Positional Notation as a Native Ultrametric Geometry of Nested Cycles

#1. Introduction

Digital systems, scientific simulations, and calendar‑processing software all rely on positional notation to encode discrete quantities. Engineers and programmers confront systematic inefficiencies: (i) clustering of temporally close dates is evaluated with Euclidean day‑count distances, which ignore the shared‑prefix structure of mixed‑radix calendar strings; (ii) error‑correction codes designed for linear bit strings overlook the hierarchical relationships among bits, leading to sub‑optimal logical error rates; (iii) data‑compression pipelines treat trailing zeros as placeholders rather than as informative hierarchical markers, inflating storage costs. The magnitude of these inefficiencies in any specific system is not quantified here; we treat the practical benefit of the ultrametric view as a hypothesis to be tested (Section 8). If positional notation were treated as an ultrametric tree, clustering would respect shared prefixes, error‑correction could exploit subtree confinement, and compression could compress redundant hierarchical patterns.

Contributions.

1. Isomorphism theorem – a formal proof that positional notation in base $b\ge2$ induces the $b$‑adic ultrametric $d_b(x,y)=b^{-v_b(x-y)}$ (Section 3).
2. Metric derivation – explicit calculation of ultrametric distances for representative integer pairs and for Gregorian calendar dates, including a $31\times$ separation between month‑ and day‑level distances (Section 4).
3. Closed‑form consequences – derivation of the expected number of shared trailing digits $\mathbb{E}[v_b]=1/(b-1)$ and of ultrametric ball cardinalities $|B_{b^{-k}}|=b^{\,n-k}$ (Section 4).
4. Error‑suppression analysis – derivation of the failure probability $p^{(1)}=0.1736704$ for an eight‑fold majority‑vote scheme at physical error rate $p=0.4$, and computation of the suppression factor $0.4/p^{(1)}\approx2.303$ (Section 4).
5. Discussion of limitations and falsification criteria (Section 7).
6. Practical pathway for software and quantum‑hardware designers, including readiness assessment and next‑step experiment (Section 8).

#2. Background and Related Work

Ultrametrics are defined by the strong triangle inequality $d(x,y)\le\max\{d(x,z),d(z,y)\}$ (Section 1 of [1]). In Hilbert geometry, a non‑Euclidean distance $d_U(p,q)=\frac12\ln\frac{\|q-q_1\|}{\|q-p_1\|}\times\frac{\|p-p_1\|}{\|p-q_1\|}$ is presented in Section 2.2 of [2]. Ultrametric ideas have also been carried into quantum error correction: [11] develops qudit quantum error correction on an ultrametric foundation, and [12] analyzes quantum error correction on Bruhat–Tits trees. The majority‑vote analysis in Section 4 of the present paper is our own derivation and does not reuse numerical suppression factors from either work. The counting of ultrametric balls used later in this paper is derived from first principles in Section 4.3 rather than borrowed from the literature; to our knowledge no cited work combines the $b$‑adic valuation metric with mixed‑radix calendar structure, which is the specific gap this paper fills.

#3. Methods

#3.1 Positional Notation as a Tree

For a fixed base $b\ge2$, any non‑negative integer $x$ admits a unique expansion

where $L$ is the length of the digit string. Interpreting the string $(a_{L-1}\dots a_0)$ as a path from the root of a $b$‑ary tree to a leaf yields a natural nested‑cycle hierarchy: each digit $a_i$ marks a branch at depth $i$.

#3.2 $b$‑adic Valuation

Define $v_b(n)$ as the largest integer $k\ge0$ such that $b^{k}\mid n$. The $b$‑adic distance between $x$ and $y$ is

If $x$ and $y$ share exactly $k$ trailing digits, then $b^{k}\mid(x-y)$ but $b^{k+1}\nmid(x-y)$, so $v_b(x-y)=k$ and $d_b(x,y)=b^{-k}$.

