運算化設(shè)計: 知識時代的新設(shè)計

撰文 (羅馬尼亞)米哈伊·納丁翻譯 高巖

當(dāng)今世界，知識正在變得越來越運算化。以前與知識的獲取、交流及批評的方式與方法都漸漸淡出，取而代之的是由數(shù)字化手段實現(xiàn)的信息探尋、知識分配與價值評估。帕斯卡（Pascal）、萊布尼茲（Leibniz）和皮爾斯（Peirce）為此變化，搭建了多項必要的概念結(jié)構(gòu)。在學(xué)習(xí)知識的動機、獲取知識的渠道及分享知識的意愿等問題上，他們提出許多相關(guān)的問題；換句話來講，他們給出了認(rèn)知方面的定義。更近的，布爾（Boole）①、維納（Wiener）[1]和凡·諾依曼（von Neumann）提供了所需的科學(xué)基礎(chǔ)，并最終由阿塔那索夫（Atanasoff）、 祖薩（Zuse）、 ?？颂兀‥ckert）與 莫奇萊（Mauchly） 制造出機器③。電腦圖像、可視化、桌面出版、CAD、多媒體、虛擬現(xiàn)實、互聯(lián)網(wǎng)、萬維網(wǎng)，已經(jīng)成為了我們生活的一部分，還有更多的知識也將來到我們身邊。在這一進程中，科學(xué)也開始變得趨向運算化：最耳熟能詳?shù)谋闶俏锢?、生物、化學(xué)的轉(zhuǎn)變；工程領(lǐng)域，在材料應(yīng)用、機器人技術(shù)，甚至是計算機生產(chǎn)及軟件自動生成的各個方面，都經(jīng)歷著同樣的變化。在這樣的大環(huán)境下，設(shè)計領(lǐng)域發(fā)生了怎樣變化？

1 現(xiàn)狀

目前的普遍現(xiàn)象是——計算機和設(shè)計師僅僅是工具與使用者的關(guān)系。事實上，在設(shè)計領(lǐng)域內(nèi)部，一直存在計算機是否能夠替代設(shè)計師的爭論，至少計算機已經(jīng)取代了鉛筆和馬克筆，以及枯燥的模型制作。平面設(shè)計師又一次走在最前端，他們樂于、也最容易開拓各種新的領(lǐng)域，比如字體設(shè)計、按需出版和電子出版。他們很快就發(fā)現(xiàn)數(shù)字技術(shù)并不僅僅代表著原有功能上的先進工具，更代表著活動領(lǐng)域的拓寬。激光顯示器、掃描儀、手繪板，硬盤，不斷嘗試整合越來越多的網(wǎng)絡(luò)工具（瀏覽器，Applets程序，F(xiàn)rames語言）?？茖W(xué)抽樣、拼接、變異、超鏈接等手段方法也加入進來，正是這樣，多媒體和網(wǎng)絡(luò)交互為印刷品增色不少。我認(rèn)為最成功的范例是傳媒設(shè)計的新應(yīng)用：虛擬設(shè)計室，也就是設(shè)計師基于已有的技術(shù)與工作方式，設(shè)計出一種新的互動工作環(huán)境。因此運算化成為設(shè)計工作的組織機制，并在實踐中根據(jù)對它的評估，不斷接受新的測試。但是，即使在平面設(shè)計領(lǐng)域，根本問題還是被回避了：我們是否認(rèn)定人類大眾是一個不會隨著科學(xué)和技術(shù)改變的群體？我們是否在“設(shè)計”我們自己的公眾，為植根社會且更具個性化的人們之間的互動，發(fā)明各種形式和手段？我們?nèi)绾纬綄τ诖蟊妭鞑サ拿詰?，令小眾傳播為目?biāo)的“設(shè)計”，在內(nèi)容與表現(xiàn)手段上都能夠與大眾傳播取得平衡？我們又該如何改變固有的想法，嘗試革新性交流的手段與動機？

技術(shù)，盡管已經(jīng)被創(chuàng)新地應(yīng)用在傳媒設(shè)計領(lǐng)域，但仍然遠遠超出我們已經(jīng)應(yīng)用的范疇。在其它設(shè)計領(lǐng)域，主要是產(chǎn)品設(shè)計和工業(yè)設(shè)計，目前的情況還不是特別明朗。傳統(tǒng)的工業(yè)設(shè)計幾乎不能提供新的就業(yè)機會，需要數(shù)字整合的教育項目的進程也比較緩慢?？杀氖?，教育工作者們已經(jīng)習(xí)慣了不懈地思考工業(yè)革命模型的硬性條件，這些條件并不是依靠和設(shè)計相關(guān)的新思路，更多是基于對制造手段的預(yù)期要求。眾所周之，在技術(shù)上的投資，諸如軟硬件、維護、培訓(xùn)、研究等，費用高得令人咋舌。很少有人敢于嘗試創(chuàng)業(yè)的風(fēng)險，更不用提獲得成功的人了。大企業(yè)為了鞏固自己的強勢地位，逐漸吸納這樣一些人，他們能夠管理當(dāng)前基于計算機或計算機輔助設(shè)計的復(fù)雜工作環(huán)境。在很多情況下，為了保護獨立的知識產(chǎn)權(quán)，企業(yè)的設(shè)計并不公開。那些利用計算機及軟件來開發(fā)先進產(chǎn)品的設(shè)計團隊，甚至不配置上網(wǎng)設(shè)備。而當(dāng)涉及到數(shù)字技術(shù)的時候，他們又不得不嘗試不同的協(xié)作設(shè)計方法，這些注定都是要避開公眾的視線隱蔽地進行。他們忽視了這種孤立的手段方法與其工作本身結(jié)構(gòu)之間的矛盾性。結(jié)果是他們可以嚴(yán)守設(shè)計秘密（如新車模型、新玩具、新家具等等），但每次發(fā)布時卻經(jīng)常發(fā)現(xiàn)自己已經(jīng)落后于市場了。

技術(shù)領(lǐng)先于設(shè)計并不僅僅表現(xiàn)在工業(yè)設(shè)計領(lǐng)域。在紡織業(yè)、時尚界、玩具商及室內(nèi)設(shè)計中也非常明顯，各種形式的設(shè)計還都是手工模式為主。難以避免的后果是各種設(shè)計缺陷都被計算機輔助設(shè)計掩蓋了，結(jié)果扔給折扣市場顧客的都是些沒有品味的產(chǎn)品。軍事和人工智能領(lǐng)域之外，無論設(shè)計初衷是多么“高檔”的、價格多么合適的小玩具，都不能在我們的文化中得體地存在。

2 設(shè)計理論的可能性

運算化設(shè)計認(rèn)同工具與使用者之間的有機聯(lián)系。運算化設(shè)計的目的在于通過這種聯(lián)系轉(zhuǎn)換出更多的可能性，并通過設(shè)計付諸實現(xiàn)。要達成這一目標(biāo)，“運算”既不能像今天這樣僅僅作為一個表達的媒介，也不能僅是產(chǎn)生變體，不管這種變體成不成系統(tǒng)，因為運算化已經(jīng)成為設(shè)計本身的構(gòu)成機制。設(shè)計理論是否真的可行的老問題，再次擺在我們的面前，如果可行，會以什么樣的形式實現(xiàn)？

