Where language models learn the statistical structure of text, world models learn the statistical structure of space and time

Focus your attention on the inner loops, where the bulk of the computationsand memory accesses occur.. Try to maximize the spatial locality in your programs by reading data objectssequentially, with stride 1, in the order they are stored in memory.. Try to maximize the temporal locality in your programs by using a data objectas often as possible once it has been read from memory.

Programs that repeatedly reference the same variables enjoy good temporallocality..For programs with stride-k reference patterns, the smaller the stride, thebetter the spatial locality. Programs with stride-1 reference patterns have goodspatial locality. Programs that hop around memory with large strides havepoor spatial locality..Loops have good temporal and spatial locality with respect to instructionfetches. The smaller the loop body and the greater the number of loop it-erations, the better the locality.

eople prefer to know who else is present in a shared space, and they usethis awareness to guide their work

Viewed from a practice perspective, the distinction be­tween cyclic and linear time blurs because it depends on the observer's point of view and moment of observation. In particular cases, simply shifting the observer's vantage point (e.g., from the corporate suite to the factory floor) or changing the period of observation (e.g., from a week to a year) may make either the cyclic or the linear aspect of ongoing practices more salient.

The past leads to and influences the future, but the future does not influence the past.Thus El Sawy s research provided a second clue that past and future are re­lated, and it even added a causal direction (i.e., “A connection to the past fa­cilitates a connection to the future” [March 1999, p. 75])·