#3.3 Proof of Isomorphism (Theorem 1)

Lemma 1. Two integers share $k$ trailing digits iff $b^{k}\mid(x-y)$ and $b^{k+1}\nmid(x-y)$. Proof. Write $x$ and $y$ in base $b$:

Equality of the $k$ least‑significant digits means $a_i=b_i$ for $i=0,\dots,k-1$, i.e. $x-y$ is a multiple of $b^{k}$. If the $k$‑th digit differs, $b^{k+1}\nmid(x-y)$. ∎

Theorem 1 (Native Ultrametric of Positional Notation). In any base $b\ge2$, the distance $d_b$ defined above satisfies the strong triangle inequality and therefore is an ultrametric on $\mathbb{Z}$. Proof. Let $x,y,z\in\mathbb{Z}$ and let $k_{xy}=v_b(x-y)$, $k_{yz}=v_b(y-z)$. The shared‑prefix property of the tree implies that $x$ and $z$ share at least $\min\{k_{xy},k_{yz}\}$ trailing digits, hence $v_b(x-z)\ge\min\{k_{xy},k_{yz}\}$. Therefore

Thus $d_b$ is an ultrametric. ∎

#3.4 Majority‑Vote Error Model

Consider a binary error model where each physical qubit fails independently with probability $p$. An eight‑fold majority‑vote decoder declares the logical bit as the majority of eight physical bits. The logical failure probability $p^{(1)}$ is the probability that at least five physical bits fail:

We evaluate this expression for $p=0.4$ in Section 4.

#4. Analysis

#4.1 Ultrametric Distance Example

Take base $b=10$ and the integers $x=1200$, $y=1250$. Their difference is

The absolute value $|x-y|=50$ is divisible by $10$ but not by $10^{2}=100$, so $v_{10}(50)=1$. Applying the distance formula,

Thus the two numbers share exactly one trailing digit (the final 0) and differ in the tens digit, hence $v_{10}(50)=1$ and $d_{10}=10^{-1}=0.1$.

#4.2 Majority‑Vote Failure Probability

Set $p=0.4$, $q=1-p=0.6$, $n=8$.

Compute each binomial term:

1. $k=5$: $\displaystyle \binom{8}{5}=56,\; p^{5}=0.4^{5}=0.01024,\; q^{3}=0.6^{3}=0.216$ $\displaystyle \text{Term}_{5}=56\times0.01024\times0.216=0.12386304.$

1. $k=6$: $\displaystyle \binom{8}{6}=28,\; p^{6}=0.4^{6}=0.004096,\; q^{2}=0.6^{2}=0.36$ $\displaystyle \text{Term}_{6}=28\times0.004096\times0.36=0.04128768.$

1. $k=7$: $\displaystyle \binom{8}{7}=8,\; p^{7}=0.4^{7}=0.0016384,\; q^{1}=0.6$ $\displaystyle \text{Term}_{7}=8\times0.0016384\times0.6=0.00786432.$

1. $k=8$: $\displaystyle \binom{8}{8}=1,\; p^{8}=0.4^{8}=0.00065536$ $\displaystyle \text{Term}_{8}=0.00065536.$

Summing:

The suppression factor relative to the raw physical error rate is

#4.3 Closed‑Form Consequences for Uniform Numerals

#4.3.1 Expected Shared Trailing Digits

Let $x$ and $y$ be independent uniformly random integers modulo $b^{n}$. For any $k\ge1$,

The expected valuation is

Thus the expected number of shared trailing digits equals $1/(b-1)$ (exactly $1$ in binary, $1/9\approx0.1111$ in decimal). This derivation follows directly from the uniformity argument and the geometric series.

#4.3.2 Ultrametric Ball Cardinalities

Consider the finite set $\{0,1,\dots,b^{n}-1\}$ equipped with $d_b$. A ball of radius $b^{-k}$ consists of all points sharing at least $k$ trailing digits with the centre, i.e. all integers congruent modulo $b^{k}$. There are exactly $b^{n-k}$ such integers. Hence

Conversely, the number of distinct balls of radius $b^{-k}$ is $b^{k}$, because each residue class modulo $b^{k}$ defines a distinct centre.

#4.4 Mixed‑Radix Calendar Ultrametric

A Gregorian date can be represented as the mixed‑radix tuple $(\text{year},\text{month},\text{day})$ with level sizes $L_1=12$ (months per year) and $L_2=31$ (maximal month length). Using the weight choice $w_i=\bigl(\prod_{j=1}^{i}L_j\bigr)^{-1}$ yields

Worked examples (all in 2024, a leap year) give:

• Same month, different day: $d_{\mathrm{mr}}=1/372\approx0.00268817$.
• Same year, different month: $d_{\mathrm{mr}}=1/12\approx0.08333333$.
• Different year: $d_{\mathrm{mr}}=1$.

