pith. sign in

arxiv: 2606.05823 · v1 · pith:TMAMEIC5new · submitted 2026-06-04 · 🧮 math.CO · math.AG· math.RT

Frayed Demazure weaves for Poisson-compatible cluster structures on Bott--Samelson charts

Pith reviewed 2026-06-28 00:57 UTC · model grok-4.3

classification 🧮 math.CO math.AGmath.RT
keywords Demazure weavesBott-Samelson varietiescluster structuresPoisson structuresquasi-cluster morphismsbraid varietiesaffine charts
0
0 comments X

The pith

Adding frayed strands to Demazure weaves produces Poisson-compatible cluster structures on further affine charts of the Bott-Samelson variety, with chart transitions becoming rational quasi-cluster morphisms.

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

Demazure weaves serve as combinatorial representations of maps between Bott-Samelson cells and already yield cluster structures on braid varieties. The paper establishes that these maps and structures are compatible with the standard Poisson structure on the Bott-Samelson variety. Introducing frayed strands extends the construction to additional affine charts. The resulting transition functions between charts become rational quasi-cluster morphisms, and the associated mutation sequences relate closely to those appearing in work on open Richardson seeds. A reader would care because the extension equips more of the variety with cluster data that respects the Poisson geometry.

Core claim

Demazure weaves represent maps between Bott-Samelson cells and construct cluster structures on braid varieties that remain compatible with the standard Poisson structure. Adding frayed strands produces Poisson-compatible cluster structures on other affine charts of the Bott-Samelson variety. The transition functions across these charts become rational quasi-cluster morphisms, and the mutation sequences constructed for these morphisms are closely related to those of Ménard for open Richardson seeds.

What carries the argument

Frayed Demazure weaves, which are ordinary Demazure weaves augmented by frayed strands, that represent maps preserving Poisson compatibility while turning chart transitions into rational quasi-cluster morphisms.

If this is right

  • Cluster structures already known on braid varieties extend to other affine charts while preserving Poisson compatibility.
  • Transition functions between the new charts and the original ones become rational quasi-cluster morphisms.
  • The mutation sequences needed for these morphisms are the same as those appearing for open Richardson seeds.
  • The construction supplies Poisson-compatible cluster data on a larger collection of affine charts than previously available.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Covering the whole Bott-Samelson variety by such charts could produce a global cluster structure respecting the Poisson form.
  • The fraying technique may apply to other varieties equipped with Poisson structures and combinatorial maps similar to Demazure weaves.
  • The relation to open Richardson seeds suggests possible overlap or unification between two families of cluster constructions on related spaces.

Load-bearing premise

The standard Poisson structure on the Bott-Samelson variety remains compatible with the maps given by frayed Demazure weaves so that cluster algebra relations are preserved after the fraying step.

What would settle it

An explicit computation on a concrete pair of charts showing that the transition map induced by a frayed Demazure weave fails to be a rational quasi-cluster morphism, or that the Poisson bracket is not preserved, would disprove the claim.

Figures

Figures reproduced from arXiv: 2606.05823 by Jon Cheah.

