{"id":"f4e1bf66-60c6-497a-a431-f27d40bb5f62","arxiv_id":"1908.00418","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"high","formal_verification":"none","parameter_count":4,"one_line_summary":"A multi-identifier internet architecture with blockchain-based governance and a hash-prefix-tree forwarding table is proposed and prototyped, but scaling claims exceed the measured 20-node, 3.5-billion-entry tests.","lead":"MIN is a proposed internet architecture that lets identity, content, location, and IP addresses coexist and be managed jointly by many organizations through a blockchain. The authors built a prototype on Chinese operators' networks and report that it can register identifiers and stream video, though the scalability claims outpace the experiments.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"APoV throughput claim is internally inconsistent: Eq. (7) predicts ~111k tx/s at n=20, not >300k; with fixed resources Eq. (12) throughput collapses as ~1/n^2, so hundreds/thousands-node scalability is unsupported.","rationale":"The reader's REJECT verdict is appropriate. The paper presents a coherent architecture and a real prototype, with deployment over operator networks, insertion/deletion algorithms for HPT-FIB, tunnel scenarios, and VoD rates; these give some support to feasibility. However, the load-bearing quantitative claims are the management-plane throughput at scale and the FIB capacity. I focused on APoV because the abstract's 'hundreds or thousands of nodes with high throughput' is the central co-governing claim, and the paper contains a concrete, checkable inconsistency at n=20: Eq. (7) predicts 1.80 s per consensus round, whereas the claimed >300k tx/s would require 0.667 s. Since the paper gives no 20-node raw data, no error bars, and no code, the reader cannot tell which is wrong. Eq. (12) also shows throughput decays with n when resources are fixed, so large-n claims require explicit resource-scaling assumptions that are absent. This is not a disagreement with consensus; it is an internal consistency and support problem. A single reproduction of the 20-node experiment, plus a statement of Figure 7's assumptions, would settle it. The FIB extrapolation from 3.5B to 'tens of billions' is also a gap, but it is secondary because the management-plane contradiction is already decisive for the REJECT verdict.","tokens_in":14911,"tokens_out":11753,"duration_ms":113586,"concrete_test":"Run the APoV benchmark at n=20 under the Table 1 parameter set (K=10,000; each node both consortium and bookkeeping node; 1 Gbps links; Xeon Silver 4114) and report the average per-round consensus time. Compare it to Eq. (7)'s prediction of 1.80 s and to the claimed >300k tx/s requirement of <=0.667 s per round. Also list the a and band values used for Figure 7 and evaluate Eq. (12) at n=100 and n=1000: if the figure assumes a or band grows superlinearly with n, that assumption must be stated explicitly and justified by the deployment plan.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing pillar is the abstract's promise that MIN's management plane can be co-governed by 'hundreds or thousands of nodes with high throughput.' Section 4.1 fits the consensus time from 3-8 node experiments as Eq. (7): t_cons=(0.0312n^3-0.1920n^2+2.0714n+11.25)/125. At n=20 this predicts 1.80 s per round, which with K=10,000 gives 111k tx/s. Yet the same section states that a measured 20-blockchain-node prototype 'can achieve a stable throughput of more than 300 thousand transactions per second,' which requires at most 0.667 s per round. The paper does not report the 20-node raw times, so the reader cannot tell whether Eq. (7) is invalid beyond n=8 or the 20-node run used different parameters. Either way, one of the two central management-plane claims is unsupported. This is not merely a concern about extrapolation: Eq. (12) has consensus time growing as O(n^3) for fixed a and band, so high throughput at hundreds or thousands of nodes depends on a and band growing superlinearly with n; the paper never states or justifies such scaling, and Figure 7's axes and parameter ranges are not specified in the text. Since the management plane is what makes MIN 'co-governing,' this unresolved inconsistency is decisive for the REJECT verdict.","agreement_with_reader":"partial"},"referee_report":null,"author_rebuttal":null,"desk_editor":null,"rs_alignment":null,"lean_confirmation":null,"pith_extraction":null,"created_at":"2026-08-14T15:57:38.159537+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":null,"supporting_citations":[],"review_version":1}