The separation ratio between month‑ and day‑level distances is

exactly the maximal month length, confirming that the ultrametric encodes the calendar cycle structure.

#5. Results

1. Ultrametric distance between $1200$ and $1250$ in base $10$ is $d_{10}=0.1$ (Section 4.1).
2. Logical failure probability for an eight‑fold majority‑vote decoder at $p=0.4$ is $p^{(1)}=0.1736704$ (Section 4.2).
3. Error‑suppression factor is $2.303$ (Section 4.2).
4. Expected shared trailing digits for uniform $b$‑ary numerals is $1/(b-1)$ (Section 4.3).
5. Ultrametric ball size of radius $b^{-k}$ in a $b^{n}$‑element domain is $b^{\,n-k}$ (Section 4.3).
6. Calendar ultrametric separates month‑ and day‑level distances by a factor of $31$ (Section 4.4).

All quantitative outcomes follow directly from the definitions and arithmetic displayed in Section 4.

#6. Extended Derivations and Worked Examples

In this section we deepen the quantitative picture introduced in Sections 4–5. We first derive the full probability distribution of the $b$‑adic valuation $v_b(x-y)$ for two independent uniform integers, then compute its variance and a confidence interval for the expected number of shared trailing digits. Next we relate the ultrametric ball cardinalities derived in §4.3 to the classical Cantor‑set construction of [7], showing that the same counting argument appears in both contexts. Afterwards we present a mixed‑radix calendar example that incorporates the irregular month length of February, demonstrating how the separation factor changes when the radix is not constant. Finally we discuss two algebraic extensions—multiplicative transposition of numerals [4] and quaternionic encoding of digit paths [5]—that illustrate how the ultrametric tree can be enriched while preserving its strong‑triangle‑inequality property. Throughout we draw analogies to symmetry reduction in geometric contexts [3] and to Noether‑type invariance under base change [8].

#1. Full distribution of the $b$‑adic valuation

Let $X$ and $Y$ be independent random variables uniformly drawn from $\{0,1,\dots,b^{n}-1\}$. For any integer $k\ge0$ define the event

Because $X\equiv Y\pmod{b^{k}}$ iff $b^{k}\mid (X-Y)$, the uniformity of $X$ and $Y$ gives

The probability that the valuation equals exactly $k$ is

Equation (2) is a geometric distribution with success probability $1/b$. The mean and variance follow from standard formulas:

Thus the variance equals the mean for every base $b\ge2$. For binary ($b=2$) we obtain $\mathbb{E}[v_2]=1$ and $\operatorname{Var}(v_2)=1$; for decimal ($b=10$) we obtain $\mathbb{E}[v_{10}=0.111\ldots$) and $\operatorname{Var}(v_{10})=0.111\ldots$.

A $95\%$ confidence interval for the empirical mean $\bar v$ of $m$ independent pairs follows from the normal approximation:

For $b=10$ and $m=10^{4}$ this yields

#2. Ultrametric balls and Cantor‑set cardinalities

Section 4.3 showed that a ball of radius $b^{-k}$ in the finite set $\{0,\dots,b^{n}-1\}$ contains exactly $b^{\,n-k}$ points. The same counting argument appears in the construction of a $b$‑ary Cantor set, as described in [7] where “Cantor sets are known to carry a natural ultrametric structure.” Consider the iterative removal process that at each stage partitions an interval into $b$ equal sub‑intervals and discards all but one. After $k$ stages the remaining set consists of $b^{k}$ disjoint intervals, each of length $b^{-k}$. The set of left‑hand endpoints of these intervals is precisely the set of integers congruent modulo $b^{k}$, i.e. a ball of radius $b^{-k}$ in the $b$‑adic ultrametric. Consequently the cardinality of a level‑$k$ Cantor approximation equals $b^{k}$, matching the number of distinct ultrametric balls of radius $b^{-k}$ in the finite domain. This duality explains why the ball‑size formula $|B_{b^{-k}}|=b^{\,n-k}$ can be interpreted both as a counting of integer residues and as a measure‑theoretic property of a self‑similar ultrametric fractal.

