of the Eighteenth European Conference on Computer Systems . EuroSys ’23. Rome, Italy: Association for Computing Ma- chinery, 2023, pp. 590–605. doi: 10.1145/3552326.3587458

    2023

  65. [72]

    Information Exposure From Consumer IoT Devices: A Multidimensional, Network-Informed Mea- surement Approach

    Jingjing Ren et al. “Information Exposure From Consumer IoT Devices: A Multidimensional, Network-Informed Mea- surement Approach”. In:Proceedings of the Internet Measure- ment Conference. Association for Computing Machinery, 2019, pp. 267–279. doi: 10.1145/3355369.3355577

    2019

  66. [73]

    Hydra: Effective Run- time Network Verification

    Sundararajan Renganathan et al. “Hydra: Effective Run- time Network Verification”. In: Proceedings of the ACM SIG- COMM 2023 Conference. ACM SIGCOMM ’23. New York, NY, USA: Association for Computing Machinery, 2023, pp. 182–

    2023

  67. [74]

    Internet Architecture Evolution: Found in Translation

    Guilherme Ribeiro et al. “Internet Architecture Evolution: Found in Translation”. In:Proceedings of the 23rd ACM Work- shop on Hot Topics in Networks . 2024, pp. 300–307

    2024

  68. [75]

    Somatic Hypermutation

    Jana Ridani et al. “Somatic Hypermutation”. In: Molecular Biology of B Cells . Elsevier, 2024, pp. 235–256. doi: 10.1016/ b978-0-323-95895-0.00012-x

    2024

  69. [76]

    SibylFS: formal specification and oracle- based testing for POSIX and real-world file systems

    Tom Ridge et al. “SibylFS: formal specification and oracle- based testing for POSIX and real-world file systems”. In: Proceedings of the 25th Symposium on Operating Systems Principles. Monterey California: ACM, Oct. 2015, pp. 38–53. doi: 10.1145/2815400.2815411

    2015

  70. [77]

    Cheating and punishment in cooperative animal societies

    Christina Riehl and Megan E. Frederickson. “Cheating and punishment in cooperative animal societies”. In: Philosoph- ical Transactions of the Royal Society B: Biological Sciences 371.1687 (Feb. 2016). doi: 10.1098/rstb.2015.0090

    2016

  71. [78]

    Lost in the Edge: Finding Your Way with DNSSEC Signposts

    Charalampos Rotsos et al. “Lost in the Edge: Finding Your Way with DNSSEC Signposts”. In: 3rd USENIX Workshop on Free and Open Communications on the Internet (FOCI 13) . Washington, D.C.: USENIX Association, Aug. 2013

    2013

  72. [79]

    The discovery and innovation of AI does not qualify as creativity

    Mark A. Runco. “The discovery and innovation of AI does not qualify as creativity”. In:Journal of Cognitive Psychology (Dec. 2024), pp. 1–10. doi: 10.1080/20445911.2024.2436362

    2024

  73. [80]

    Pathways to mutualism break- down

    J SACHS and E SIMMS. “Pathways to mutualism break- down”. In: Trends in Ecology and Evolution 21.10 (Oct. 2006), pp. 585–592. doi: 10.1016/j.tree.2006.06.018

    2006 doi

  74. [81]

    PROSPER: Ex- tracting Protocol Specifications Using Large Language Mod- els

    Prakhar Sharma and Vinod Yegneswaran. “PROSPER: Ex- tracting Protocol Specifications Using Large Language Mod- els”. In: Proceedings of the 22nd ACM Workshop on Hot Topics in Networks. ACM, 2023. doi: 10.1145/3626111.3628205

    2023

  75. [82]

    Contextual and combinatorial structure in sperm whale vocalisations

    Pratyusha Sharma et al. “Contextual and combinatorial structure in sperm whale vocalisations”. In:Nature Commu- nications 15.1 (May 2024). doi: 10.1038/s41467-024-47221-8

    2024 doi

  76. [83]

    The economics of a trillion connected devices

    Philip Sparks. The economics of a trillion connected devices . 2017

    2017

  77. [84]

    How to own the internet in your spare time

    Stuart Staniford, Vern Paxson, and Nicholas Weaver. “How to own the internet in your spare time”. In: 11th USENIX Security Symposium. USENIX, 2002

    2002

  78. [85]

    Your botnet is my botnet: analysis of a botnet takeover

    Brett Stone-Gross et al. “Your botnet is my botnet: analysis of a botnet takeover”. In: Proceedings of the 16th ACM con- ference on Computer and Communications Security . ACM, Nov. 2009, pp. 635–647. doi: 10.1145/1653662.1653738

    2009

  79. [86]

    How Many Species of Insects and Other Terrestrial Arthropods Are There on Earth?

