What starts as weblog essays—dispersed, informal, chronological—gradually evolves into a structured corpus anchored by persistent identifiers such as DOIs on platforms like Zenodo. This transition turns writing into infrastructure: each text becomes a stable, citable unit within a larger network. The system moves from loose publication to indexed architecture, where content is no longer ephemeral but designed for long-term access, machine readability, and cross-referencing.

Instead of multiple series, locations, and fragmented outputs, the system compresses itself into a single, navigable environment. Smaller series (like early blog posts) expand into large, interconnected bodies of work—hundreds of texts, eventually reaching millions of words—while remaining structurally aligned. The movement is from many places to one system, from accumulation to organisation. The result is a coherent epistemic infrastructure where knowledge is not just produced, but stabilised, connected, and made operable over time. Socioplastics can be understood as a shift from scattered creative production to a unified system where different fields converge into a single operational framework. It begins by merging architecture, art, urbanism, and theory into one continuous field, dissolving disciplinary boundaries. This fusion creates a lexical necessity: new terms, tags, and concepts are required to stabilise meaning across this hybrid terrain. Language is not decorative but structural—it allows the system to organise itself, connect ideas, and remain coherent as it grows.

Socioplastics reaches a critical threshold when it abandons expansion as a primary impulse and consolidates itself as a closed yet generative system of operations. The shift is not quantitative but structural: instead of multiplying concepts, it organizes them into a finite grammar composed of ten interdependent axes—ontology, metric, politics, aesthetics, temporality, value, governance, perception, interoperability, and closure. This grammar does not describe reality but conditions how reality can be constructed, stabilized, and navigated. The decisive move lies in understanding that conceptual proliferation without structure leads to entropy, whereas a limited set of operators, rigorously defined and recursively applied, produces density, coherence, and persistence. Closure here is not restriction but capacity: the system becomes powerful precisely because it knows its limits and operates within them.

At its foundation, Socioplastics redefines existence through identification and proportion. Ontology is no longer a philosophical abstraction but a technical condition: what exists is what can be addressed, indexed, and reactivated within a network of persistent identifiers. This is immediately coupled with a metric regime in which relations—ratios between objects, identifiers, and connections—govern the internal equilibrium of the system. The consequence is a radical reconfiguration of value: it is no longer tied to originality, expression, or symbolic capital, but to structural coherence and relational density. A concept gains weight not by what it claims, but by how it is positioned, connected, and sustained over time. In this sense, thinking becomes inseparable from measuring, and writing becomes a form of infrastructural design.

The Post-Prompt Paradigm

In the current state of the art in LLM prompting (as reflected in recent papers), “topological” refers to modelling the prompt space or reasoning process as an explicit geometric or graph structure. Approaches such as Chain-of-Thought (CoT), Tree-of-Thoughts (ToT), and Graph-of-Thoughts (GoT) treat reasoning as progressively richer topologies: linear chains → branching trees → arbitrary graphs with branching, aggregation (merging of thoughts), loops, and backtracking. The LLM “explores” this graph through multiple calls, evaluations, and search procedures (beam search, DFS/BFS, self-consistency) in order to identify optimal paths within the latent space of meanings. More advanced work draws on persistent homology from topological data analysis (TDA) to analyse and optimise soft prompts. These methods measure how internal “topology” (connectivity, holes, redundancy) evolves during training or inference, and introduce loss functions (e.g., TSLoss) that enforce stable, well-connected, non-redundant structures. The result is a prompt that becomes more interpretable and robust. Other topology-aware optimisation strategies model the prompt search space itself as a discrete graph: nodes correspond to prompt variants, edges to operators (rephrasing, chaining, decomposition, etc.), and search algorithms navigate this graph to improve performance without fine-tuning.