这篇文章（日期为 2026 年 2 月 13 日）重新回顾了海森堡在 1920 年代末提出的「切割」以及波耳的哥本哈根观点：量子力学描述的是观测，而经典物理描述的是经验到的现实；接著主张现代实验能探测许多长度尺度，包括约数奈米的中尺度区域，甚至是光学显微镜下可见的物体，迫使人们更尖锐地说清量子到经典的转变，以及机率看似「塌缩」成单一结果的过程。

文章综述了彼此竞争的塌缩叙事（客观的随机塌缩）、德布罗意-玻姆（de Broglie-Bohm）的导引波动力学，以及 Everett 在 1957 年提出的多世界分支；随后将焦点放在 Wojciech Zurek 以标准量子力学建构的研究纲领：纠缠（由 Schrodinger 于 1935 年命名）无所不在，测量只是系统与环境的交互作用，而退相干会迅速把叠加态扩散到庞大的环境自由度中，使量子干涉在实务上几乎无法恢复；一个量化例子是，一粒尘埃因光子与气体碰撞而在约 10^-31 秒内退相干，约比光穿越一个质子的时间短 10^6 倍。

Zurek 的「量子达尔文主义」补充说，某些稳健的「指标态」（pointer states）可以在不被打乱的情况下被冗余地复制到环境中，因此经典性质之所以能客观可得，是因为许多观察者读取的是同一份纪录；在他与 Jess Riedel 于 2010 年的估算中，阳光可在 1 microsecond 内把一粒尘埃的位置烙印约 10,000,000 次，而且理论预测资讯会「饱和」，因此只需少数环境碎片即可，且初步测试支持此点。文中提出这可能透过一种由退相干选择的「epiontic」观点，调和哥本哈根的认识论波函数与多世界的本体论立场，但也标示出未解问题（为何会选定特定结果、选定何时变得不可逆、以及如何更严格地检验）并提到专家反应不一，包括 Sally Shrapnel 的热情与 Renato Renner 的怀疑；Renner 强调存在实验上可行的情境，观察者可能无法达成一致。

The article (dated February 13, 2026) revisits the “cut” proposed by Heisenberg in the late 1920s and Bohr’s Copenhagen view that quantum mechanics describes observations while classical physics describes experienced reality, then argues that modern experiments can probe many length scales, including mesoscale regimes around a few nanometers and even objects visible under an optical microscope, forcing a sharper account of the quantum-to-classical transition and the apparent “collapse” of probabilities into a single outcome.

It surveys competing collapse stories (objective random collapse), de Broglie-Bohm pilot-wave dynamics, and Everett’s 1957 many-worlds branching, then centers on Wojciech Zurek’s program built from standard quantum mechanics: entanglement (named by Schrodinger in 1935) is ubiquitous, measurements are just system-environment interactions, and decoherence rapidly spreads superpositions into vast environmental degrees of freedom, making quantum interference practically unrecoverable; a quantitative example is a dust grain decohering in about 10^-31 seconds from photon and gas collisions, roughly 10^6 times shorter than light’s transit time across a proton.

Zurek’s “quantum Darwinism” adds that certain robust “pointer states” can be redundantly copied into the environment without being scrambled, so classical properties become objectively accessible because many observers read the same records; in a 2010 estimate with Jess Riedel, sunlight can imprint a dust grain’s location about 10,000,000 times within 1 microsecond, and the theory predicts information “saturates” so only a few environmental fragments are needed, with preliminary tests supporting this. The piece suggests this could reconcile Copenhagen’s epistemic wave function with many-worlds’ ontic stance via an “epiontic” view selected by decoherence, but flags open problems (why a specific outcome is selected, when selection becomes irreversible, and how to test more rigorously) and notes mixed expert reactions, including enthusiasm from Sally Shrapnel and skepticism from Renato Renner, who highlights experimentally feasible scenarios where observers may not agree.