    Nigel E. Stork. “How Many Species of Insects and Other Terrestrial Arthropods Are There on Earth?” In: Annual Review of Entomology 63.1 (Jan. 2018), pp. 31–45. doi: 10. 1146/annurev-ento-020117-043348

    2018

  80. [87]

    The origin and evolution of mycorrhizal symbioses: from palaeomycology to phyloge- nomics

    Christine Strullu-Derrien et al. “The origin and evolution of mycorrhizal symbioses: from palaeomycology to phyloge- nomics”. In: New Phytologist 220.4 (Mar. 2018), pp. 1012–

    2018

  81. [88]

    How Decentralization Affects User Agency on Social Platforms

    Aditya Surve, Aneesh Shamraj, and Swapneel Mehta. How Decentralization Affects User Agency on Social Platforms . ICWSM, June 2024. doi: 10.36190/2024.74

    2024 doi

  82. [89]

    How mycorrhizal associations drive plant population and community biology

    Leho Tedersoo, Mohammad Bahram, and Martin Zobel. “How mycorrhizal associations drive plant population and community biology”. In: Science 367.6480 (Feb. 2020). doi: 10.1126/science.aba1223

    2020 doi

  83. [90]

    Understanding and applying biologi- cal resilience, from genes to ecosystems

    Rose Thorogood et al. “Understanding and applying biologi- cal resilience, from genes to ecosystems”. In:npj Biodiversity 2.1 (Aug. 2023). doi: 10.1038/s44185-023-00022-6

    2023 doi

  84. [91]

    Establishing a synthetic orthogonal replication system enables accelerated evolution in E. coli

    Rongzhen Tian et al. “Establishing a synthetic orthogonal replication system enables accelerated evolution in E. coli”. In: Science 383.6681 (Jan. 2024), pp. 421–426. doi: 10.1126/ science.adk1281

    2024

  85. [92]

    Phonetic and Lexical Discovery of Ca- nine Vocalization

    Theron S. Wang et al. “Phonetic and Lexical Discovery of Ca- nine Vocalization”. In:Findings of the Association for Compu- tational Linguistics: EMNLP 2024 . Ed. by Yaser Al-Onaizan, Mohit Bansal, and Yun-Nung Chen. Miami, Florida, USA: Association for Computational Linguistic...

    2024 doi

  86. [93]

    Consequences of environmental stress on oak: predisposition to pathogens

    PM Wargo. “Consequences of environmental stress on oak: predisposition to pathogens”. In:Annales des sciences forestières. Vol. 53. 2-3. EDP Sciences. 1996, pp. 359–368. Steps towards an Ecology for the Internet AAR 2025, August 18–22, 2025, Aarhus N, Denmark

    1996

  87. [94]

    CHERI: A hybrid capability- system architecture for scalable software compartmental- ization

    Robert NM Watson et al. “CHERI: A hybrid capability- system architecture for scalable software compartmental- ization”. In: 2015 IEEE Symposium on Security and Privacy . IEEE. 2015, pp. 20–37

    2015

  88. [95]

    Ten recent insights for our understand- ing of cooperation

    Stuart A. West et al. “Ten recent insights for our understand- ing of cooperation”. In: Nature Ecology and Evolution 5.4 (Jan. 2021), pp. 419–430. doi: 10.1038/s41559-020-01384-x

    2021 doi

  89. [96]

    Mycorrhizal Fungi Respond to Resource Inequality by Moving Phosphorus from Rich to Poor Patches across Networks

    Matthew D. Whiteside et al. “Mycorrhizal Fungi Respond to Resource Inequality by Moving Phosphorus from Rich to Poor Patches across Networks”. In: Current Biology 29.12 (June 2019), 2043–2050.e8. doi: 10.1016/j.cub.2019.04.061

    2019 doi

  90. [97]

    Wikipedia. Ecology. http://en.wikipedia.org/w/index.php? title=Ecology&oldid=1275647999. 2025

    2025

  91. [98]

    Internet

    Wikipedia. Internet. http://en.wikipedia.org/w/index.php? title=Internet&oldid=1275606545. 2025

    2025

  92. [99]

    TCP ex machina: computer-generated congestion control

    Keith Winstein and Hari Balakrishnan. “TCP ex machina: computer-generated congestion control”. In: Proceedings of the ACM SIGCOMM 2013 Conference on SIGCOMM . SIG- COMM ’13. Hong Kong, China: Association for Comput- ing Machinery, 2013, pp. 123–134. doi: 10.1145/2486001. 2486020

    2013 doi

  93. [100]

    Peter Wohlleben.The hidden life of trees: What they feel, how they communicate—Discoveries from a secret world . Vol. 1. Greystone Books, 2016

    2016

  94. [102]

    Examining the Centralization of Email Industry: A Landscape Analysis for IPv4 and IPv6

    Luciano Zembruzki et al. “Examining the Centralization of Email Industry: A Landscape Analysis for IPv4 and IPv6”. In: 2023 IEEE Symposium on Computers and Communications

    2023

  95. [103]

    Privacy-preserving activity and health monitoring on databox

    Yuchen Zhao et al. “Privacy-preserving activity and health monitoring on databox”. In: Proceedings of the Third ACM International Workshop on Edge Systems, Analytics and Net- working. 2020, pp. 49–54

    2020

  96. [194]

    doi: 10.1145/3603269.3604856

  97. [732]

    doi: 10.1126/science.1157719

    doi

  98. [1030]

    doi: 10.1111/nph.15076

    doi

  99. [2021]

    doi: 10.1145/3487552.3487821

  100. [2023]

    doi: 10.1109/ISCC58397.2023.10217834

    2023

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.