Figure 1
Figure 1. Figure 1: The three types of vertices in classical Demazure weaves of simply-laced type. We keep to the convention in [7] of using blue, red, and green to denote edges labelled i, j, k where i and j are adjacent roots, and i and k are not adjacent [PITH_FULL_IMAGE:figures/full_fig_p014_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: shows an expanded set of permitted vertices for simply-laced type. These will be further expanded upon in Appendix A, but the arguments in this section will only require these initial six. i i i j i j k i j k i −i i −j i j −k i i −k i i −i j −i [PITH_FULL_IMAGE:figures/full_fig_p024_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Factoring a classical trivalent vertex as a fraying followed by a twining. In [PITH_FULL_IMAGE:figures/full_fig_p024_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Equivalent frayed weaves from (i, k, −i) to (i, k). z1z3 − z2 z1 − z4 z3 z1 z1 − z4 z2 − z3z4 z3 z1 − z4 z2 − z3z4 z3 z1 z2 z3 z4 z4 z1 z2 z3 z4 ∼ [PITH_FULL_IMAGE:figures/full_fig_p026_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Equivalent frayed weaves from (i, j, i, −j) to (i, j, i). Corollary 5.5. In simply-laced type, we can apply leftward moves as described above to get from the signed expression (w0, −i) to (i ∗ , −i ∗ , v), where v = w0si and i ∗ is such that si ∗ = w0siw0 [PITH_FULL_IMAGE:figures/full_fig_p026_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: A frayed weave (w0, 2, 3, 1, −2, 2, −1) → (w0, 2, 3, 1, 2) repre￾senting the map ρ : O(2,3,1,−2,2,−1) → O(2,3,1,2). For convenience, we’ve chosen the reduced word w0 = (2, 1, 3, 2, 1, 3) for the longest element to minimise braid moves. The map does not depend on the choice of word for w0 nor on the coordinates for those strands. Note that how strand labels propagate in the non-simply-laced case depends on … view at source ↗
Figure 7
Figure 7. Figure 7: Frayed weaves in type A3 and the octavalent vertices they fold to in type B2. Example 5.20. In type B2, with Cartan matrix  2 −1 −2 2  , consider the signed ex￾pression γ = (1, 2, 1, −2, 2), so that γ + = (1, 2, 1, 2). In the Bott–Samelson coordinates z1, . . . , z4 for Oγ + , the exchange matrix E(γ +) and seed s(γ +) are:   0 2 −1 0 −1 0 1 −1 1 −2 0 1 0 1 − 1 2 0   and z2 z2z4 − z 2 3 z1 z1z3 −… view at source ↗
Figure 8
Figure 8. Figure 8: The types of vertices allowed for frayed weaves in simply-laced type, and their effect on the strand labellings. As before, blue and red strands represent adjacent roots i and j, and green strands to denote a root k not adjacent to i. The vertices in the first two rows correspond to isomorphisms between charts. Those in the third row correspond to maps which are invertible on an open set given by the non-v… view at source ↗
Figure 9
Figure 9. Figure 9: Frayed weaves in type A3 and the octavalent vertices they fold to in type B2. Example A.10. In type B2, recall the seed s0 = s(1, 2, 1, 2) which was shown in Example 5.20. The mutable part of the valued quiver is of mutation type B2, and its exchange graph consists of six distinct clusters arranged in a cycle. We list these in the [PITH_FULL_IMAGE:figures/full_fig_p048_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Two equivalent frayed weaves from (1, −1, −1) to (1). B.3. Isotopy of raking rays. In [6, §5] and [7, §5.1.5.], it is checked that the relative heights of trivalent vertices does not affect the labellings on weaves (or the configuration of flags the weave represents). Specifically, the labellings are preserved when hexavalent and trivalent weave vertices are moved by isotopy past a raking ray. The same st… view at source ↗
Figure 11
Figure 11. Figure 11: Eight equivalences arising from moving a fraying vertex past a tetravalent or hexavalent vertex. For the hexavalent vertices in the bottom row that do not correspond to isomorphisms, note that they shares the same indeterminacy locus as that of the fraying vertex above it [PITH_FULL_IMAGE:figures/full_fig_p053_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: When the variable at the top of the first strand [PITH_FULL_IMAGE:figures/full_fig_p055_12.png] view at source ↗
Figure 12
Figure 12. Figure 12: Embeddings of weave moduli for other isotopies of fraying vertices [PITH_FULL_IMAGE:figures/full_fig_p056_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: Mutation of Demazure weaves from [7] compared against the frayed weaves arising from commuting the order of fraying vertices. Conjecture B.6. For any word w, the frayed weave consisting only of fraying vertices wrev : (i1, i2, . . . , iℓ−1, iℓ) → (i1, i2, . . . , iℓ−1, −iℓ) → (i1, i2, . . . , −iℓ−1, −iℓ), → · · · → (i1, −i2, . . . , −iℓ−1, −iℓ) → (i1, −i2, . . . , −iℓ−1, −iℓ), corresponding to the reverse… view at source ↗
read the original abstract