之前當(dāng)這個問題還處在構(gòu)想階段的時候，理論只能是遵從實際的設(shè)計。最好的設(shè)計師，或者至少是那些將自己的想法透過文字表達出來的人，能夠?qū)⒆约旱某删秃侠砘?。也就是說，他們僅僅是在設(shè)計被認(rèn)同或者被公眾認(rèn)可之后再對自己的工作進行評價，這樣的情況大家都早已司空見慣了。眾所周之，設(shè)計從手工制作開始，在進化的過程中，設(shè)計在各類人類的嘗試中第一次獲得了公正的稱號。但是隨著手段與方法的拓展，設(shè)計又創(chuàng)造出自己的依據(jù)和概念的范疇。隨著設(shè)計評論與設(shè)計歷史的出現(xiàn)，設(shè)計理論與設(shè)計教育也一同建立起來。最初的理論必然是分析性的，通過觀察和歸納獲取新知識，這與設(shè)計普遍化進程中派生出的演繹和推導(dǎo)相互補充，結(jié)果是設(shè)計理論家能夠嘗試綜合各類理論。俄羅斯構(gòu)成主義、包豪斯、美國二戰(zhàn)后的設(shè)計都是新概念領(lǐng)域的典型例子。其中有些取自形態(tài)學(xué)、結(jié)構(gòu)主義、符號學(xué)，有些甚至源于心理學(xué)、語言學(xué)、社會學(xué)和工程學(xué)，還有一些由功能設(shè)計派生出來。近些年，系統(tǒng)運算式、啟發(fā)式程序甚至遺傳學(xué)都能在設(shè)計理論中有所體現(xiàn)。另外，“設(shè)計假設(shè)”同樣利用運算化進行建模和檢測。我自己的設(shè)計機器系統(tǒng)（Design Machine?）工作室也是這個方面的例子之一。

推論式的理論保持了結(jié)論性等式的魅力：即首先會有個設(shè)計的原因——設(shè)計項目，然后是設(shè)計的結(jié)果——設(shè)計衍變成可識別的物體，其特性通過商務(wù)交易和文化認(rèn)同體現(xiàn)出來。其結(jié)果是，設(shè)計必須遵循闡述“如何設(shè)計”，以及“什么才是一個好設(shè)計”的理念。它的前景似乎越來越迷茫。首先，語言作為我們相互理解的最佳媒介，卻不是保證人類活動的唯一必然因素，因為人類的本性并不單純是語言的歸納與總結(jié)；其次，往往人們都認(rèn)為好的設(shè)計之所以好，是由所處文脈決定的（比如形式上的、功能上的、結(jié)構(gòu)上的等等）。顯然，好的設(shè)計理論應(yīng)該可以解釋為什么有些設(shè)計就是不好的。由于這些問題都是借助語言的幫助而產(chǎn)生的，我們也認(rèn)識到設(shè)計理論其實處在學(xué)科間，是跨學(xué)科的。我們可以對設(shè)計投以各種熱情洋溢的褒義詞藻，但是它們并不會對設(shè)計理論的實踐及設(shè)計本身有太大幫助。盡管如此，我們并不否認(rèn)設(shè)計對于語言的依賴。人類工程學(xué)、功能學(xué)、心理學(xué)、社會學(xué)與經(jīng)濟學(xué)各種概念，也開始影響與設(shè)計相關(guān)的問題和設(shè)計教育的課程設(shè)置。設(shè)計師經(jīng)常向潛在客戶陳述人類工程、文化或者符號學(xué)等方面的內(nèi)容，而不是設(shè)計本身。之后，設(shè)計實踐又開始在更傾向于保護書面內(nèi)容的社會中，對建筑產(chǎn)品權(quán)益進行保護，而不再是含糊不清的視覺傳達，因此，律師的語言也被借用到設(shè)計理論中來。

3 設(shè)計知識庫

運算化設(shè)計已經(jīng)告別了進退兩難的境地，像其他運算化知識一樣，融入到人類務(wù)實的生存狀態(tài)中。我們都知道運算物理學(xué)（Computational Physics）既是理論，也是實踐。作為理論，運算物理學(xué)從宇宙起源開始提出各種假設(shè)；作為實踐，通過模擬各種假設(shè)來驗證其真實性，并最終轉(zhuǎn)化為各類探索宇宙的工具。模擬能夠為我們研究宇宙提供各類新知識，同時也能幫助我們認(rèn)識自身活動中的知識，這個過程中并不需要考慮我們是否是物理學(xué)家或者其它領(lǐng)域的專家（生物學(xué)、化學(xué)、哲學(xué)、藝術(shù)）。這些知識在實際工作中都能夠為我們開拓思路提供前攝性（一種更主動面對未來變化的態(tài)度，往往通過自身主動的變化導(dǎo)致變化，而不是等待預(yù)料之外的變化而不知所措）的幫助，很容易讓我們聯(lián)想到在外太空開展的植物、動物，食物甚至是藝術(shù)的實驗。運算工程學(xué)（Computational Engineering）讓設(shè)計師整合了新材料與許多有趣的事物，為未來創(chuàng)造出更多的可能性：從分子或原子出發(fā)進行假設(shè)以構(gòu)建新結(jié)構(gòu)，并在其它自然資源加工處理之前，進行運算化測試。運算遺傳學(xué)（Computational Genetics）這項實踐活動的中心，就是為人類謀求更多的福祉。

運算化設(shè)計的主要意圖是想借助超越主觀的外力驅(qū)動設(shè)計，這些外力讓設(shè)計在需求性評估、可能性評估、和體現(xiàn)人類特性的方式評估中變得可能且必要。這些評估都以數(shù)據(jù)的形式進行，更準(zhǔn)確地講是非常復(fù)雜的數(shù)據(jù)庫。其他設(shè)計理論的本質(zhì)是被動反應(yīng)，因此經(jīng)常是投機性的；而運算化設(shè)計理論基于大量的數(shù)據(jù)，本質(zhì)是前攝性的。運算化設(shè)計具有不可避免的局限性，一方面指在需要同時兼顧質(zhì)量和數(shù)量的前提下，設(shè)計人員本身有限的搜集和組織數(shù)據(jù)的能力；另一方面指我們處理數(shù)據(jù)的高端運算化程序的能力非常有限。像其它運算化理論一樣，運算化設(shè)計同時也是一種實踐，是遠遠不同于我們?nèi)粘；顒有问皆诟鼜V義的環(huán)境下進行的精確設(shè)計實驗。他的成敗取決于測試的結(jié)果，核心是關(guān)注人類最重要的資產(chǎn)——知識。運算化設(shè)計需要配備一個相應(yīng)的知識庫，這將超越目前所有設(shè)計博物館、收藏館及各類書籍、文章關(guān)于設(shè)計的一切內(nèi)容。此外，這個知識庫需要以今天人類全球化生存為依據(jù)，設(shè)置導(dǎo)航、搜索、檢索的功能。我們需要從更廣義的文化角度來審視人造物，連同它的規(guī)劃及孕育它的設(shè)計。這個知識庫還應(yīng)該包含涉及視覺表達、運動、色彩、人體工程學(xué)，及其整合他信息傳達手段（聲音、材質(zhì)、氣味等）的運算化表達知識（Computational Expressed Knowledge）。實現(xiàn)這些目標(biāo)任務(wù)艱巨，但又無法逃避。不幸的是，大多數(shù)被我們看做是“歷史學(xué)?！钡牟┪镳^和收藏館，就好比一個垃圾場，并沒有收集設(shè)計真正所需的知識庫。