#3. Mixed‑radix calendar with irregular month lengths

The mixed‑radix model in §4.4 assumed a uniform month size $L_{1}=12$ and a maximal day size $L_{2}=31$, yielding a separation factor of $31$. Real Gregorian calendars contain February with $28$ days (or $29$ in leap years). To incorporate this irregularity we define a radix vector

where the second component is a list of day‑lengths for each month in a non‑leap year. The weight for the day level becomes month‑dependent:

where $L_{2}^{(m)}$ is the length of month $m$. For February ($m=2$) we have $L_{2}^{(2)}=28$, giving

The month‑level weight remains $w_{\text{month}}=1/12\approx0.08333333$. The separation ratio for February is therefore

exactly the number of days in February. For a 31‑day month the ratio is $31$, as previously derived. Thus the ultrametric automatically adapts to the actual radix of each sub‑cycle, preserving the interpretation that the distance between two dates reflects the size of the smallest differing cycle.

#4. Multiplicative transposition of numerals

Transposition of musical frequencies is performed by multiplying each member of a set by a single factor; this operation is formalized in [4], which states that “transposition of music is carried out by multiplying each member of a set of frequencies by a single frequency.” In the ultrametric tree, a numeral $x$ corresponds to a path $(a_{L-1}\dots a_{0})$ from the root. Multiplying $x$ by a constant $c$ can be interpreted as a tree‑automorphism that relabels each leaf according to the map

where $\oplus$ denotes digitwise addition modulo $b$ and $\text{code}(c)$ is the base‑$b$ expansion of $c$. Because digitwise addition preserves the depth of the lowest common ancestor, the $b$‑adic distance is invariant under $\phi_c$:

Thus multiplicative transposition is a symmetry of the ultrametric space, echoing the Noether‑type correspondence between symmetries and conserved quantities cited in [8]. The conserved quantity here is the ultrametric distance, which remains unchanged under the group of digitwise translations.

#5. Quaternionic encoding of digit paths

Quaternions are introduced in [5] as hyper‑complex numbers $q=x_{0}+x_{1}i+x_{2}j+x_{3}k$ with $x_{\mu}\in\mathbb{R}$. We can embed a $b$‑ary digit path of length $L$ into a quaternion by interpreting each digit $a_{i}\in\{0,\dots,b-1\}$ as a coefficient of a basis element in a four‑dimensional vector:

where $(e_{0},e_{1},e_{2},e_{3})=(1,i,j,k)$. The norm $|q_{\text{path}}|$ then aggregates the digit magnitudes across the path. Because the quaternion norm satisfies $|pq|=|p|\,|q|$, concatenation of two paths corresponds to quaternion multiplication:

The $b$‑adic valuation of the difference of two integers can be expressed as the exponent of the highest power of $b$ dividing the scalar part of $q_{\text{path}_1}-q_{\text{path}_2}$. This provides an alternative algebraic viewpoint on the ultrametric that is compatible with the quaternionic formalism of [5].

#6. Symmetry reduction and tree simplification

The use of symmetry reduction to obtain Einstein metrics on odd‑dimensional spheres is reported in [3] (“symmetry reduction yields Einstein metrics on compact odd dimensional manifolds”). Analogously, the $b$‑adic ultrametric exploits the inherent symmetry of the $b$‑ary tree: all branches at a given depth are interchangeable, and the distance depends only on the depth of the lowest common ancestor. By quotienting the full tree by the action of the symmetric group $S_{b^{k}}$ at each level $k$, one obtains a reduced model in which each equivalence class corresponds to a ball of radius $b^{-k}$. The reduction preserves the strong‑triangle‑inequality because the metric is defined solely by the depth of the common ancestor, which is invariant under permutations of sibling subtrees. This formal analogy justifies the claim that the ultrametric tree is the natural “symmetry‑reduced” representation of positional numerals.

