Yann Vonlanthen

We present the first longitudinal, cross-network measurement study of 36 public blockchain peer-to-peer (P2P) networks, spanning nine months from late 2024. We deployed 15 active crawlers with hourly connectivity probes, performed Internet-wide scans, and decomposed shared discovery substrates to identify 19 auxiliary networks. We find that active node counts span three orders of magnitude (10,000) and vary substantially in reachability, discovery coverage, and deployment structure. To characterize these differences, we quantify discovery effectiveness, churn dynamics, peer-table coverage, and AS- and geographic concentration (via the Herfindahl--Hirschman Index). Our results reveal pronounced heterogeneity in network structure and highlight fundamental observability limitations in widely deployed peer-discovery mechanisms. This extended abstract accompanies the full paper [10].

We present the first longitudinal, cross-network measurement study of 36 public blockchain peer-to-peer (P2P) networks, spanning nine months from late 2024. We deployed 15 active crawlers with hourly connectivity probes, performed Internet-wide scans, and decomposed shared discovery substrates to identify 19 auxiliary networks. We find that active node counts span three orders of magnitude (10,000) and vary substantially in reachability, discovery coverage, and deployment structure. To characterize these differences, we quantify discovery effectiveness, churn dynamics, peer-table coverage, and AS- and geographic concentration (via the Herfindahl--Hirschman Index). Our results reveal pronounced heterogeneity in network structure and highlight fundamental observability limitations in widely deployed peer-discovery mechanisms. This extended abstract accompanies the full paper [10].

Blockchain technologies underpin an expanding ecosystem of decentralized applications, financial systems, and infrastructure. However, the fundamental networking layer that sustains these systems, the peer-to-peer (P2P) layer, of all but the top few ecosystems remains largely opaque. In this paper, we present the first longitudinal, cross-network measurement study of 36 public blockchain networks. Over 9 months (since late 2024), we deployed 15 active crawlers, sourced data from two additional community crawlers, and conducted hourly connectivity probes (e.g., pings and protocol-level handshakes) to observe the evolving state of these networks. Furthermore, by leveraging Ethereum's discovery protocols, we inferred metadata for an additional 19 auxiliary networks that utilize the Ethereum peer discovery protocol. We also explored Internet-wide scans, which only require probing each protocol's default ports with a simple, network-specific payload. This approach allows us to rapidly identify responsive peers across the entire address space without having to implement custom discovery and handshake logic for every blockchain. We validated this method on Bitcoin and similar networks with known ground truth, then applied it to Cardano, which we could not crawl directly. Our study uncovers dramatic variation in network size from under 10 to more than 10,000 active nodes. We quantify trends in IPv4 versus IPv6 usage, analyze autonomous systems and geographic concentration, and characterize churn, diurnal behavior, and the coverage and redundancy of discovery protocols. These findings expose critical differences in network resilience, decentralization, and observability. Beyond characterizing each network, our methodology demonstrates a general framework for measuring decentralized networks at scale. This opens the door for continued monitoring, benchmarking, and more transparent assessments of blockchain infrastructure across diverse ecosystems.

Voting advice applications (VAAs) help millions of voters understand which political parties or candidates best align with their views. This paper explores the potential risks these applications pose to the democratic process when targeted by adversarial entities. In particular, we expose 11 manipulation strategies and measure their impact using data from Switzerland’s primary VAA, Smartvote, collected during the last two national elections. We find that altering application parameters, such as the matching method, can shift a party’s recommendation frequency by up to 105%. Cherry-picking questionnaire items can increase party recommendation frequency by over 261%, while subtle changes to parties’ or candidates’ responses can lead to a 248% increase. To address these vulnerabilities, we propose adversarial robustness properties VAAs should satisfy, introduce empirical metrics for assessing the resilience of various matching methods, and suggest possible avenues for research toward mitigating the effect of manipulation. Our framework is key to ensuring secure and reliable AI-based VAAs poised to emerge in the near future.

Voting advice applications (VAAs) help millions of voters understand which political parties or candidates best align with their views. This paper explores the potential risks these applications pose to the democratic process when targeted by adversarial entities. In particular, we expose 11 manipulation strategies and measure their impact using data from Switzerland's primary VAA, Smartvote, collected during the last two national elections. We find that altering application parameters, such as the matching method, can shift a party's recommendation frequency by up to 105%. Cherry-picking questionnaire items can increase party recommendation frequency by over 261%, while subtle changes to parties' or candidates' responses can lead to a 248% increase. To address these vulnerabilities, we propose adversarial robustness properties VAAs should satisfy, introduce empirical metrics for assessing the resilience of various matching methods, and suggest possible avenues for research toward mitigating the effect of manipulation. Our framework is key to ensuring secure and reliable AI-based VAAs poised to emerge in the near future.

This paper presents Banyan, the first rotating leader state machine replication (SMR) protocol that allows transactions to be confirmed in just a single round-trip time in the Byzantine fault tolerance (BFT) setting. Based on minimal alterations to the Internet Computer Consensus (ICC) protocol and with negligible communication overhead, we introduce a novel dual mode mechanism that enables optimal block finalization latency in the fast path. Crucially, the modes of operation are integrated, such that even if the fast path is not effective, no penalties are incurred. Moreover, our algorithm maintains the core attributes of the ICC protocol it is based on, including optimistic responsiveness and rotating leaders without the necessity for a view-change protocol. We prove the correctness of our protocol and provide an open-source implementation of it. Banyan is compared to its predecessor ICC, as well as other well known BFT protocols, in a globally distributed wide-area network. Our evaluation reveals that Banyan reduces latency by up to 30% compared to state-of-the-art protocols, without requiring additional security assumptions.

Ethereum introduced Transaction Access Lists (TALs) in 2020 to optimize gas costs during transaction execution. In this work, we present a comprehensive analysis of TALs in Ethereum, focusing on adoption, quality, and gas savings. Analyzing a full month of mainnet data with 31,954,474 transactions, we found that only 1.46% of transactions included a TAL, even though 42.6% of transactions would have benefited from it. On average, access lists can save around 0.29% of gas costs, equivalent to approximately 3,450 ETH (roughly US$ 5 Mio) per year. However, 19.6% of TALs included by transactions contained imperfections, causing almost 11.8% of transactions to pay more gas with TAL than without. We find that these inaccuracies are caused by the unknown state at the time of the TAL computation as well as imperfect TAL computations provided by all major Ethereum clients. We thus compare the gas savings when calculating the TAL at the beginning of the block vs. calculating it on the correct state, to find that the unknown state is a major source of TAL inaccuracies. Finally, we implement an ideal TAL computation for the Erigon client to highlight the cost of these flawed implementations.