semishen / 3sth.net

3sth.net

社群織衍計畫 | 3sth.net

計畫影片： 3sth[DOT]net

計劃簡介

「一切事物的初始狀態，都是些看似無關的碎片、不可測的擾動，然而當此混沌結束後，這些碎片卻又自主有序的聚合成一個整體。」- Wikipedia

2006年小說《三體》的問世，讓這一古老的天文物理難題再度來到世人面前，然而三體問題所引發的思辨，除了可以朝著更大尺度的外太空（Outer Universe）與更為深邃的時空間（Deep Spacetime）外，亦或可反轉為當代文明的運作模式，即原先的自然 -- 人類雙體文明，在科技逐步全面化，透過種種量化技術，解構原先的現代二元對立架構，建立了更為多層次的體系。

若以一種線性觀點檢視前述說法：最初人們對於世界的詮釋歸因於一體 -- 自然；近代科學與哲學先驅笛卡爾，則嘗試架構雙體的世界觀，即廣義上代表人類的心與代表自然的物。然而當經驗逐漸轉變成以數據為基礎，則不論農工產業、以至於更高級的服務業體系，都因為數據累積的資訊而科技化、數位化，最終顯現為時代集體的典範轉移 -- 一個由三體（自然、人、科技）共構的複雜網絡。

在古典物理的三體架構中，主要的問題是探討宇宙天體間，假設僅受萬有引力作用時，則相較於兩天體的週期性**關係，三體**無法計算出所有解，即其**方式須藉由渾沌模型描述。

由此檢視當代網絡的三體問題，則複雜性的來源是量化的數據，即數據由原先作為物與人衍生物，逐漸轉換為具有主體與能動性的存在，成為混沌的基礎。換言之，科技的飛速進展，重構原有的二元關係，令自然界與人類社會的互動變得不可預測。物理真實與虛擬真實以數據為媒體相互交融疊加，成了一張大型網狀構造的概念。

在藝術史上，藝術透過對技術、程序與行為的分析與考察，再現了時間、空間、關係以及系統等概念並進而探討其內涵，然而藝術是否能夠面對集體永恆的不確定性？藝術又該如何在三體構成的動態風險社會中，發掘此一狀態中所生的感知特性？

社群織衍計畫 a.k.a 3sth.net 作為一個藝術性的研究計畫，探討三者間的互動將會形成什麼樣貌的文化織體，以及此間的運作方式：透過自行於網絡與現實環境搭建的三體網絡，讓藻類（自然）、使用者（人類）與人工智慧（科技）得以透過數位介面交流，其中所產生的相關數據經由蒐集、演算處理後，由搭載自動演算的機器人手臂（robot arm）將其編織成網，此一參數資料並將轉化為影像，投影於網狀織體上。人類在參與及觀看的位置間擺動，透過自身實踐經驗觀看認識，思考與人工智慧與數據，對原有的雙體關係，帶來了什麼樣的變化？而所謂的三體關係，又將產生什麼新的現象？數據為三體交互作用的內在運作結果，而織體為三體交互作用衍生的表徵架構，AI所帶來的不穩定性，讓實驗不僅受限於藝術計畫建立的框架中，並也由網絡的混沌決定；參與者不再受到既定的互動機制掌握，而將一同體驗創作者實驗過程中的未知與不可預測。

Introduction

At the very beginning, the origin of things seems to consist merely of irrelevant bits and unpredictable turbulence. Yet, as this chaos comes to an end, the fragments gather together and autonomously form a bigger structured whole. - Wikipedia

The Three-Body Problem, a science fiction novel by the Chinese writer Liu Cixin came out in 2006, putting discussions of the century-old astronomical three-body problem under the spotlight. Relevant dialogues not only are significant in addressing the outer universe and deeper spacetime, but also could be employed to direct the interrogation on how we see our modern-day civilization. As technological advancements had reached nearly every aspect of our lives, the dualism (of human and nature, of subjectivity and objectivity) since Enlightenment seemed to give way to a more complex, multi-dimensional system emerging as a result of the ever-increasing presence of quantifiable data and technology.

Under the presumption of a linear history, we could see it as a progression from monisms which generally see the universe (nature) as consisting of a single entity, to Cartesian dualism which distinguishes between the subjective human consciousness (human) and the objective physical world (nature). In the new era where everything generates data and is understood through the analysis of data, the digitalization eventually manifests itself in the form of a shifting paradigm -- a new system out of triangular dynamics between digital representatives of nature, human and technology.

Originally, the classical three-body problem describes how three objects would move under one of the physical forces -- for example, how three celestial bodies interact with one another gravitationally (dictated by Newton's laws of motion and of universal gravitation). Unlike two-body problems, three-body problems are, in general, not analytically solvable. Chaos theory is therefore developed to describe the dynamical situation.

The contemporary interpretation of the three-body problem, on the other hand, looks into quantified data and its consequential complications. Previously existing as by-products of the physical world and human activities, data gradually gains its agency, and, maybe someday, subjectivity, under rapid technological advancements. The change further introduces indeterministic complexity (chaos) into our world, disrupts the dualism of human-nature relation, and urges us to re-examine, reconstruct or redefine new multilateral relations to include virtual existences. Physical reality and virtual reality now has overlapped with each other and merged into a giant mesh.

Through addressing contemporary technologies, systems, and dynamics— both analytically and aesthetically -- art has been representative of the paradigm of its time, manifesting in itself our idea of time, space, relation, and structure. However, is art capable of interrogating this particular everlasting uncertainty of collectivity? How does art venture into the unknown of new three-body relations and bring the representation to the audience without losing its poetic sensitivity?

As an experimental research project bridging between art and technology, 3sth.net centers around new three-body movements at the conceptual level, and investigates the interactions between algae (nature), users (human) and artificial intelligence (technology) in practicality. Through digital interfaces and connected bioreactors, the data of interactions is collected, processed, and visualized. Moreover, it is used to control parameters of a robotic arm to fabricate a textile sculpture, as well as to generate images -- symbols created by AI as though the history were written in a different way -- to be projected onto the sculpture. Whereas the physical representation invites perception, the digital interface allows participation. Oscillating between these two positions, between active participation and passive spectatorship, the audience is thus left to contemplate over relevant questions. How do data and artificial intelligence play a role in this complication? What could be generated through the new multilateral relations? Data, as the product of internal mechanisms of the digital universe, and the sculpture, as the external embodiment, furthermore, are complementary to each other both in terms of conception and representation. Last but not least, as generative models of AI bring indeterminacy into the digital universe, the project is allowed to exist both in and outside of the context of fine art -- participants are not constrained by our authorial intention and artistic framework; they are part of the online community which experiences and actively contributes to the unknown and unpredictability of this experiment.

技術架構

社群織衍計畫 | 3sth.net 技術上包含下列項目：

• 3sth.net (WebApp)

  • KeJiRungYi_2018 (WebApp 前後端設計)
  • 3sth_backend (藻類、聊天機器人、deep ai)

• Meta AI (GAN model, 字符結構生成)

  • C_TextImgGenerator (字符結構生成範例)

• 展場投影系統

  • C_3sthProjection (三階段投影內容)

• 機械臂數位編織

• 藻類IOT生物反應器