A continuation fragment about opposite Weyl chambers and full dimension

Definitions and mathematical statement

This record is the middle of a sentence after Conjecture 33, not another conjecture. The supplied source assessment and neighboring fragments identify its scope: D. Kleinbock observed that for two diagonal rays in opposite Weyl chambers, the argument of Kleinbock--Margulis [108] can be adapted to produce the witnesses of Conjecture 33, with a full-Hausdorff-dimensional witness set. Reference [108] itself is described by the aid as a one-flow thickness/avoidance result, not a theorem asserting an arbitrary two-ray intersection. The missing words before and after the fragment are not promoted to a standalone problem.

The mathematical lemma developed in the aid is local dimension subtraction. If $E,B$ are subsets of a $d$-dimensional manifold and, in a nonempty open $U$, \[ \dim_H(E\cap U)=d,\qquad\dim_H(B\cap U)<d, \] then \[ \dim_H((E\setminus B)\cap U)=d. \] This follows from $E\cap U=((E\setminus B)\cap U)\cup(E\cap B\cap U)$ and the maximum rule for the Hausdorff dimension of a finite union. Thus if $E$ is thick, meaning full dimension $d$ in every nonempty open set, it stays thick after deleting a globally lower-dimensional $B$, or a $B$ which is strictly dimension-subcritical in each such open set. Equality $\dim_HB=d$ would not suffice.

For the supplied modern application, let $X_3=SL(3,{\mathbb{R}})/SL(3,{\mathbb{Z}})$, of dimension $8$, and let $A$ be the full diagonal group, with positive identity component $A^0$. Fix a finite or countable family of nontrivial rays $A_i^+=\{\exp(tH_i):t\geq0\}$ and their two-sided one-parameter subgroups $A_i$. Put \[ E=\{x:A_ix\text{ is bounded for each prescribed }i\},\qquad B=\{x:A^0x\text{ is bounded}\}. \] Bounds and compact sets may depend on $x$ and $i$; a single compact set for all $i$ is not required. The aid invokes the An--Guan--Kleinbock joint-bounded-orbit theorem for thickness of $E$, and the Einsiedler--Katok--Lindenstrauss dimension-$2$ theorem for $B$. Finite components do not change full-$A$ boundedness. Therefore $E\setminus B$ is thick: every prescribed ray orbit is bounded there, while the full diagonal orbit is unbounded. Taking two rays gives the witness conclusion in the neighboring Conjecture 33. This is a conditional deduction from these precisely stated modern inputs, not a claim that [108] alone proves their countable intersection.

The original fragment concerns its historical opposite-chamber argument. The broader modern deduction uses different sources, whose complete hypotheses and citations must be checked; the elementary subtraction lemma by itself constructs no common bounded-orbit set.

Short English statement

The entry continues a historical claim about opposite-Weyl-chamber rays and is not an independent problem. The supplied aid isolates a valid dimension argument: deleting a strictly lower-dimensional set from a locally full-dimensional set preserves thickness. Combined with separate modern joint-boundedness and full-diagonal dimension theorems, it yields the desired witnesses without attributing a two-ray theorem to a one-flow source.

Sources

• Local UnsolvedMath registry, permanent ID $20003317$, source code catalogue entry; canonical location aim-other-notes.json, notes[123]. Neighboring records $20003315,20003316,20003318$ provide Conjecture 33 and the surrounding sentence boundaries.
• AIM Emerging applications of measure rigidity, continuation after Conjecture 33, https://aimath.org/WWN/measrigid/measrigid.pdf. It credits D. Kleinbock's adaptation of the argument in [108], identified in the aid as Kleinbock--Margulis, without a full bibliographic entry.
• An--Guan--Kleinbock's 2015 common-bounded-orbit theorem and Einsiedler--Katok--Lindenstrauss, Theorem 10.2, for the dimension-$2$ full-diagonal bounded locus, as named by the supplied aid and neighboring records. Full publication details are absent.
• Research attempt 1 is marked context-only/not-a-problem, proved, and low novelty confidence, despite the registry conjecture/open labels. Its explicit caveat says the aid is machine generated, not peer reviewed, and requires independent verification before citation.