設(shè)計知識庫的典型例子之一，是已經(jīng)在設(shè)計領(lǐng)域里廣泛應(yīng)用的計算機程序。事實上，一個CAD程序、字體設(shè)計程序、多媒體編輯程序，或者網(wǎng)絡(luò)瀏覽器都是高度抽象的理論表達。在這些程序里，我們需要設(shè)定幾何關(guān)系、材質(zhì)特性、調(diào)節(jié)光線，完成動作、角度的設(shè)置，建立圖片與聲音文件的聯(lián)系，整合文字內(nèi)容和其它設(shè)計因素。當(dāng)然這些工作只通過一個程序是完成不了的，但是至少可以建立起設(shè)計的一致性，或者令我們能夠更好的理解設(shè)計?；谶@些“理論”的設(shè)計實踐，才是真正對設(shè)計任務(wù)的研究，并通過人造物（即設(shè)計的物質(zhì)對象）數(shù)字化后的性能表現(xiàn)來評價設(shè)計。得益于成功的設(shè)計經(jīng)驗，一些較為成功的程序版本可以變得越來越好。在我寫下這些話的時候，Netscape? 3.0正式發(fā)布；這次它吸納了遠程電話會議技術(shù)，這點讓我接下去的話變得不言自明了。在設(shè)計假設(shè)不斷延續(xù)的進程中，有一些設(shè)計理論被證明做了不恰當(dāng)?shù)募僭O(shè)而逐漸消失。就在兩年前，電話會議技術(shù)，這項傳媒設(shè)計的主流創(chuàng)意，還擁有數(shù)十億美元的潛在市場；但今天，它已經(jīng)成為一項標(biāo)準(zhǔn)的瀏覽器功能。

我想對于視“使用程序”為設(shè)計活動說得更清楚一些： Macromedia Director?、Phontographer?、Alias?、Vellum?，或用于桌面出版的程序（Quarkexpress?、Pagemaker?），還有材質(zhì)設(shè)計軟件、珠寶設(shè)計軟件等，這些都是我們能夠在商店購買到的軟件，然后應(yīng)用在不同的工作中。但相對于鉛筆、畫筆、美工刀、木材或金屬、膠水等，這些設(shè)計師原來的常用工具，程序就是他們曾經(jīng)擁護和發(fā)明（比如電話會議）的理論精華。顯然程序并都不能窮盡設(shè)計的各個環(huán)節(jié)，但卻可以表述和整合與興趣、多媒體、字體設(shè)計、CAD、出版物設(shè)計或在線廣告相關(guān)的設(shè)計活動。編寫程序的大多是由程序員、心理學(xué)家、設(shè)計師等組成的大型團隊，他們需要綜合各類物理學(xué)、數(shù)學(xué)、美學(xué)、符號學(xué)及人體工程學(xué)等知識。事實上，任何一個程序都是一次理論假設(shè)，再利用程序來檢驗假設(shè)。程序生成的最終產(chǎn)品能夠與應(yīng)用運算化工程創(chuàng)造出的新材料，或運算化遺傳學(xué)研制出的新醫(yī)藥相媲美。

4 計算機并不僅僅是工具

事實上，新材料、新醫(yī)藥、新基因的創(chuàng)造都可以被理解成“設(shè)計”。我之所以用這個標(biāo)題，意在表達在運算化時代下，設(shè)計變?yōu)橐粋€涉獵范圍更廣的詞。如果我們不能理解運算化設(shè)計的必要性，我們可以繼續(xù)形而上學(xué)地認(rèn)為計算機僅僅是一種工具。我們可以繼續(xù)對設(shè)計的誕生進行詩意的描述，即設(shè)計源于設(shè)計師的大腦，就像神話中金星維納斯誕生于土星朱庇特頭顱中一樣。毋庸置疑，程序的致命弱點是無法實現(xiàn)人類直覺所能實現(xiàn)的，這并不是因為它們沒有直覺（它們也不必有），而是因為設(shè)計師只是在簡單機械地使用，而非創(chuàng)造性地使用它們。

我并不回避這個問題：缺少了計算機，設(shè)計的許多方面仍然可以很好地展現(xiàn)出來。這些方面并不屬于運算化設(shè)計的陣營，畢竟運算化設(shè)計并不能替代設(shè)計，而只是在新務(wù)實的環(huán)境下延續(xù)設(shè)計、拓寬設(shè)計。設(shè)計目前面對的最大挑戰(zhàn)在于運算化形式下缺少新的設(shè)計知識，造成千篇一律的解決方案。哈伯望遠鏡的設(shè)計和它后期的修理，可以在仍有缺陷的狀態(tài)下就被發(fā)射出去，開始它圍繞地球的旅行。這依賴了運算化設(shè)計模型（主要涉及虛擬現(xiàn)實的方法和手段），它能夠發(fā)現(xiàn)那些致命的設(shè)計錯誤，也可以生成包括完成特定任務(wù)的工具設(shè)計在內(nèi)的、對其進行優(yōu)化改進的程序步驟。這就是運算化設(shè)計不僅引入建模、渲染、動畫，還需要模擬（包括虛擬現(xiàn)實）的原因。這僅僅是運算化設(shè)計的一點點成果，但足以讓我們想象，利用綜合運算化設(shè)計完成的數(shù)字化模型，比起簡單的塑料模型、木質(zhì)模型或是3D模型，意義大得多，更何況它仍在不斷向前發(fā)展。在溝通層面上來看，實物模型在設(shè)計表達的即時性上有出色的表現(xiàn)，但是它的生產(chǎn)過程，同把實際建筑縮減到一個模型一樣，貧乏可憐。迅速發(fā)展的原型理念遠遠比其他模型手段更先進。無論是利用CNC工具，還是簡單的立體印刷技術(shù)，運算化設(shè)計都能讓設(shè)計師對設(shè)計進行有效的評估，而這一點在機械制造廠房是不可能達到的。盡管有些設(shè)計師和建筑師仍然雇用好的木匠來完成模型，然而設(shè)計與工具已經(jīng)由網(wǎng)絡(luò)連接起來，我們可以利用虛擬現(xiàn)實或物理3D模型完成遠程原型設(shè)計。

5 什么是原型（Prototype）？

為了更好地闡述設(shè)計的應(yīng)用，我們需從一個基本概念談起。設(shè)計在很多場合下，其實并不是為了制作真正的東西，而是先設(shè)計原型，再轉(zhuǎn)變?yōu)檎嬲漠a(chǎn)品，比如報紙、自行車，或是一個新的時尚路線。以前，生產(chǎn)線很長，設(shè)計線就相應(yīng)的很長。但現(xiàn)在情況已經(jīng)不同了，我們生活在一個日新月異的時代，更強調(diào)“就趁現(xiàn)在”或是“即時市場”。從最初的設(shè)計理念到運輸、發(fā)行，時間大大縮短了。設(shè)計與生產(chǎn)相互獨立。過程簡化會帶來一些風(fēng)險。

在設(shè)計階段之后的快速原型制造，也成為運算化的組成部分。在這方面，平面設(shè)計師又一次走在了前端，他們最先利用數(shù)字技術(shù)的“快速原型制造”進行打樣和印前測試。各地服務(wù)部門都可以遠程執(zhí)行從排版設(shè)計、校色到印前設(shè)計的工作，使設(shè)計師從“藝術(shù)家”走向了客戶端。近些年，已經(jīng)出現(xiàn)了紡織原型的“虛擬織機”，為產(chǎn)品開發(fā)組建的快速原型制造服務(wù)部門也相繼成立。圣地亞哥超級計算機中心對網(wǎng)絡(luò)遠程原型制造提供了相應(yīng)的技術(shù)支持。