Demazure weaves are combinatorial representations of maps between Bott--Samelson cells and have been used to construct cluster structures on braid varieties. We show the compatibility of these maps and the resulting cluster structures with the standard Poisson structure on the Bott--Samelson variety. Adding frayed strands to Demazure weaves, we further construct Poisson compatible cluster structures on other affine charts of the Bott--Samelson variety in a manner that transition functions across charts become rational quasi-cluster. The mutation sequences we construct for these quasi-cluster morphisms are closely related to those of M\'enard for open Richardson seeds.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript claims that Demazure weaves are compatible with the standard Poisson structure on the Bott--Samelson variety and that adding frayed strands produces Poisson-compatible cluster structures on additional affine charts, with transition functions across charts becoming rational quasi-cluster morphisms. The constructed mutation sequences are stated to be closely related to those of Ménard for open Richardson seeds.

Significance. If the central claims hold, the work extends prior combinatorial constructions of cluster structures from braid varieties to a wider collection of affine charts on the Bott--Samelson variety while preserving Poisson compatibility and producing explicit rational quasi-cluster transition maps. The combinatorial (rather than data-fitted) definition of the frayed objects and the explicit link to Ménard’s mutation sequences are strengths that support potential applications in Poisson geometry and cluster algebra theory on flag varieties.

major comments (2)
  1. [§3] §3 (definition of frayed strands): the claim that fraying preserves Poisson compatibility and yields quasi-cluster morphisms after mutation is load-bearing for the extension beyond braid-variety charts; the manuscript must supply an explicit verification that the added strands do not alter the Poisson bracket relations in a way that breaks the cluster algebra structure, rather than relying solely on the combinatorial extension of the weave.
  2. [§5] §5 (mutation sequences for quasi-cluster morphisms): the asserted close relation to Ménard’s sequences for open Richardson seeds is central to the transition-function claim; a concrete comparison (e.g., explicit sequence of mutations or a table of corresponding exchanges) is required to confirm that the sequences indeed produce rational quasi-cluster maps rather than merely analogous combinatorics.
minor comments (2)
  1. [Introduction] The introduction would benefit from an early, self-contained statement of the main theorem (including the precise statement of Poisson compatibility) before the technical definitions.
  2. [§2] Notation for the standard Poisson structure on the Bott--Samelson variety should be fixed once in §2 and used consistently thereafter to avoid ambiguity when discussing compatibility.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive suggestions. Both major comments identify places where explicit verification would strengthen the manuscript; we agree and will incorporate the requested material in the revision.

read point-by-point responses
  1. Referee: [§3] §3 (definition of frayed strands): the claim that fraying preserves Poisson compatibility and yields quasi-cluster morphisms after mutation is load-bearing for the extension beyond braid-variety charts; the manuscript must supply an explicit verification that the added strands do not alter the Poisson bracket relations in a way that breaks the cluster algebra structure, rather than relying solely on the combinatorial extension of the weave.

    Authors: We agree that an explicit verification of Poisson compatibility for the frayed strands is required. In the revised manuscript we will add, in §3, a direct computation of the Poisson brackets on the new variables introduced by fraying. The calculation shows that these variables behave as frozen variables whose brackets are compatible with the existing cluster algebra structure on the braid-variety charts, thereby confirming that the combinatorial extension preserves the Poisson relations. revision: yes

  2. Referee: [§5] §5 (mutation sequences for quasi-cluster morphisms): the asserted close relation to Ménard’s sequences for open Richardson seeds is central to the transition-function claim; a concrete comparison (e.g., explicit sequence of mutations or a table of corresponding exchanges) is required to confirm that the sequences indeed produce rational quasi-cluster maps rather than merely analogous combinatorics.

