A periodic counterexample to the diffuse allocation characterization
Read the paper (PDF) · Download PDF · Supporting ZIP · Zenodo record
Plain-English explanation · Read on this page · Collaboration
Abstract
We record an explicit diffuse measure on ℝ² whose rooted law is invariant under every measurable, pointwise-covariant preserving allocation of the original observed state, but is not mass-stationary. The measure consists of periodic parallel lines of alternating densities 1 and 2. Covariance reduces every allocation to two line translations; mass resampling changes the root type with probability 2/3. This gives a negative answer to the historical unrestricted diffuse allocation question. The full negative conclusion is also an elementary consequence of Last–Thorisson's atomic counterexample and Lebesgue-product proposition, through the allocation transfer explained below. The periodic model is an illustration of these established mechanisms; no novelty or historical-firstness claim is made.
The idea in plain English
Take an alternating pattern of light and heavy parallel lines, and place the origin on a light line. Deterministic moves that respect the observed pattern and preserve the measure must leave it on a light line. Sampling directly from mass gives the heavier lines more weight and can move the origin to one.
The precise result
The rooted law of this diffuse periodic measure is invariant under every measurable, pointwise-covariant, preserving allocation of its original observed state. It fails mass stationarity. A mass sample from the shifted unit square moves the root to a heavy line with probability 2/3, whereas the original root is light.
What the two densities force
Covariance leaves one translation for each of two line families. Their unequal weights rule out a swap or a collapse of the two families under a preserving allocation. A direct sample by mass still sees the ratio 1:2.
Scope
The allocations are deterministic and use the observed measure as it stands. The claim concerns unrestricted diffuse measures. Results using diffuse balancing assumptions or independent stationary background randomness have different hypotheses.
Background and prior work
Last and Thorisson’s atomic counterexample and their Lebesgue-product proposition already imply the historical negative conclusion by lifting allocations. The periodic model gives an explicit geometric illustration.
Status and AI use
Unrefereed preprint. AI tools assisted substantially in developing, writing and checking the work. Independent human review is welcome.
Read the paper
Open the published PDF · Download PDF · Original PDF on Zenodo
Supporting materials
Download diffuse-allocation-verification.zip · Original diffuse-allocation-verification.zip on Zenodo
Collaborate on this result
I welcome independent verification of the preservation argument or a comparison of how additional background randomness changes the class of allocations.
Authorship and responsibilities would be discussed in relation to substantive contributions. Contact Alec.
Cite this paper
Kriebel, Alec (2026). A periodic counterexample to the diffuse allocation characterization. Version 1.0.0. Zenodo. 10.5281/zenodo.23397026.
Licensing
The deposited work is available under Creative Commons Attribution 4.0.