當(dāng)然，相對于傳達設(shè)計和材質(zhì)設(shè)計的打樣階段，工業(yè)設(shè)計中的3D原型技術(shù)要復(fù)雜得多。比如，要驅(qū)動激光打印機，需要生成“后腳本”（Postscript）文件，但是我們往往不知道該如何做好驅(qū)動快速原型制造設(shè)備的.STL文件（該類型的文件會把模型表面轉(zhuǎn)變成為三角面網(wǎng)），為構(gòu)建3D模型服務(wù)?？焖俪尚图夹g(shù)最初是作為數(shù)控機床切割消減的程序，就好比雕塑家處理大理石或者木材中多余的部分一樣。另外一個應(yīng)用是立體打印技術(shù)（適當(dāng)光照下液態(tài)感光），與消減性技術(shù)不同，它是一種添加性的裝置，分為選擇性燒結(jié)（利用激光束融合熱塑性粉末）和霧滴性沉積（在薄陶瓷層或金屬粉末上添加粘合劑）兩種。我們也可以將添加技術(shù)和消減技術(shù)相結(jié)合，就好像熔融沉積模型（融化熱塑性材料在由設(shè)計的形式進行“打印”）和層壓實體制造（通過層層成型然后壓合得到層壓物體）。

顯然，設(shè)計師并不需要是熱塑性融合或立體打印技術(shù)的專家。他們了解計算機輔助設(shè)計及快速原型制造技術(shù)的原因是，設(shè)計表達與加工制造（計算機輔助制造）的關(guān)系越來越緊密。而且，設(shè)計師也必須認(rèn)識到正是由于這項技術(shù)的出現(xiàn)，設(shè)計任務(wù)已經(jīng)由最初的從已有形式中選取，轉(zhuǎn)向了發(fā)明新的形式，設(shè)計新分子、新基因、新材料，甚至是和人互動的新形式。事實上，運算化設(shè)計的背景下，設(shè)計需要完全融合審美需求與功能要求，形式不再跟隨功能，而是成為功能。為了達成這個目的，設(shè)計師不再僅僅作為訂單和美學(xué)的代理人，不再把以前的那些“臟活”留給工程師去做。

我還想提一下“理想化”這個詞，主要考慮到很多設(shè)計師還保有懷舊情結(jié)。我要強調(diào)的是，實際上理想的運算化模型，是各種特征都可以通過可變參數(shù)模擬出來。有人把這看成是運算化設(shè)計的缺點，但我卻反而覺得這真是運算化設(shè)計的強項。過去，模型只能展示出在選定材料前提下的有限的形式特點；但運算化設(shè)計卻能夠體現(xiàn)多種可能合適材料的形式可能性，讓設(shè)計師超越固有的限制。那些對數(shù)字化表述持懷疑態(tài)度的人，其實是沒有認(rèn)識到人類活動的主體，是處在認(rèn)知的理想領(lǐng)域之中，而不是僅僅滿足需求的層面，后者會限制某些新技能的培訓(xùn)（體現(xiàn)在仍然使用昔日的機器和工具上）。

6 設(shè)計與希冀

作為創(chuàng)意主體的人類，最大的強項并不是對外在世界和自然變化的反應(yīng)，而是希冀（對未來變化的希望并為此做出行動，譯者注）。運算化設(shè)計的本質(zhì)就是希冀，即前攝性。換句話說，運算化設(shè)計強調(diào)，由未來決定系統(tǒng)現(xiàn)有狀態(tài)的事實，定義的概念領(lǐng)域。這可能這聽起來難以讓人相信，讓我們的思緒轉(zhuǎn)向不是預(yù)言，就是技術(shù)。但是仔細想一下，我們就會意識到如果沒有希冀的因素，那設(shè)計只能是一種被動跟隨的游戲，一種對變化的消極反應(yīng)，而失去了作為變化中介的角色。昨天那個決定性的口號——設(shè)計是一個解決問題的過程——依然縈繞于耳，以致我們難以確定是否已經(jīng)真正實現(xiàn)了它，這簡直就是一場游戲。形成鮮明對比的是不斷的再包裝（一系列基于同種元素只是風(fēng)格不同的咖啡機、烤箱、汽車、收音機和計算機），運算化設(shè)計需要且支持發(fā)明創(chuàng)造。在環(huán)保意識日益加強的今天，運算化設(shè)計挑戰(zhàn)了一次性解決所有問題的設(shè)計策略。通過問題生成，運算化重新定位了在當(dāng)今環(huán)境和瞬息萬變的社會生活中的個人。運算化設(shè)計平等對待個體與所處的環(huán)境文脈，并將最后終結(jié)大規(guī)模生產(chǎn)的時代，讓人類進入一個提供個性定制化的解決方案的新時代。為了更好地解釋這個問題，我需要再回到之前關(guān)于“實用(pragmatic)”的討論中。

實用的大環(huán)境是對工作中某些特定的“力”、開發(fā)的能源和社會的政治結(jié)構(gòu)的響應(yīng)。史前獵人和強盜的設(shè)計需求、期望與農(nóng)業(yè)和畜牧業(yè)時期的人肯定大不相同。即使在今天，因為設(shè)計定義了工匠和工廠勞工所處的環(huán)境和工作，他們同設(shè)計的關(guān)聯(lián)，必定不同于教師、物理學(xué)家、科學(xué)家和藝術(shù)家那樣。工業(yè)革命引起了許多同設(shè)計相關(guān)的問題，把世界變分成了許多不相關(guān)的板塊。想想家里各種電器，或者辦公室和工廠的各種設(shè)施設(shè)備，各自構(gòu)成了圍繞個體的“世界”本身，有自己的生活法則。信息時代帶來的是世界一體化，把之前斷裂的“板塊”連成一個復(fù)雜而高效整體。設(shè)計只有考慮了人在各類不同環(huán)境下的差異，整合各種任務(wù)，才能更好地解決能源消耗、環(huán)境問題，更好地實現(xiàn)人與人的互動。

運算化設(shè)計也相應(yīng)地為此構(gòu)建出理論框架，并透過實踐達成上述目標(biāo)。顯然，一體化帶來了信息繁雜的問題。越來越多的按鈕和按鍵，無論設(shè)計得多么優(yōu)雅，我們還是很難掌握機器復(fù)雜的使用方法。因此，設(shè)計師應(yīng)該通過設(shè)計，更好地控制復(fù)雜性。如今的情況就是，每個機器都只能發(fā)揮20%的作用。設(shè)計僅僅停留在玩味復(fù)雜的形式，并不能充分利用現(xiàn)有的技術(shù)有效地幫助用戶解決問題。

7 設(shè)計與廣泛存在的運算化

20世紀(jì)早期，電力的飛速發(fā)展與網(wǎng)絡(luò)技術(shù)的普及，使得運算化設(shè)計不斷推動全球經(jīng)濟的進步。電力、電話與電視構(gòu)成了世界的底層結(jié)構(gòu)。同樣，數(shù)百萬人已經(jīng)通過各種事物，從網(wǎng)絡(luò)數(shù)字互動及先進的運算一體化中獲益。運算化在電話、無線通訊、手表、家用電器、汽車卡車、飛機、自動柜員機、娛樂和教育等方面都有體現(xiàn)。與運算化設(shè)計機密相關(guān)的數(shù)字化技術(shù)的各種應(yīng)用，還都只是在起步階段。運算化設(shè)計應(yīng)該積極承擔(dān)起加快進程的作用。一兩個設(shè)計師決定用不用計算機進行設(shè)計無關(guān)緊要，普遍的變化并不以小范圍的異議為轉(zhuǎn)移。昨日許多設(shè)計師還在宣稱抵制桌面出版程序，然而現(xiàn)在，盡管有些程序還是以前的狀態(tài)，甚至有些顯示出較大的缺點，但是一個眾所周知的事實是，現(xiàn)實中沒有程序使用技能的設(shè)計師在設(shè)計領(lǐng)域中，已經(jīng)很難找到工作了。工作需要以及全球經(jīng)濟特點，都顯示出如果我們能夠認(rèn)識到現(xiàn)在的實情，就會有更多的選擇、更多的可能性。運算化設(shè)計的前景相當(dāng)樂觀。