    Authors: We accept that a concrete side-by-side comparison is needed. The revised §5 will contain an explicit table listing the mutation sequences arising from our frayed Demazure weaves together with the corresponding sequences from Ménard’s construction for open Richardson seeds. The table will highlight the precise exchange relations that guarantee the transition maps are rational quasi-cluster morphisms. revision: yes

Circularity Check

0 steps flagged

No significant circularity identified

full rationale

The paper defines frayed Demazure weaves combinatorially as an extension of prior Demazure weave constructions for braid varieties, then shows compatibility with the standard Poisson structure on the Bott-Samelson variety and constructs transition functions as rational quasi-cluster morphisms. The mutation sequences are stated to be closely related to Ménard's (distinct author) work on open Richardson seeds. No equation or definition reduces by construction to its own inputs, no fitted parameter is relabeled as a prediction, and no load-bearing premise rests solely on self-citation. The derivation chain remains self-contained against external combinatorial and Poisson-geometric benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The paper relies on the existence of a standard Poisson structure on the Bott-Samelson variety and on the prior combinatorial definition of Demazure weaves; no new free parameters or invented entities are introduced in the abstract.

axioms (2)
  • domain assumption The Bott-Samelson variety carries a standard Poisson structure compatible with its cell decomposition.
    Invoked when claiming compatibility of the cluster structures with the Poisson bracket.
  • domain assumption Demazure weaves represent maps between Bott-Samelson cells that can be extended by frayed strands while preserving cluster relations.
    Central to the construction of the new quasi-cluster morphisms.

pith-pipeline@v0.9.1-grok · 5624 in / 1471 out tokens · 21671 ms · 2026-06-28T00:57:57.913173+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

38 extracted references · 21 canonical work pages

  1. [1]

    Upper cluster structure on Kac–Moody Richardson varieties.arXiv preprint arXiv:2506.10382, 2025

    Huanchen Bao and Jeff York Ye. Upper cluster structure on Kac–Moody Richardson varieties.arXiv preprint arXiv:2506.10382, 2025

  2. [2]

    Quantum cluster algebras.Advances in Mathematics, 195(2):405–455, 2005

    Arkady Berenstein and Andrei Zelevinsky. Quantum cluster algebras.Advances in Mathematics, 195(2):405–455, 2005. doi: https://doi.org/10.1016/j.aim.2004.08.003

  3. [3]

    Fomin, and Andrei Zelevinsky

    Arkady Berenstein, Sergey. Fomin, and Andrei Zelevinsky. Cluster algebras III: Upper bounds and double Bruhat cells.Duke Mathematical Journal, 126(1):1–52, 2005. doi: 10.1215/S0012-7094-04-12611-9

  4. [4]

    On Leclerc’s conjectural cluster structures for open Richardson varieties.arXiv preprint arXiv:2207.10184, 2022

    Peigen Cao and Bernhard Keller. On Leclerc’s conjectural cluster structures for open Richardson varieties.arXiv preprint arXiv:2207.10184, 2022

  5. [5]

    Legendrian weaves:N-graph calculus, flag moduli and applications.Geometry & Topology, 26(8):3589–3745, 2022

    Roger Casals and Eric Zaslow. Legendrian weaves:N-graph calculus, flag moduli and applications.Geometry & Topology, 26(8):3589–3745, 2022. doi: 10.2140/gt.2022.26. 3589. 58 JON CHEAH

  6. [6]

    Algebraic weaves and braid varieties.American Journal of Mathematics, 146(6):1469–1576, 2024

    Roger Casals, Eugene Gorsky, Mikhail Gorsky, and Jos´ e Simental. Algebraic weaves and braid varieties.American Journal of Mathematics, 146(6):1469–1576, 2024. doi: 10.1353/ajm.2024.a944357

  7. [7]

    Cluster structures on braid varieties.Journal of the American Mathematical Society, 38(2):369–479, 2025

    Roger Casals, Eugene Gorsky, Mikhail Gorsky, Ian Le, Linhui Shen, and Jos´ e Si- mental. Cluster structures on braid varieties.Journal of the American Mathematical Society, 38(2):369–479, 2025. doi: 10.1090/jams/1048

  8. [8]

    Sherman- Bennett, and Jos´ e Simental

    Roger Casals, Pavel Galashin, Mikhail Gorsky, Linhui Shen, Melissa. Sherman- Bennett, and Jos´ e Simental. Comparing cluster algebras on braid varieties.arXiv preprint arXiv:2508.03816, 2026

