{"id":"5eedd369-3fe6-44d9-b6a3-39aa9200b2b2","arxiv_id":"1908.05395","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"By allowing dihedral angles of zero, Gott introduces envelope polyhedra, a new class of regular polyhedral surfaces with many finite and infinite examples.","lead":"This paper defines a new class of polyhedra called envelope polyhedra, where the same arrangement of regular polygons meets at every vertex but some faces are glued back-to-back. A generalist might read it to see a new family of geometric surfaces that bridges classical polyhedra and spongelike structures.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The informal regularity definition makes the {4,14} vertex-transitivity claim unverifiable; an explicit coordinate check is needed.","rationale":"After a good-faith reading, the finite constructions are sound and the paper deserves credit for concrete models and correct Euler characteristics for the main examples. The serious weakness is definitional: 'identical arrangement' is a term of art that is doing all the work, and the paper supplies no certificate or algorithm for checking it. This is precisely the reader's weakest assumption, so I agree with the CONDITIONAL verdict. I do not see a reason to reject: the main finite example is explicit and counts are internally consistent, and the infinite constructions are plausible and may be checkable. The proposed coordinate test would convert the conditional acceptance into a verified existence proof for at least the paper's headline infinite example.","tokens_in":17117,"tokens_out":28516,"duration_ms":298979,"concrete_test":"Generate an explicit periodic coordinate model of the Section 5 {4,14} construction: take a regular triangular tiling, erect right triangular prisms whose height equals the triangle side, delete top and bottom triangles, and insert the east-west back-to-back square fins in the described dashed pattern; repeat vertically. Then compute the link of every vertex as a cyclic list of the incident face sectors, each labeled by face polygon, side (exterior/interior/fin side), and the dihedral angle to the next sector. Verify (i) every edge has exactly two incident sectors (single surface, no free edges) and (ii) all vertex links are isomorphic as labeled cyclic sequences.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing point is not the Euler arithmetic: the finite {4,6} example in Section 4 is a coherent double of a sphere with eight holes, and its counts check out. The load-bearing point is the unformalized regularity condition in Section 3: 'the arrangement of polygons (creating a single surface) around each vertex must be identical' is used to certify every example, but the paper never defines the object that must be identical. It does not say whether the vertex figure is a cyclic sequence of face sectors, whether side labels (exterior/interior) and dihedral angles are part of the data, whether back-to-back coincident faces are modeled as a double cover, or whether equivalence is combinatorial, metric, or ambient-isometric. This matters most for the flagship infinite construction {4,14} in Section 5, which is described only verbally ('add east-west fins ... in a dashed pattern') and has no coordinates or diagram from which the claimed 14-square cyclic arrangement at every vertex can be checked. If some vertices receive fins differently, or if the fins create a second surface component or a free edge, the object is not a regular envelope polyhedron under even the paper's own informal rule. The same gap propagates through the appendix catalog, where many entries are referenced only to WBK page numbers.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a class of 'envelope polyhedra': polyhedra made of regular polygons whose vertex figures are all identical, but whose dihedral angles may vary and may be 0 degrees, meaning paired faces lie back to back. The main finite example is squares–6 around a point {4,6}, obtained by deleting the triangular faces of a rhombicuboctahedron to produce a hollow polyhedron with 36 square faces, 72 edges, 24 vertices, and genus 7. Further examples include octagons–4 around a point and decagons–4 around a point, and the paper claims infinite examples such as squares–14 around a point and triangles–18 around a point. The Appendix catalogs many additional examples, many obtained by deleting polygons from previously published pseudopolyhedra, and Section 7 introduces a separate subclass with mirror-image vertex figures.","tokens_in":17370,"tokens_out":2753,"duration_ms":31172,"significance":"If the constructions are valid, the paper enlarges the classical taxonomy of regular polyhedra in an interesting direction, connecting to the historical work of Coxeter and Petrie, Gott, Wells, and Wachman-Burt-Kleinmann. The finite examples are genuinely checkable: the Euler characteristic computation for {4,6} (F=36, E=72, V=24 gives genus 7) and the decagon example (genus 19) are consistent, and the paper honestly credits prior discoveries. However, the central notion of regularity for envelope polyhedra is never formalized, and the infinite flagship examples are described only verbally, so