在我們正在經(jīng)歷根本性變化的環(huán)境下，設(shè)計的新任務(wù)，源自人類務(wù)實的認(rèn)知感，設(shè)計教育也會受到影響。因此設(shè)計實踐與設(shè)計教育都需要做到前攝性，不能僅僅作為技術(shù)進步的反映，就是說要將運算化或其它形式的信息處理媒介，都變成設(shè)計的一個有機組成部分。簡言之，在工作室或大學(xué)課程設(shè)計中，書籍、海報、宣傳冊、汽車、烤箱、椅子、臺燈都可以作為設(shè)計的課題。相反，只知道如何設(shè)計這些產(chǎn)品，并不代表設(shè)計師有足夠的應(yīng)對新問題的能力。僅僅利用計算機來美化設(shè)計，把它當(dāng)作傳統(tǒng)表現(xiàn)工具一樣使用，必然效果不大、不盡人意。在設(shè)計新產(chǎn)品時，電腦必須創(chuàng)造性地整合到設(shè)計的過程中。為此，整個計算機工業(yè)雖然一直竭盡所能在做，但卻一直沒有找到行之有效的辦法。計算機行業(yè)的人，最多只能想道要在設(shè)計領(lǐng)域達成這個目標(biāo)，必須有更快的芯片、更大的儲存容量及更好的壓縮方案，僅此而已。因此，無所不在的計算機革命進程中，運算化設(shè)計將推動設(shè)計師成為計算機領(lǐng)域的合作伙伴。

功能主義的思想仍然回響在運算化設(shè)計項目中。拋開辦公桌上笨重的設(shè)備，遠離人人變成打字員的苦惱，運算化的普及，提出了與各類無形的數(shù)字設(shè)備互動的新觀點。運算化取代了對更好界面的迷戀，通過計算機的集成能力，在提供實現(xiàn)人類公平和任務(wù)合理的設(shè)備與工具方面，可以更好地表現(xiàn)。和任務(wù)脫離的計算機應(yīng)該引起我們額外的關(guān)注，因為只有重新和任務(wù)的目的聯(lián)系，數(shù)字化技術(shù)才能充分實現(xiàn)我們的意愿。運算化設(shè)計的主要目的就是讓信息處理集成能力，成為人類能力和思想的有效補充。要有電燈泡，并不需要知道電廠如何工作，也不需要了解如何處理變壓器；要使用洗衣機也不用考慮那些集成運算；要獲得天氣預(yù)報、旅行援助和游覽信息，也是如此。新產(chǎn)品、新汽車、錄像機、家具的設(shè)計都應(yīng)該了解用戶的需求，醫(yī)療設(shè)備應(yīng)該同時為護士和病人所有，甚至各種智能工具，也應(yīng)該所有人都可以操作的。運算化就應(yīng)該像運動鞋一樣，穿在誰的腳上都能很舒服。我們應(yīng)該信手拈來直接使用，不需要什么過多的培訓(xùn)或者教材。其接口使用界面的設(shè)計是運算化設(shè)計的關(guān)鍵，這點應(yīng)該是顯而易見的，界面設(shè)計一定是運算化設(shè)計的重要組成方面。而界面設(shè)計同設(shè)計本身一樣，是無形的，與設(shè)計的對象和信息形成有機的整體。在技術(shù)飛速更新?lián)Q代的今天，目標(biāo)決定了設(shè)計的主要任務(wù)。

8 設(shè)計研究：變化的動力

隨著運算化設(shè)計的出現(xiàn)，設(shè)計邁入了一個劃時代的新世紀(jì)。作為建立人類活動新務(wù)實大環(huán)境的參與者，設(shè)計革新可能進一步分化我們的工作。因此，權(quán)利開始下放，等級結(jié)構(gòu)也逐漸消失。設(shè)計領(lǐng)域內(nèi)早就開始了這樣的變化，雖然沒有我們想象中的那么順利，但已逐漸顯現(xiàn)出一些效果。還有更多的變革即將到來，也許過程會更費力，但會影響到整個行業(yè)，因為過程中需要尋求更高效的水準(zhǔn)，以維持和給養(yǎng)全球經(jīng)濟。我們所處的時代，變化的速度與創(chuàng)新的速度持平，設(shè)計師不得不走到最前沿。這也正是為什么延緩的策略，可以在變化緩慢的社會里生存，但今天必然不能奏效。那些不能適應(yīng)快速變化的手段和方法將最終被淘汰。壞消息是，當(dāng)今的競爭環(huán)境下，設(shè)計領(lǐng)域的破產(chǎn)率空前的高；好消息是，越來越多的革新派設(shè)計師，開始以各種形式運用運算化設(shè)計，在激烈的市場競爭中找到解決之道，成為進程中的標(biāo)兵。昨天還是格林威治一個普通的小商店，今天就可以利用新媒體、新材料、新交互形式提供各式的服務(wù)。名片和辦公用品的設(shè)計師，某天被那些酒店大堂、公共汽車站和火車站里的投幣機取代，也不會讓人大跌眼鏡。新設(shè)計將越發(fā)關(guān)注人類的心智。網(wǎng)站可能不是個人能擁有的最高目標(biāo)，但是如果站在人類正在空前的彼此鏈接、互動合作的層面上，網(wǎng)站要遠比那些高級汽車、燈具或者明信片上為文盲閱讀的白癡般的文字，更有意義。

隨著運算化設(shè)計的到來，設(shè)計終于可以破天荒地確立屬于自身的研發(fā)領(lǐng)域，不再需要等待其他學(xué)科的發(fā)展和需求。運算化讓設(shè)計研發(fā)本身變成一股新變化的力量。

Knowledge is becoming increasingly computational. Previous means and methods for the acquisition, communication, and criticism of knowledge are being replaced by inquiry,dissemination, and evaluation carried out by digital means. Pascal, Leibniz, and Peirce, among others, prepared the conceptual framework for this fundamental change. They asked questions regarding our motivation to know, our way of acquiring knowledge, and our desire to share it. In other words, they defined the cognitive horizon. Closer to our time, Boole, Wiener,and von Neumann provided the scientific foundations. Finally, Atanasoff, Zuse, Eckert and Mauchly (among others) built the machines.The rest is already part of our lives: computer graphics, visualization, desktop publishing, CAD,multimedia, virtual reality, Internet, World Wide Web with more to come. In the process, sciences became computational: physics, biology, chemistry,to name the best known. Many engineering endeavors took the same turn with the synthesis of materials, robotics, even the production of computers, and the automatic generation of software. What happened to design in this context of fundamental change?

A Snapshot of the Current Situation

As things stand, computers and design are merely an association of tools and users. Indeed,within the design community, the discussion still goes on whether the computer will ever replace the designer, or if it will at least replace the pencil and the marker, not to mention the tedious process of model building. Graphic designers are very much ahead of the rest, plowing happily in the new territories of typeface design, print on demand,and electronic publishing. They discovered very quickly that digital technology means not only better tools for old functions, but also a broadening of the scope of their activity. The laser writer, the scanner, the plotter, the compact disk, and more recently network tools (browsers, applets, frames)were integrated in a new practical effort. So were the methods and means of science sampling,splicing, mutations, hyperlinking. As a result,printed paper is complemented by multimedia and Internet-based communication. Exemplary of the effort I am referring to is also the new practice of communication design: the virtual design office. Indeed, in this case designers designed their own new context of interaction based on the technologies and the methods they work with. Thus the computational becomes constitutive of the work, and is tested as the work itself is subjected to evaluation. But even in graphic design, fundamental issues are still avoided: Do we address a generic human being, who has remained the same as science and technology have changed? Or do we "design"our own public, i.e., invent forms and means for more individualized, and still socially rooted, forms of human interactions? How do we transcend the dominant obsession with mass communication(broadcasting) and make narrowcasting a design goal equally significant in respect to contents and expressive means? Do we improve on what we inherited or do we participate in the renewal of the motivations and means of communication?