  9. [9]

    Positroid links and braid varieties.Journal f ur die reine und angewandte Mathematik (Crelles Journal),

    Roger Casals, Eugene Gorsky, Mikhail Gorsky, and Jos´ e Simental. Positroid links and braid varieties.Journal f ur die reine und angewandte Mathematik (Crelles Journal),

  10. [10]

    doi: doi:10.1515/crelle-2026-0007

  11. [11]

    Bott–Samelson varieties and Poisson Ore extensions

    Bal´ azs Elek and Jiang-Hua Lu. Bott–Samelson varieties and Poisson Ore extensions. International Mathematics Research Notices, 2021(14):10745–10797, 07 2019. doi: 10.1093/imrn/rnz127

  12. [12]

    Brick manifolds and toric varieties of brick polytopes.The Electronic Journal of Combinatorics, 23(P2.25), 2016

    Laura Escobar. Brick manifolds and toric varieties of brick polytopes.The Electronic Journal of Combinatorics, 23(P2.25), 2016. doi: 10.37236/5038

  13. [13]

    V. V. Fock and A. B. Goncharov. Cluster ensembles, quantization and the diloga- rithm.Annales scientifiques de l’ ´Ecole Normale Sup´ erieure, Ser. 4, 42(6):865–930,

  14. [14]

    doi: 10.24033/asens.2112

  15. [15]

    Fomin, L

    S. Fomin, L. Williams, and A. Zelevinsky. Introduction to Cluster Algebras. chapter 6.arXiv preprint arXiv:2008.09189, 2020

  16. [16]

    Double bruhat cells and total positiv- ity.Journal of the American Mathematical Society, 12(2):335–380, 1999

    Sergey Fomin and Andrei Zelevinsky. Double bruhat cells and total positiv- ity.Journal of the American Mathematical Society, 12(2):335–380, 1999. doi: 10.1090/S0894-0347-99-00295-7

  17. [17]

    Cluster algebras I: Foundations.Jour- nal of the American Mathematical Society, 15(2):497–529, 2002

    Sergey Fomin and Andrei Zelevinsky. Cluster algebras I: Foundations.Jour- nal of the American Mathematical Society, 15(2):497–529, 2002. doi: 10.1090/ S0894-0347-01-00385-X

  18. [18]

    Quasi-homomorphisms of cluster algebras.Advances in Applied Math- ematics, 81:40–77, 2016

    Chris Fraser. Quasi-homomorphisms of cluster algebras.Advances in Applied Math- ematics, 81:40–77, 2016. doi: 10.1016/j.aam.2016.06.005

  19. [19]

    Braid variety cluster structures, I: 3D plabic graphs.arXiv preprint arXiv:2210.04778, 2024

    Pavel Galashin, Thomas Lam, Melissa Sherman-Bennett, and David Speyer. Braid variety cluster structures, I: 3D plabic graphs.arXiv preprint arXiv:2210.04778, 2024

  20. [20]

    Braid variety cluster structures, II: general type.Inventiones mathematicae, 243:1079–1127, 2026

    Pavel Galashin, Thomas Lam, and Melissa Sherman-Bennett. Braid variety cluster structures, II: general type.Inventiones mathematicae, 243:1079–1127, 2026. doi: 10.1007/s00222-025-01390-5

  21. [21]

    F. A. Garside. The braid group and other groups.The Quarterly Journal of Mathe- matics, 20(1):235–254, 1969. doi: 10.1093/qmath/20.1.235

  22. [22]

    Kac–Moody groups and cluster algebras.Advances in Mathematics, 228(1):329–433, 2011

    Christof Geiß, Benard Leclerc, and Jan Schr¨ oer. Kac–Moody groups and cluster algebras.Advances in Mathematics, 228(1):329–433, 2011. doi: 10.1016/j.aim.2011. 05.011

  23. [23]

    Cluster algebras and Poisson geometry.Moscow Mathematical Journal, 3(3):899–934, 2003

