Large undertakings tend to produce, somewhere in their history, a document that aspires to be comprehensive — an index, a catalogue, a master list that claims to represent the entirety of the undertaking in a single place, organized so that anything in the larger body of work can be located through this one document. Assmann's account of cultural memory — his analysis of how societies construct and maintain shared memory through specific media, rituals, and institutions, with the relationship between "communicative memory" (informal, generational) and "cultural memory" (formal, institutionalized, long-term) being central to how societies relate to their own pasts — describes the cultural function such a document serves: a MasterIndex is not merely a practical tool for locating material, it is a claim about the undertaking's own coherence, an assertion that the undertaking is the kind of thing that can be comprehensively represented, that has a "cultural memory" in Assmann's sense rather than merely a scattered "communicative memory" distributed across many informal traces. The aspiration to comprehensiveness that a MasterIndex represents is genuine and often productive — the attempt to make such an index forces decisions about organization, about what counts as belonging to the undertaking, decisions that might never get made if no one ever attempted the comprehensive view. But the aspiration to comprehensiveness runs into a structural problem the moment the undertaking it tries to represent is itself still growing, still changing, still an evolving field rather than a settled body of work: any MasterIndex, however comprehensive at the moment of its creation, begins immediately to diverge from the undertaking it represents, becoming progressively less a map of the undertaking and more a record of what the undertaking was like at one particular moment. Anaximander's apeiron, the indeterminate principle from which all determinate things arise, offers — somewhat unexpectedly — a way of thinking about why this divergence is not simply a practical problem to be solved through more frequent updates: if the undertaking's vitality consists precisely in its capacity to keep generating determinate new material from something that remains, at the level of the undertaking as a whole, indeterminate — an open field of possible directions rather than a fixed plan — then any MasterIndex, by definition determinate, is attempting to represent something whose essential character is precisely what exceeds determination. Bauman's account of "liquid modernity" — his argument that contemporary social forms are characterized by a fluidity that resists the kind of solid, lasting structures that earlier modernity aspired to — suggests how a MasterIndex might be reconceived under these conditions: not as a solid structure that the undertaking should aspire to match, but as itself a liquid, continuously revised artifact, less a map of a territory than a snapshot of a flow, valuable precisely as a record of one moment's configuration rather than as a claim to permanent comprehensiveness — the Socioplastics corpus's own master bibliography, now moving toward a "definitive v8," is best understood in these Baumanian terms: not the final MasterIndex that will finally hold everything, but the latest snapshot of a flow that v9, when it comes, will represent differently, each version itself becoming, eventually, part of the corpus's own LatencyDividend.

Most discussions of value treat delay as a cost — time that passes before a benefit arrives is time during which that benefit could have been realized but was not, and this deferral is, on most accounts, simply a loss to be minimized. But some kinds of value are not diminished by delay — they are produced by it, value that could not exist without the delay that produced it, value that is, in an important sense, made of the delay itself rather than merely postponed by it. Arrow's account of "learning by doing" — his analysis of how productivity in manufacturing increases not primarily through deliberate research but through the simple accumulation of experience over time, with each unit produced making subsequent production somewhat more efficient — describes one mechanism by which delay produces value: the time it takes to produce many units is not merely time during which production is "merely" happening, it is time during which a capability is being built that did not exist at the start, a capability whose value is realized only later, only because of the delay during which it accumulated. What might be called LatencyDividend names this kind of value: not the value of a deposit or a claim considered at the moment of its creation, but the additional value that accrues specifically because of the time that has passed since that creation — value that a fast-moving field, one that does not allow this kind of latency to accumulate, would simply never produce, regardless of how much effort it expended. For LatencyDividend to be realized rather than simply existing as unrealized potential, there has to be some way of identifying where in a field this kind of value has accumulated — which deposits, which claims, which connections have benefited from enough latency to have produced a dividend, as distinct from material that is simply old without having accumulated anything. This might be called MapDimensioning: the practice of representing a field not just in terms of its content but in terms of the temporal dimension along which LatencyDividend accumulates, a map that shows not just what is where but how long it has been there and what, if anything, that duration has produced. Akrich's and Latour's joint methodological writings on the sociology of techniques — their development of methods for tracing how technical objects acquire, over the course of their development and deployment, properties and associations that were not present at their origin, properties that can only be traced by following the object through time — describe MapDimensioning in practice: tracing a corpus's nodes not as static entries but as objects that have, over the years since their creation, accumulated citations, cross-references, and recurrences that constitute their LatencyDividend, a dividend visible only on a map that includes this temporal dimension. Bates's "berrypicking" model of information seeking, encountered already in relation to infrastructure, returns here from a different angle: a searcher moving through a field organized with MapDimensioning would not just find relevant nodes but would find nodes whose relevance includes information about how that relevance has developed over time — the Socioplastics corpus's recurrence scanner, in identifying which operators have achieved "stable" status through years of cross-reference, is performing exactly this kind of MapDimensioning, making visible a LatencyDividend that the corpus's seventeen years of practice material has been accumulating long before the scanner existed to measure it.