the claimed classification in Table 1 cannot currently be independently verified. The paper offers no machine-checked proofs or explicit coordinates; its value depends on a definition and verification standard that are not supplied.","major_comments":[{"comment":"The defining condition for a regular envelope polyhedron is informal. The phrase 'the arrangement of polygons (creating a single surface) around each vertex must be identical' does not specify what data constitute the arrangement: is it the cyclic order of face sectors, the list of interior angles, the dihedral angles, the distinction between interior and exterior sides, or some quotient where back-to-back coincident faces are identified? The paper also does not state whether vertex figures are considered up to combinatorial isomorphism, metric congruence, or ambient isometry. This is load-bearing because every example and every row of Table 1 is certified by appeal to this condition.","section":"Section 3"},{"comment":"The flagship infinite construction {4,14} is described only in words: 'add east-west fins of back-to-back squares above this single layer of cells in a dashed pattern' and 'these fins are then connected to another plane of squares–12 around a point triangular cell layer above it.' No coordinates, no diagram, and no explicit vertex-by-vertex accounting are provided, so the claim that every vertex has the same cyclic 14-square configuration cannot be checked. In particular, the construction must rule out free edges, additional surface components, and vertices where the fins attach differently, none of which is demonstrated. A formal construction with coordinates or an unambiguous polyhedral complex is needed before {4,14} can be accepted as an envelope polyhedron under the paper's own definition.","section":"Section 5, squares–14 around a point"},{"comment":"The catalog in the Appendix relies extensively on page references to Wachman, Burt, and Kleinmann (1974) without reproducing the starting structures or verifying the deletion process. For example, 'Squares–8 around a point (1): Start with the semi-regular pseudopolyhedron shown by WBK on page 9 at the top' gives the reader no way to check that deleting the triangles yields a single non-self-intersecting surface with all vertices congruent. Since Table 1 claims many entries on the basis of these constructions, the paper needs either a general lemma stating sufficient conditions under which deleting a specified set of faces from a WBK pseudopolyhedron yields an envelope polyhedron, or a case-by-case verification for every listed entry. Without this, the completeness and correctness of Table 1 are not established.","section":"Appendix A and Appendix B"},{"comment":"The mirror-vertex subclass changes the equivalence relation for vertex figures, but the paper does not define when a vertex figure and its mirror image are to count as 'congruent' in three dimensions. Section 7 says some structures have 'pairs of mirror vertices whose vertex figures are not identical except under mirror reflection,' yet it also argues that mirror-image vertices cannot be superimposed in 3D. This is a different regularity notion from the one used for Section 3 examples and Table 1, and the paper needs to state explicitly whether mirror congruence is considered admissible in the main definition or only in a separate weaker class. The current text leaves the boundary between 'identical vertices' and 'mirror vertices' unclear, which affects the interpretation of the entire classification.","section":"Section 7"}],"minor_comments":[{"comment":"The reference to Coxeter is misspelled as 'Coexter' in the bibliography; the spelling should be corrected.","section":"References"},{"comment":"Some LaTeX artifacts remain, such as 'Schl¨ afli' and 'Pellicer and Shulte'; these should be cleaned up before publication.","section":"Throughout"},{"comment":"The figure for squares–14 around a point is cited as the visual evidence for the construction, but the text does not explain how the photograph encodes the 'dashed pattern' of fins or how the fin placement is uniform at every vertex. Adding a schematic diagram with the fins highlighted would help.","section":"Figure 5"},{"comment":"The table uses symbols (⋄, ◦, •, +) with a legend below, but the legend is easy to miss because it appears after the table; moving the legend into the caption would improve readability.","section":"Table 1"}],"recommendation":"major_revision","confidential_remarks":"This is an informal, historically flavored geometry paper whose finite examples are largely sound but whose central definition and infinite constructions need substantial formalization. The revision burden is significant but probably within the scope of a rewrite: supply a rigorous definition of vertex-figure congruence for envelope polyhedra, give an explicit construction for at least the {4,14} example, and either prove a general deletion lemma or verify the appendix entries. If the venue is primarily expository, the editor may weigh whether the informal style is acceptable; for a research journal, the current level of rigor is insufficient for acceptance."