#7. Noether‑type invariance under base change

Noether’s theorem, as described in [8], establishes a correspondence between continuous symmetries and conserved quantities. In the present context the continuous symmetry is the freedom to choose the base $b$ as a parameter of the representation, while the conserved quantity is the ultrametric distance after appropriate rescaling. Formally, let $b\mapsto b'$ be a change of radix and define the scaling factor $\sigma=b'/b$. The transformed distance is

If we simultaneously rescale the valuation by $\sigma^{-1}$, i.e. set $v_{b'}(x-y)=\sigma^{-1}v_{b}(x-y)$ (which holds for integers when $b'$ is a power of $b$), then

Thus the distance changes only by a predictable multiplicative factor, a conserved quantity under the symmetry of base rescaling. This invariance mirrors the Noether correspondence cited in [8].

#8. Summary of extended results

These derivations reinforce the central claim that positional notation possesses a native ultrametric geometry, that this geometry is robust under algebraic transformations (multiplicative transposition, quaternionic encoding), and that it aligns with broader mathematical structures such as Cantor sets, symmetry‑reduced manifolds, and Noetherian invariants. The quantitative formulas above are all obtained directly from the definitions and arithmetic displayed in Section 4, without recourse to external data.

#7. Discussion

#7.1 Limitations

• Base dependence (L1). The ultrametric $d_b$ requires an explicit base $b$; systems that implicitly assume base 10 must be re‑engineered to expose $b$ as a configurable parameter.
• Independence assumption (L2). The majority‑vote analysis assumes independent physical errors; correlated noise would invalidate the binomial model.
• Finite‑precision effects (L3). Implementations must handle overflow and rounding, which may introduce deviations from the ideal ultrametric.
• Applicability scope (L4). The ultrametric is most relevant for discrete integer‑valued data; continuous measurements require discretization before the metric can be applied.
• Scope limitation (L5). The derivations cover integer domains and fixed mixed‑radix hierarchies; extensions to continuous domains or adaptively learned hierarchies remain untested and are listed as open questions in Section 7.3.

#7.2 Falsification Criterion

If empirical clustering of calendar dates using the shared‑prefix ultrametric does not yield a statistically significant reduction in intra‑cluster variance compared with Euclidean day‑count clustering (e.g., paired $t$‑test $p\ge0.01$ on a representative dataset), the central claim that the ultrametric better captures phenomenological groupings would be falsified.

#7.3 Open Questions

• Interaction between the ultrametric and existing run‑length or dictionary compression schemes.
• Integration of the shared‑prefix distance into kernel functions for machine‑learning models on temporally structured data.
• Extension of the mixed‑radix ultrametric to calendars with irregular month lengths (e.g., lunar calendars) and its impact on clustering quality.

#8. Practical Implications and Engineering Pathways

#(a) Intended Users and Design Change

Software engineers developing date‑handling libraries, data‑compression toolchains, and quantum‑error‑correction firmware can replace linear distance measures with the $b$‑adic ultrametric. In calendar applications, clustering functions would use $d_{b}$ instead of absolute day differences, yielding groups that respect shared month‑day prefixes. In quantum hardware, decoder algorithms can prioritize syndrome patterns that respect subtree confinement, as illustrated by the majority‑vote analysis.

#(b) Technology Readiness

The work is purely analytical and has not undergone peer review; according to NASA’s Technology Readiness Level definitions, it resides at TRL 2 (technology concept formulated).

#(c) Minimal Next Experiment

Implement a prototype date‑clustering module that computes $d_{b}$ for mixed‑radix Gregorian dates and compare intra‑cluster variance against a baseline Euclidean day‑count clustering on a public holiday dataset (e.g., U.S. federal holidays). The experiment falsifies the claim if the variance reduction is not statistically significant ($p\ge0.05$).

#(d) Engineering Constraints

• Explicit base handling. All software must expose the base $b$ as a configurable parameter; legacy systems hard‑coded to decimal require refactoring.
• Integer representation limits. For very large integers, 64‑bit overflow may truncate tree depth; arbitrary‑precision libraries are needed.
• Error‑model fidelity. The majority‑vote analysis assumes independent Bernoulli errors; hardware designers must verify this for their qubit technology.
• Regulatory compliance. Calendar‑related software in financial contexts may be subject to standards prescribing Gregorian day counts; adopting $d_{b}$ would require regulatory approval.