    Michael Gekhtman, Michael Shapiro, and Alek Vainshtein. Cluster algebras and Poisson geometry.Moscow Mathematical Journal, 3(3):899–934, 2003. doi: 10.17323/ 1609-4514-2003-3-3-899-934

  24. [24]

    the American Mathematical Society, 2010

    Michael Gekhtman, Michael Shapiro, and Alek Vainshtein.Cluster algebras and Pois- son geometry, volume 167 ofMathematical Surveys and Monographs. the American Mathematical Society, 2010. FRAYED DEMAZURE WEAVES 59

  25. [25]

    A generalized cluster struc- ture onGL n via birational Poisson maps.arXiv preprint arXiv:03.12486, 2026

    Misha Gekhtman, Michael Shapiro, and Alek Vainshtein. A generalized cluster struc- ture onGL n via birational Poisson maps.arXiv preprint arXiv:03.12486, 2026

  26. [26]

    Regular pullback of gener- alized cluster structures.arXiv preprint arXiv:2603.12496, 2026

    Misha Gekhtman, Michael Shapiro, and Alek Vainshtein. Regular pullback of gener- alized cluster structures.arXiv preprint arXiv:2603.12496, 2026

  27. [27]

    K. R. Goodearl and M. Yakimov. Poisson structures on affine spaces and flag varieties. II.Transactions of the American Mathematical Society, 361(11):5753–5780, 2009. doi: 10.1090/S0002-9947-09-04654-6

  28. [28]

    K. R. Goodearl and M. T. Yakimov. Cluster algebra structures on Poisson nilpotent algebras.arXiv preprint arXiv:1801.01963, 2018

  29. [29]

    B. Leclerc. Cluster structures on strata of flag varieties.Advances in Mathematics, 300:190–228, 2016. doi: 10.1016/j.aim.2016.03.018

  30. [30]

    Mutation matrices for Poisson CGL extensions

    Jiang-Hua Lu and Yipeng Mi. Mutation matrices for Poisson CGL extensions. 2026. In preparation

  31. [31]

    B. R. Marsh and K. Rietsch. Parametrizations of flag varieties.Representation Theory, 8:212–242, 2004. doi: 10.1090/S1088-4165-04-00230-4

  32. [32]

    Cluster algebras associated with open Richardson varieties: an algorithm to compute initial seeds.arXiv preprint arXiv:2201.10292, 2022

    Etienne M´ enard. Cluster algebras associated with open Richardson varieties: an algorithm to compute initial seeds.arXiv preprint arXiv:2201.10292, 2022

  33. [33]

    Locally acyclic cluster algebras.Advances in Mathematics, 233(1): 207–247, 2013

    Greg Muller. Locally acyclic cluster algebras.Advances in Mathematics, 233(1): 207–247, 2013. doi: 10.1016/j.aim.2012.10.002

  34. [34]

    Rietsch and L

    K. Rietsch and L. Williams. Newton–Okounkov bodies, cluster duality, and mirror symmetry for Grassmannians.Duke Mathematical Journal, 168(18):3437–3527, 2019. doi: 10.1215/00127094-2019-0028

  35. [35]

    A superpotential for Grassmannian Schubert varieties.arXiv preprint arXiv:2409.00734, 2025

    Konstanze Rietsch and Lauren Williams. A superpotential for Grassmannian Schubert varieties.arXiv preprint arXiv:2409.00734, 2025

  36. [36]

    Shen and Daping Weng

    Linhui. Shen and Daping Weng. Cluster structures on double Bott-–Samelson cells. Forum of Mathematics, Sigma, 9:e66 1–89, 2021. doi: 10.1017/fms.2021.59

  37. [37]

    J. R. Stembridge. On the fully commutative elements of Coxeter groups.Journal of Algebraic Combinatorics, 5:353–385, 1996. doi: 10.1007/BF00193185

  38. [38]

    Maximal green sequences for quantum and Poisson CGL extensions

    Milen Yakimov. Maximal green sequences for quantum and Poisson CGL extensions. arXiv preprint arXiv:2603.14376, 2026. Department of Mathematics, The University of Hong Kong, Pokfulam Road, Hong Kong Email address:joncheah@connect.hku.hk