Technology, even as it is creatively applied in communication design, is still ahead of us. In other design activities, and primarily in what is called product or industrial design, the situation to date is less promising. While the old-fashioned industrial design practically stopped generating employment opportunities, educational programs are slow in acknowledging the need for integrating the digital.The educators involved still think in the solid terms of the model of the Industrial Revolution, terms that are based on formal expectations of crafting but not on the need for new design thinking. As we know, the investment in technology hardware,software, maintenance, training, research of new avenues is prohibitively high. Few have dared to take the risks of entrepreneurship, and even fewer have succeeded. Big companies consolidated their controlling positions, and literally sucked in everyone able to manage the complexity of computer-based or computer-aided design.In many cases, instead of making design more transparent, they insulated themselves under the very convincing argument of protecting intellectual property instead of disseminating it. It is not unusual that advanced product design teams using advanced computers and sophisticated software do not even have access to the Internet. While those involved in digital technology attempt to produce viable methods of cooperative design work, such teams are predicated to a monastic type of activity.More often than not they do not even notice the contradiction between the means used and the methods and structures of work. Consequently,they maintain the secrecy (of new car models, new toys, new furniture, etc.), but are always late on the market.

Technological lead over design considerations is radical not only in the area of industrial design. It is also manifest in textile, fashion, toy, and interior design, all forms of design still close to the paradigm of craftsmanship. Consequently, monstrosities of all kinds, conceived with the aid of some computer programs, spill over to the consumer in the supermarkets of discounted bad taste.No matter how "noble" the intention of making affordable every gadget that until now was in the exclusive realm of the military and the intelligence communities, it only rarely justifies their presence in our culture.

About the Possibility of Design Theory

Computational design acknowledges the association between tools and users. However,its goal is to turn this into an association of new possibilities, which should become realities through design. To achieve this goal, computation cannot be only, as it is today, a medium of representation and unsystematic, or even systematic, variations. It has to become constitutive of design. This brings to the forefront the older question of whether design theory is possible, and if yes, which form it can take.

In the past, to the extent it was formulated,theory has followed the practice of design. The best designers, or at least those able to articulate their thoughts in writing, rationalized their achievements;that is, they discussed what they did and how only after their design was acknowledged or received public acclaim. This situation should not surprise anyone. Design evolves, as we all know, from the crafts and in this evolution, it first has to acquire legitimacy among many other human endeavors.But as it develops its means and methods, it also produces its justification and conceptual horizon.With the emergence of design criticism and design history, obviously in connection with the establishment of design education, the possibility of theory is established. Such a theory had to be analytical at the beginning. In time, induction acquisition of knowledge through observation was complemented by deduction derivation of new knowledge from design generalizations.As a result, design theoreticians were able to venture into synthesis. The example of the Russian Constructivists, or of the Bauhaus, or of American design after World War II belong to the domain of new concepts. Some were adopted from morphology, structuralism, semiotics, and even from psychology, linguistics, sociology, and engineering. Others were derived from within, the best example being functionalist design. In recent years, algorithmic thinking, heuristic procedures,and even genetics found their way in the theory of design. Moreover, design hypotheses were computationally modeled and tested. My own Design Machine? can be mentioned as an example in this direction.

Theories attached to discursive reasoning remain captive to the deterministic equation: there is a cause, i.e., design work, and there is a result,i.e., designs that become identifiable objects traded or culturally recognized for their characteristics.So it ought to follow that a theory should explain how people design and what good design is. Here things get murky. First of all, because language as we know it might be the best medium for our reciprocal understanding, but not necessarily for handling human activities that by their nature are not reducible to language. Second, because the romantic assumption within discursive reasoning is that good design "good" being defined in a given context (formal, functional, structural, etc.) is also successful. Obviously, a good design theory should explain why sometimes this is not the case. As this kind of questioning in and with the help of language is established, we have learned that design theory is inter- and transdisciplinary. These are good words to use in applying for a grant, but not necessarily helpful in practicing design theory, or in designing.Nevertheless, the result of this understanding explains the import of specialized language in design. Ergonomic, functional, psychological,sociological, and economic concepts invade the dialogue on design issues and the curriculum of design education. Designers speak to future clients more about ergonomic, cultural, or symbolic aspects than about design itself. More recently, the language of lawyers is being added to the wholesale package of design theory, since the practice of design also means protecting its products in a society inclined to protect the written, but not necessarily the more ambiguous visual expression.

A Design Knowledge Base

Computational design escapes this Catch-22 situation. It is, like any other form of computational knowledge, anchored in the pragmatics of human existence. As we know, computational physics is at the same time theory and practice. As theory, it produces hypotheses regarding the beginning of the universe, for example. As practice, it simulates them in order to test the validity of the premise,and it eventually transforms them into new tools for the investigation of the universe. Simulations serve further to derive new knowledge regarding our inquiry of the universe. They also help us to understand the meaning of this knowledge for our own activity, regardless of whether we are physicists or professional involved in other fields (biology,chemistry, philosophy, art). Such knowledge is proactive, in the sense of opening new avenues for practical endeavors. Think about the many experiments with plants, animals, food or even with art performed in outer space. Computational engineering synthesized new materials some very interesting for designers and as a result also opened new avenues towards the future. It starts from hypotheses at the molecular or atomic level. Its results are the new structures modeled and tested in computational form before any other natural resources are processed. Computational genetics is a practical activity having at its center human wellbeing.

Computational design means, then, design activity driven by the forces that make design possible and necessary in the first place: assessment of needs, assessment of possibilities, assessment of means as they embody human characteristics. The assessment takes the form of data, in particular,complex databases. But while any other design theory is by its nature reactive, based on opinion,and thus often speculative, a computational design theory is based on processed data and is by its nature proactive. Its limits are the limits of our ability to collect and meaningfully organize data regarding quantity as well as quality, and our ability to design effective computational procedures for their processing. Like any other computational theory it is at the same time practice, more precisely design practice in the broader context of extremely differentiated forms of human activity,such as those we experience today. It is subject to confirmation by test, and it is, first of all, centered on knowledge, the most important asset human beings have. Accordingly, it requires that we establish a design knowledge base that extends beyond the poor, or even less than poor, design museums and collections, books and articles about design. Furthermore, it requires that we design procedures for navigation, search, and retrieval in such a knowledge base, evidently conceived at the global level of human existence today. Artifacts,along with the plans and designs from which they were derived, need to be seen together from a broad cultural perspective. Such a knowledge base should also contain computationally expressed knowledge regarding visual representation, movement, color,ergonomy, the integration of other means of communication (sound, texture, smell, etc.). All these objectives are a tall order, but unavoidable.Unfortunately, the majority of our design museums and collections, the places where we look at design as a "school of the past", resemble a junkyard more than a knowledge base for design.

As examples of what belongs in our design knowledge base, as it started to become a reality, are the computer programs that the design community uses. Indeed, a CAD program, or one for the production of a new font, a multimedia composer,or a net browser is already a theoretic expression of high abstraction. Within such a program, we describe geometry, material characteristics, optics;we describe movement, perspective, associations of images and sounds, ways to integrate text, and many other components of design. Not all of them are captured together in such programs, of course, but at least those about which a design consensus has been established. Or those we understand better.The practice of design based on such "theories"is, then, the research of actual design assignments.And the evaluation of the design is the performance of the artifact digitally conceived. In successive versions, benefiting from the experience of design such programs improve. As I write these lines,Netscape? 3.0 is being announced; it will integrate teleconferencing, which makes my next statement self-explanatory. In the succession of design hypotheses, some disappear because the theory they advance proved inappropriate. Only two years ago,teleconferencing, a major communication design idea, was a potential multibillion dollar market.In our days, it is becoming a standard browser function.