#(e) Projected Cost/Performance Bounds (Projection)

Assuming dates are stored as base‑$b$ digit arrays without padding, a projection based on the expected number of shared trailing digits $\mathbb{E}[v_b]=1/(b-1)$ derived in Section 4.3 suggests that prefix‑aware encoding can eliminate on average $1/(b-1)$ redundant trailing digits per numeral pair; for $b=10$ this is $1/9\approx0.111$ digits per pair (projection; actual savings depend on the dataset's prefix structure and are not measured here). The logical error rate after one level of ultrametric‑aware majority voting is projected to be $p^{(1)}=0.1736704$ for $p=0.4$, yielding the suppression factor $0.4/p^{(1)}\approx2.303$ derived in Section 4.2 under the $p=0.4$ independent‑error model.

#(f) Risks and Misuse

Applying the ultrametric to continuous sensor streams without appropriate discretization may introduce aliasing artifacts. Over‑reliance on shared‑prefix clustering could mask semantically important outliers that differ only in low‑significance digits (e.g., leap‑second adjustments).

#9. Conclusion

We have demonstrated that positional notation intrinsically carries a $b$‑adic ultrametric structure, proved its strong‑triangle‑inequality property, and derived concrete numerical consequences for both data clustering and error correction. The ultrametric view restores the native geometry of numerals, offering a principled alternative to the prevailing linear Archimedean abstraction. Future work will focus on empirical validation in calendar‑processing pipelines and on extending the tree‑topology error‑correction framework to realistic quantum hardware.

#Provenance and AI Involvement

• This paper was drafted, reconciled, reviewed, revised and checked by the QNFO automated research pipeline (qnfo-research-exec 0.27.3-revision-first). Language models were tools in that process; they are not authors. The named author is accountable for the content.
• Models: drafting by openai/gpt-oss-120b, zai-org/glm-5.3-flash; reconciliation, completion, review, revision and verification by openai/gpt-oss-120b, zai-org/glm-5.3-flash, moonshotai/kimi-k2.6, with the QNFO model router as a fallback.
• Sources: every reference is drawn from a bibliography retrieved from arXiv and the QNFO corpus before drafting; every cited work's full text was retrieved (arXiv HTML, or the QNFO paper's body) and the writers were given excerpts of its own sections; statements about a cited work are restricted to its retrieved title, abstract and those excerpts.
• Checks passed before publication: required sections present and finished, cross-references resolve, reference list rendered from the retrieved bibliography, no assistant residue or placeholders, adversarial model review with its blocking findings resolved, and a citation existence check at publication.
• Independent recomputation: a separately generated script recomputed 7 quantitative claims from the paper's stated inputs; 0 differed from the stated value at the stated precision.
• Not checked: whether each cited source supports the sentence that cites it beyond the retrieved full-text excerpts; derivation steps other than the recomputed claims. This paper has not been peer reviewed, and this version was published by the pipeline without a prior manual review by the author.

#References

[1] An embedding, an extension, and an interpolation of ultrametrics. arXiv:2008.10209v2. https://arxiv.org/abs/2008.10209v2 [2] Asymptotic Properties of Hilbert Geometry. arXiv:0711.0446v1. https://arxiv.org/abs/0711.0446v1 [3] Sasakian Geometry and Einstein Metrics on Spheres. arXiv:math/0505221v1. https://arxiv.org/abs/math/0505221v1 [4] Transposition of Notations in Just Intonation. arXiv:1508.07739v4. https://arxiv.org/abs/1508.07739v4 [5] Differential geometry using quaternions. arXiv:2404.03765v1. https://arxiv.org/abs/2404.03765v1 [7] Ultrametric Cantor Sets and Growth of Measure. arXiv:1002.3951v4. https://arxiv.org/abs/1002.3951v4 [8] Noether's Theorem in Multisymplectic Geometry. arXiv:1705.05818v2. https://arxiv.org/abs/1705.05818v2 [11] https://papers.qnfo.org/papers/ultrametric-quantum. QNFO: Qudit Quantum Error Correction. [12] https://papers.qnfo.org/papers/ultrametric-quantum-error-correction-on-bruhat-tits-trees-energy-barriers-classi. QNFO: Ultrametric Quantum Error Correction on Bruhat–Tits Trees: Energy Barriers, Classical Evidence, and the Quantization Question.

#Appendix A. Divergence report

The reconciliation process found the drafts to be in agreement on all substantive claims.

#Appendix B. Claim attribution

Spotted a typo or a possible error? Select the passage and choose Report, or report it here.