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This paper introduces a new class of regular polyhedral surfaces—envelope polyhedra—by allowing dihedral angles of 0 degrees while keeping the cyclic arrangement of faces at each vertex identical. That is a genuine extension beyond dihedra, pseudopolyhedra, and the Coxeter–Petrie skew polyhedra, and it produces some elegant new examples. The constructions are self-contained and the Euler arithmetic for the main finite examples is correct: the {4,6} shell from deleting triangles from the rhombicuboctahedron has F=36, E=72, V=24, genus 7, and the truncated-cube {8,4} and decagon examples match their stated genera. The paper is also historically careful, crediting Coxeter, Petrie, Wells, WBK, and the author's own 1967 pseudopolyhedra without overstating priority. The soft spot is the working definition of regularity. \"The arrangement of polygons touching each vertex must be identical\" is never made precise. The paper does not say whether the vertex figure is a cyclic list of face sectors, whether interior/exterior side labels or dihedral angles are part of the data, or how back-to-back coincident faces are treated topologically. As a result, the claim that each deletion construction yields a single non-self-intersecting surface is unverified. This matters most for the flagship infinite {4,14}, which is described only verbally (\"add east-west fins in a dashed pattern\") with no coordinates, diagram, or symmetry description. I cannot confirm from the text that every vertex has the same 14-square cyclic order or that the surface is one connected component. The appendix lists dozens of additional examples, many referenced only to WBK page numbers, and the instructions are often too loose to be checked without the original book in hand. A few smaller overstatements also creep in: calling 1260 degrees the \"most found for any polyhedron so far\" implies an exhaustive search that is not presented. The historical anecdotes are pleasant but do not affect the mathematics. The paper is aimed at geometers and polyhedra enthusiasts, and it will be of real interest to that community. With a formal definition of vertex congruence and computational verification of at least the infinite cases and the appendix, it could become a solid reference. As it stands, it deserves a serious referee rather than a desk rejection—but the referee should send it back for a definition and verification the reader can actually check.","headline":"Envelope polyhedra are a real but modest extension of the regular polyhedron family; the finite examples check out, but the informal definition leaves the infinite flagship examples unverified.","tokens_in":741,"tokens_out":1848,"would_cite":false,"duration_ms":37796,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["52B10","51M20"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper defines envelope polyhedra, a new class of regular polyhedra with identical vertex arrangements but variable, sometimes zero, dihedral angles.","keywords":["envelope polyhedra","regular polyhedra","dihedral angles","pseudopolyhedra","infinite polyhedra","vertex figures","polyhedral genus","Schläfli symbols"],"falsifier":"Build the rhombicuboctahedron with its eight triangular faces removed and trace, at every one of the 24 vertices, the circuit of an ant tethered to the vertex; if any vertex visits fewer or more than six square faces, the cyclic order differs between two vertices, or the surface self-intersects, the squares-6 example fails. The same check, together with explicit face, edge, and vertex counts and the Euler characteristic, can be applied to every row of Table 1.","tokens_in":16921,"feed_emoji":"🔷","tokens_out":9480,"duration_ms":88576,"temperature":0.7,"pith_summary":"Envelope polyhedra are proposed as a new class of regular polyhedra: every face is a regular polygon and the arrangement of polygons around each vertex is the same, but dihedral angles between faces may differ, including dropping to zero so that faces lie back to back. The central examples are finite, such as squares-6 around a point made by deleting the triangular faces from a rhombicuboctahedron, and infinite, such as squares-14 around a point and triangles-18 around a point. If the definition is accepted, it substantially widens the known universe of regular polyhedral surfaces, fitting between the classical Platonic solids and the author's earlier infinite pseudopolyhedra under one rule.","feed_headline":"Zero-degree folds create a new class of regular polyhedra","feed_subtitle":"Identical vertices, variable dihedral angles, and back-to-back faces yield finite and infinite new forms.","key_machinery":"The central object is the envelope polyhedron, defined by identical vertex arrangements while allowing dihedral angles of 0 degrees. The generative machinery is the deletion construction: remove a chosen set of faces from a known regular, semiregular, or pseudopolyhedral network so that those positions become holes joining the exterior surface to the interior surface. Around each resulting vertex, the ant's circuit picks up both sides of the remaining faces, doubling the face count. The Euler characteristic