Let me make the idea of design as program more clear: The Macromedia Director?, or the Phontographer?, or Alias?, or Vellum?,or those programs used for desktop publishing(Quarkexpress?, Pagemaker?), for textile design,for jewelry, etc., are programs we can buy in stores and use for particular jobs. But as opposed to the pencil, brush, exakto knife, wood or metal type, composer stick, etc. that designers used in the past, such programs are condensed theories of the activity they support or invent (as was the case of teleconferencing). None describes design completely. They describe and synthesize design activities related to our interest and need for multimedia, font design, or for CAD, for publication design or for on-line advertisement.Those who authored such programs, quite often large teams of programmers, psychologists,designers, etc. integrate in them knowledge of physics, mathematics, aesthetics, semiotics, of ergonomy, etc. In fact, each such program is a theoretic hypothesis. Those using them test this hypothesis. The products that are finally generated are comparable to the products that result after computational engineering is applied for creating new materials, or computational genetics for creating new medicines.

Computers Are NOT Only Tools

In fact, new materials, new medicines, and new genes are designed. I use this term to suggest that design is becoming a very broad endeavor in the age of computation. If we do not understand the necessity of computational design, we only continue the metaphysical talk about how computers are only tools. Or we continue the poetic description of how design originates, like Venus from the head of Jupiter, in the head of designers. Or how intuition explains what indeed some programs still cannot achieve, not because they do not have intuition(which they don't have to have), rather because in using them, we are not yet as comfortable with them as to use them creatively.

Let us face it: many aspects of design can be carried out perfectly without any use of computers. Such aspects are not really the object of computational design. After all, computational design does not replace design, it continues and broadens design in a new pragmatic context. The real challenging aspects of design in our times are exactly in the realm where without the new design knowledge in its computational form, we could not come to viable solutions. Consider the design of the Hubble telescope, and consider further its fixing, after it was launched in a defective state and started its journey around the earth. It was in a computational design model, involving means and methods of virtual reality, that the design error that almost rendered the telescope useless was diagnosed and procedures for improvement,including design of tools appropriate to the task at hand, generated. This is why computational design integrates modeling, rendering, animation, but also simulation (including virtual reality). That this level is only timidly reached should not prevent us from understanding that the digital model resulting from a comprehensive computational design work is infinitely more telling than the Styrofoam, or wood,or polymer 3D artifacts that so many continue to idealize. As conversational pieces, models convey a beautiful quality of immediateness. However, for the production of the real objects, they are as poor as any reduction of the real to a model. Moreover,the emerging rapid prototyping is far ahead of any other modeling endeavor. Whether driving CNC tools or even performing modest stereolithography,computational design allows a designer to reach a level of evaluation that is not possible in the mechanicas shop. Instead of hiring a good carpenter, as some designers and architects still do,we can perform, even today, remote prototyping either in the form of virtual reality or in physical 3D.Design and tools can be connected via networks.

What Is a Prototype?

In order to clarify the design implications, let us start with a conceptual framework. To design is not to make the "real" thing, but the prototype of what will become, for example, a newspaper, a bicycle, a new fashion line. In previous times, when production cycles were long, design cycles were also relatively long. This situation has changed. We live in a day-and-age described by "just-in-time"or "time-to-market." From concept to shipment and distribution, time has been reduced by many orders of magnitude. The design process and the fabrication process are interdependent. With the risk of some simplification, generic diagrams give an idea of the process.

Rapid prototyping everything following the design phase as a computational component,deserves at least some words of explanation. First of all, graphic designers were again in the forefront since they started "rapid prototyping" by using digital technology for proofing and pre-press evaluation. Service bureaus all over the world perform, remotely, everything from typesetting to color correction and pre-press functions all that it takes for a design to make it from the "artist" to the client. In recent years, textile prototyping on "virtual looms" became possible and rapid prototyping service bureaus for product development started opening, too. The San Diego Supercomputer Center supports remote prototyping on the Internet.

Sure, prototyping in 3D, for industrial design purposes, is a more complex enterprise than proofing for communication design, of for textile design. We know how to generate good postscript files to drive laser printers, for example. But we are far less good in generating the so-called .STL files that drive RP devices. Such files employ a surface representation defined by triangles and serve in the fabrication of 3D models. RP technology started as a subtractive process a numerically controlled (NC)machine chiseled away, pretty much like a sculptor does working on marble or wood, what was not necessary. Today it offers additive mechanisms in the form of stereolithography (liquid photopolymers solidify under the appropriate light), selective sintering (the fusing together of thermoplastic powder by using a laser beam),droplet deposition (laying down of an adhesive liquid over a thin layer of ceramic or metal powder).We even have a combination of additive and subtractive processes, such as in fused deposition modeling (the melting of a thermoplastic material and its further "printing" in the designed form) and laminated object manufacturing (a laminated object is processed from layers of paper).

Obviously, designers do not have to be experts in thermoplastic fusion or in stereolithography.But they need to think in terms of computer-aided design (CAD) and rapid prototyping (RP), because the connection between representation (in design)and actual fabrication (through computer-aided manufacturing CAM) is getting tighter. Moreover,they need to realize that due to such technology,design tasks shift from the traditional expectation of giving form, of Gestalt, to inventing new forms,some as exotic as the design of new molecules,new genes, new materials, new forms of human interaction. Indeed, in the computational design context, aesthetic considerations and functional characteristics need to fuse. In order to accomplish this goal, designers can no longer restrict themselves to being agents of order and beauty, leaving the"dirty job", as to how things work, to engineers.

Having mentioned the word idealize in reference to the nostalgic view some designers still have, I need to confirm that, in effect, the digital model is in the realm of the ideal, where characteristics are simulated and can be optimized by varying many parameters. Some see here the shortcoming of computational design, although it is its strength. In the past, models could only display characteristics of available materials.Computational design models make the question of appropriateness of materials possible. They challenge the designer to go beyond what is available. Those who feel insecure about the ideal nature of the digital representation fail to realize that the majority of human activity is in the ideal domain of the cognitive, not in the necessary, but somehow limiting training of skills (quite often on machines and tools of yesteryear).

Design and Anticipation

The strength of the human being, as a creative entity, is in anticipating, not in reacting to the outside world and its natural changes.Computational design is by its nature anticipatory,proactive. In other words, it addresses a conceptual realm defined by the fact that the current state of a system depends on its future. At first, the thought sounds dubious. It brings to mind predestination,or teleology. But once we consider the idea, we understand that without the planning element,which is anticipation, design remains a catch-up game, a form of reaction to change, instead of being an agent of change. Design as problem solving, the slogan of a deterministic past so close to us that we are not sure whether we have overcome it, was such a game. In contrast to continuing the line of a practice of re-packaging (all the series of coffee machines, toasters, cars, radios, and computers,based on the same components but stylized differently), computational design involves and supports invention. It challenges the once-andfor-all solution, especially in view of an increased ecological awareness. It generates problems as it takes an active role in repositioning the individual in our environment and in an extremely dynamic social life. It does justice to the individual and to the particular context of existence as it brings mass production to an end and facilitates customized solutions. To explain this component, I need to briefly revisit previous pragmatic contexts.

