| paper | d4-journals-geb-2017-0124 |
| authors | — |
| venue | ? |
| filed under | unclassified |
| judged by | gpt-6-luna / xhigh (triple__luna__xhigh__c6r1) |
| judge confidence | high |
| authors would recognise it | unclear |
The paper’s named theorems and corollary are axiomatic characterizations; none asserts an algorithm, complexity bound, or computational hardness. The repeated-hauler sketch suggests a possible new continuous evaluation setting, but gives no concrete mirror of a named computational result, so bit (a) fails.
fails bit a — no named computational result to mirror
No results are mirrored; the axiomatic characterizations and the Spanish network illustration remain outside an anchored computational mirror.
The strongest positive intuition is a repeated-hauler regime: a large modular gas network with many small operators occupying a small number of standardized network roles. A type would have to include an operator’s complete role in the network—its owned pipes, flows, volumes, and path position. Mass could represent the fraction of haulers of each role, and the output could be the total loss allocated to each type under one of the paper’s rules. This is a plausible extension of the paper’s allocation problem, though much less convincing for its Spanish example, which has seven haulers and one dominant operator.
I cannot make that into a valid anchored continuous mirror under your criteria. The paper contains no named computational result asserting tractability, hardness, or another complexity classification. Theorem 1 says the edge’s rule is the unique rule satisfying independence of edge sectioning, independence by sales, and flow proportionality on edges; Theorem 2 characterizes the tracing rules using independence of unused edges, independence by sales, volume proportionality on paths, and tracing additivity. Corollary 1 specializes that characterization to the proportional tracing rule, and Theorem 3 characterizes that rule using independence of edge multiplication. These are axiomatic uniqueness results, proved in the paper, rather than complexity results.
The formulas for the rules and the Spanish-network calculations do not supply a numbered complexity anchor either. Turning rule evaluation into a continuous computational problem would therefore be a new question, not a continuous version of a named computational result in this paper. The proposed repeated-hauler scenario also requires a replicated network structure; encoding the full topology in each type may leave little separation between the number of haulers and the number of types.
So there is a conceivable population-continuum extension here, but no anchor-specific problem I can honestly hand to a prover or classify in the programme’s trichotomy. The weakness is central: the paper’s named results are axiomatic, precisely the kind of result the programme excludes as a target for continuization.
The repeated-hauler regime is the strongest rescue, and I would grant that a network made of many standardized modules could have a sensible mass distribution over hauler roles. The type would need to include the hauler’s complete role: its pipes, flows, volumes, and position in the network. But granting that model does not supply a computational result from this paper to mirror.
Theorem 1, Theorem 2, Corollary 1, and Theorem 3 characterize allocation rules by axioms. They do not analyze a decision, search, or optimization problem, or claim a complexity classification. Recasting the axioms for masses of hauler roles would still be an axiomatic question. A measure-zero issue with individual sales or mergers can be repaired by defining those changes for positive mass; that makes the model more coherent, but does not turn the characterizations into computational results.
One could ask how to compute a rule’s allocation on a large network. That is a newly posed evaluation problem, not a mirror of a result the paper proves. The rules are given by explicit sums, and the proportional-tracing rule uses flow decomposition on the paper’s acyclic network. A more ambitious task—such as choosing a rule to optimize a fairness criterion—would change the subject from analyzing the stated rules to designing a new one. Either could be a separate project, but neither is anchored by the paper’s named results.
So the negative case is strong under the programme’s stated gate: even the best repeated-hauler scenario yields no anchor-specific computational question here. Its limit is that it cannot establish that no later, newly designed computational problem about loss allocation would be worthwhile.
The adversarial triple: the proponent anchors on up to three named results; the opponent sees that case and must defeat every anchor; the judge decides which case convinced it. These are the pipeline’s own outputs, generated by tools/triple_run.py — no human edited them. The paper’s own text is not reproduced here beyond the quoted statement above.