identity $F-E+V=2(1-g)$ is then used to compute genus for finite examples, and the sum of face angles around a vertex places each structure in the positive, zero, or negative curvature class.","core_discovery":"The paper claims that there is a larger class of regular polyhedra than the standard lists, obtained by keeping the requirement that an ant tethered to a vertex encounters the same sequence of polygon faces while circling it, but dropping the requirement of equal dihedral angles and allowing some to be exactly zero. Deleting selected faces from a polyhedron or pseudopolyhedron opens holes that connect exterior and interior faces, so around each vertex the ant's circuit doubles. The flagship finite example is squares-6 around a point: remove the eight triangles from a rhombicuboctahedron, obtaining a hollow polyhedron with 36 square faces, 72 edges, 24 vertices, and genus 7. Infinite examples include squares-14 around a point, with 1260 degrees of angle around each vertex, and triangles-18 around a point; the appendix records many more, including structures whose vertices are congruent only by mirror reflection.","pith_inferences":["The paper leaves vertex congruence informal; a precise version, requiring the same cyclic order of face angles at every vertex up to rotation or reflection, would turn the table into a testable classification and would show which mirror-vertex entries belong in the same list.","The same delete-faces-to-open-holes recipe should generalize to higher dimensions, producing envelope polytopes with back-to-back facets from regular polytopes in four or more dimensions.","Because these surfaces concentrate large negative curvature at vertices, they are natural candidates for lightweight lattice structures or for simplified topological models of porous media, an application the paper leaves for the future."],"forward_implications":["Finite envelope polyhedra such as squares-6 around a point give explicit all-regular-polygon models of negatively curved, multiply connected surfaces, with genus 7.","Infinite envelope polyhedra extend the pseudopolyhedron catalog to higher per-vertex angle sums, with squares-14 around a point reaching 1260 degrees around each vertex.","The deletion recipe gives a systematic way to generate new regular polyhedral surfaces from any known regular, semiregular, or pseudopolyhedral network.","Envelope polyhedra with exactly 360 degrees around each vertex, such as squares-4 around a point, connect the planar tessellations to finite toroidal polyhedra under the same definition.","If mirror-vertex structures are admitted as regular, the total list of known regular polyhedral surfaces grows further, as organized in the paper's Table 1."],"supporting_citations":[{"why":"Supplies the prior regular skew polyhedra and mentions dihedra, setting the regularity standard that envelope polyhedra relax.","marker":"Coxeter, 1937"},{"why":"Introduces pseudopolyhedra and the angle-sum curvature categories, and provides starting networks such as triangles-10 around a point for later deletions.","marker":"Gott, 1967"},{"why":"Provides the semiregular infinite polyhedra, the N-M nomenclature, and most of the page-by-page starting structures from which faces are deleted.","marker":"Wachman, Burt, and Kleinmann (WBK)"},{"why":"Supplies additional pseudopolyhedral starting structures, including snub-cube arrays, used in the appendix constructions.","marker":"Wells (1977)"}],"fun_headline_variants":["Zero-angle folds: new regular polyhedra","Back-to-back faces build hollow regular polyhedra","New regular polyhedra from zero dihedral folds","Envelope polyhedra: regular, with zero-degree folds"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the phrase 'the arrangement of polygons around each vertex is identical' is precise enough to recognize by inspection, and that every deletion listed in the appendix yields one continuous, non-self-intersecting surface with exactly the stated counts.","fun_headline_variants_meta":{"raw":{"variants":["Zero-angle folds: new regular polyhedra","Back-to-back faces build hollow regular polyhedra","New regular polyhedra from zero dihedral folds","Envelope polyhedra: regular, with zero-degree folds"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000228,"raw_usage":{"total_tokens":1459,"prompt_tokens":911,"completion_tokens":548,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":527,"completion_tokens_details":{"reasoning_tokens":484}},"tokens_in":527,"tokens_out":548,"duration_ms":5614,"temperature":1.0,"reasoning_tokens":484,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:14:48.267307+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Build the rhombicuboctahedron with its eight triangular faces removed and trace, at every one of the 24 vertices, the circuit of an ant tethered to the vertex; if any vertex visits fewer or more than six square faces, the cyclic order differs between two vertices, or the surface self-intersects, the squares-6 example fails. The same check, together with explicit face, edge, and vertex counts and the Euler characteristic, can be applied to every row of Table 1.","supporting_citations":[],"review_version":1}