Pragmatic contexts correspond to specific forces at work, energy sources tapped, social and political structures. The prehistoric hunters and foragers had design needs and expectations very different from those of the humans involved in agriculture and animal husbandry. Craftsmen and factory laborers, even in our day, relate differently to design as it defines their living environment and their work than do teachers, physicians, scientists,artists. The Industrial Revolution posed many design problems. It also broke the world into many unrelated pieces. Think of all the appliances in one's home, or of the many tools in our offices and factories. Each makes up a world in itself,with its own rules for performing appropriately.The information age brings about the possibility of integration. Issues of energy consumption,environment, and better human interaction, issues of cultural diversity can be better addressed if we design with the aim of integrating human tasks without ignoring the differences among people living under different conditions.

Computational design should accordingly constitute the conceptual framework for such a task and become the practice of accomplishing it. Evidently, as integration takes place, we have problems in dealing with complexity. More buttons and more keys, no matter how elegantly designed, do not help in our command of the new complex machines. Accordingly, designers need to work on giving through design a better control of complexity. Otherwise, each wonderful new machine will only be used to 20 percent of its actual capacity which is the situation today. Design stuck on formal considerations does not effectively help users get the most out of what is technically possible today.

Design and Ubiquitous Computing

The expansion of computation through networking, which contributes to the dynamics of the global economy, and through ever increasing performance parallels the deployment of electricity as it took place earlier in the 20th century. Electricity, telephony, and television form an integral part of the underlying structure in many parts of the world. Similarly, millions of people already benefit from digital interaction through networks and from the progressive integration of computation in human transactions of all kinds. Computation is integrated in the telephone, in many services associated with wireless communication, in wristwatches, in home appliances, in trucks and automobiles, in airplanes,in automatic teller machines, in entertainment and edutainment. Compared to the state of computation, the creative use of digital technology is only at its beginning. Computational design should assume the goal of actively speeding up the process. It is irrelevant whether one or another designer decides not to use the computer. The dynamics of the process is such that the broader change does not depend upon such decisions.Many designers resisted the change announced by the desktop publishing programs of yesterday.As primitive as some of these programs were, and some failed in the meanwhile, they opened a new horizon and led to a reality expressed in the simple fact that those who do not master such a program cannot find a job in the design industry. Forces at work, characteristic of the global economy, define further directions which, if acknowledged and properly understood, allow for more variety and the unfolding of more possibilities. The underlying dimension of computational design is optimism.

The new tasks of design in the context of the fundamental change we are experiencing result from the recognition of the new fundamental pragmatic condition of the human being. The tasks of design education cannot be less affected by this condition.Therefore, to practice design and design education proactively, not merely in reaction to technological developments, means to make the medium of computation, and any other information processing medium, part of design. In short: not that books,posters, brochures, or cars, toasters, chairs, and lamps are invalid design subjects, in the studio or in college education. Rather, knowing only how to design such items does not prepare a designer for those qualitatively new problems we are facing. To use the computer for design cosmetics,doing what traditional tools can do just as well, is unproductive and unsatisfying. The computer has to be creatively integrated in the design process, in the new products designed. This is something the computer industry does not know how to do but is trying desperately to achieve. Those who work in the computer industry know that faster chips, more storage capacity, and better compression schemes are only means to a goal that is fundamentally in the realm of design. Accordingly, computational design will make designers become partners in the ubiquitous computing revolution.

The functionalist thought is echoed in the ubiquitous computing design program. Instead of the bulky machine on everyoneas desk, and instead of turning each user into a typist, ubiquitous computing offers the perspective of natural interaction with many "invisible" digital devices.It replaces the obsession with better interfaces,as a hope for better user performance, through integration of computer capabilities in appliances and tools that do justice to the human being and to the task at hand. A computer isolated from the task at hand requires excessive attention. Once reconnected to the purpose, digital technology enhances our ability to fulfill the purpose. The integration of information processing capabilities in ways that complement people's abilities and their ways of thinking is a major goal of computational design. In order to benefit from the electric bulb,one does not have to learn how a power plant works, even less how to operate a high voltage transformer. The same should be the case for people using active maps to obtain weather reports,travel assistance, or tourist information. Or for those using the new washing machine that integrates fuzzy logic computing. New products cars, VCRs,furniture that "learn" the behavior of the user,hospital equipment that assists the nurse as well as the patient, intelligent tools of all kind, should not require a college degree to operate. Computation should fit us as comfortably as a pair of sneakers.And we should be able to use it when necessary without having to study volumes of printed matter or to go through extensive training. That interface design is a major aspect of computational design should be obvious. Less obvious is the fact that the best interface design, like design itself, is invisible,i.e., integrated in the object or message designed.These are goals that define design tasks in a context of fast technological renewal.

Design Research: a Force for Change

With the advent of computational design,design enters a new phase of its remarkable history. As a participant in the establishing of a new pragmatic framework for human activity,design innovation makes possible distributed work.Accordingly, it contributes to decentralization,and to the disappearance of hierarchic structures.Within the design community such changes already take place, not always as smoothly as we would hope for, but definitely with the effect of a higher sense of responsibility. Much more will take place,and probably even more painful changes will affect the profession as it seeks its justification in a society determined to achieve levels of efficiency high enough for the sustenance of the global economy.As we reach the time when the rate of change equals that of innovation, designers are forced into the forefront. This is why procrastination, a survival tactic in times of less fast change, will not do. This is also why means and methods not adapted to these fast cycles of change fail. The bad news is that in the competitive context of today's world,the bankruptcy rate in design is higher than ever.The good news is that more and more innovative designers, definitely aware of computational design or practicing it in some form or another, make their way in the competitive market of innovation and become icons in the process. Where yesterday in Greenwich Village were the gadget shops, today design shops offer a variety of services based on new media, new materials, new forms of human interaction. By no accident are the designers of business cards and stationery replaced by coinoperated machines placed in hotel lobbies, bus depots, and train stations. New design addresses our minds more and more. Maybe a Website for an individual is not the highest goal one can have,but to think in terms of human interconnectedness and cooperative effort is of a higher order than to stylize cars, lamps, or to produce idiotic messages on postcards for illiterates.

With the advent of computational design,design finally defines its own domain of research and development. As a result, instead of waiting for other disciplines to define its agenda or scope of inquiry, computational design makes design research a force of change.

注釋：

Note:

①Boole, George. (1815-1864) conceived of a logical calculus in An Investigation of the Laws of Thought on which are founded the Mathematical Theories of Logic and Probabilities (London, 1854).

②von Neumann, John, the legendary mathematician, was also instrumental in the paradigm of sequential computing. He was aware of the ENIAC (Electronic Numerical Integrator and Calculator) built by J. P. Eckert and John Mauchly) and in 1945 wrote the famous First Draft of a Report to the EDVAC (Electronic Delay Storage Automatic Computer).

③Regarding the Design Machine (a research project carried out in 1985-1988).

④Regarding Anticipation.

⑤Nadin Mihai. Mind Anticipation and Chaos (German-English parallel text, from the series Milestones in Thought and Research)[M]. Stuttgart/Zürich: Belser Verlag,1991. Develops a cognitive model based on chaos and anticipation.

Reference:

[1]Wiener Norbert. Cybernetics[M]. Cambridge MA: MIT Press,1948.

[2]Nadin, Mihai, Marcos Novak. MIND A Design Machine, in Intelligent CAD Systems, Vol. 1[M].Ten Hagen, T. Tomiyama,Eds.. Berlin/New York: Springer Verlag,1987.

[3]Rosen Robert. Anticipatory Systems: Philosophical,Mathematical & Methodological Foundations[M]. Oxford/New York: Pergamon Press,1985.