何為空間的句法？作為界面的建筑及其設(shè)計(jì)

作者： John Peponis

John Peponis，佐治亞理大學(xué)建筑學(xué)院教授，希臘注冊(cè)建筑師，Kokkinou 與Kourkoulas建筑師事務(wù)所合作建筑師。

序言：空間句法和界面概念

人們通常認(rèn)為空間具有雙重性。我們將空間看作無(wú)限延伸或物體存在的場(chǎng)所。在日常生活中，我們認(rèn)定空間本質(zhì)上具有同質(zhì)性：在溫度等其他條件相同的情況下，物體在任何兩地的形狀和大小相同。然而除此之外，我們會(huì)從關(guān)系層面來(lái)理解空間?？臻g中的任何事物與其他事物可能存在遠(yuǎn)近、相鄰、前后、上下、外部、表面或旁邊等位置關(guān)系。通過(guò)設(shè)置邊界，可將空間進(jìn)行細(xì)分或重新連接。事物可位于其它事物或分隔結(jié)構(gòu)的內(nèi)部、之間、對(duì)面或外部。波特蘭?羅素（1897）在其最早期有關(guān)幾何學(xué)基礎(chǔ)的著作中指出，空白空間（可能存在的關(guān)系）和空間順序（物體存在時(shí)產(chǎn)生的實(shí)際關(guān)系）兩者間存在明顯不同。由于物體、邊界或表面的存在，空間的延伸無(wú)時(shí)不刻不受限制。因此，棲居空間具有句法，即從空間中的某個(gè)位置看到或進(jìn)入另一個(gè)位置的特定方式?！翱臻g句法”是一種有關(guān)棲居空間在物理和社會(huì)層面的認(rèn)知與理解的描述性理論，它關(guān)注空間環(huán)境在感知、 關(guān)聯(lián)和功能等方面的可供性。這一定義考慮了空間句法學(xué)說(shuō)的演化。最初，空間句法被定義為研究不同社會(huì)和文化的建筑空間基本組織原則的理論（Hillier, Leaman, Stansall and Bedford, 1976）。其后，為說(shuō)明其對(duì)社會(huì)表現(xiàn)的影響，空間句法又被用作比較分析建筑空間的定量分析方法（Hillier, Hanson, Peponis, et al., 1983）。在《空間的社會(huì)邏輯》（Hillier and Hanson, 1984）一書(shū)中，作者對(duì)上述兩種觀點(diǎn)進(jìn)行了綜合，并指出：定量分析抓住了生成建筑空間的組織原則可被觀測(cè)到的效果。

我對(duì)“空間句法”的定義側(cè)重于空間關(guān)系的表述，它界定了空間句法的方法特點(diǎn)：線網(wǎng)模型（Peponis, Wineman, Bafna et al., 1998; Turner, Penn, Hillier, 2005）、不同類型的凸空間分割（Peponis, Wineman, Rashid, et al.， 1997）、視域多邊形(Benedikt, 1979; Batty, 2001)，視域關(guān)系（Turner, Doxa, O’Sullivan et al.,2001）。這些基于空間關(guān)系的基本表述定義出一系列句法的度量方法，包括鄰近度中心性或曰整合度、中介度中心性或曰選擇度和觸及度。不論是根據(jù)方向變化（Hillier and Hanson, 1984; Peponis, Bafna, Zhang, 2008）、邊界穿越（Hillier and Hanson, 1984; Peponis, Wineman, Rashid, et al. 1997）、路徑的角度變化（Turner,2007）或公（米）制的路徑長(zhǎng)度計(jì)算距離，上述方法均適用?？臻g句法的表述和度量方法描述了空間的結(jié)構(gòu)，而空間結(jié)構(gòu)界定了（人在空間中）可能的運(yùn)動(dòng)、進(jìn)入、可視、共同在場(chǎng)或相遇（機(jī)會(huì)）。

“空間句法”一詞暗示了空間的可供性，正是因?yàn)檫@種可供性，空間就像一種語(yǔ)言，一種我們可以通過(guò)設(shè)計(jì)空間關(guān)系來(lái)滿足社會(huì)或個(gè)人生活方式的媒介。將組構(gòu)定義為考慮其他關(guān)系的關(guān)系（Hillier, 1996; Peponis, Bafna, Dahabreh et al., 2015），強(qiáng)調(diào)了個(gè)體空間關(guān)系之間以模式化的方式相互作用、彼此支撐。例如：兩個(gè)房間之間的關(guān)系可以是不對(duì)稱的（如果需要先穿過(guò)一個(gè)房間，方能到達(dá)另一個(gè)房間），也可以是對(duì)稱的（如果從房間外部可獨(dú)立地到達(dá)任何一個(gè)房間）（Hillier, 1996）?？臻g句法的度量方法抓住了模式的結(jié)構(gòu)特征，當(dāng)中的每一種關(guān)系都可能受到其他關(guān)系的影響。根本的要點(diǎn)在于，建筑空間的人文功能建立在這一強(qiáng)烈和可被理解的關(guān)聯(lián)性基礎(chǔ)上。

我們現(xiàn)在對(duì)作為空間語(yǔ)言理論的空間句法和它的用途做個(gè)區(qū)分；二者分別關(guān)于社會(huì)空間系統(tǒng)的理解和設(shè)計(jì)。有哪些概念能夠幫助我們理解空間句法語(yǔ)言陳述的觀點(diǎn)？我用“界面”的概念來(lái)回應(yīng)這個(gè)問(wèn)題。

希列爾和漢森（1984）曾在兩個(gè)語(yǔ)境中使用“界面”一詞。在分析聚落時(shí)，他們用界面描述系統(tǒng)中開(kāi)放和封閉部分之間的關(guān)系?！敖缑鎴D”表示每幢獨(dú)立建筑與該建筑連接的公共可達(dá)的凸空間單元之間的聯(lián)系，以及凸空間之間相互連接形成的供出行和共同在場(chǎng)的連續(xù)空間系統(tǒng)。界面圖暗示了居民與陌生人之間的關(guān)系：即有權(quán)進(jìn)入或擁有至少一幢建筑的人員與僅能進(jìn)入公共開(kāi)放空間的人員之間的關(guān)系。

希列爾和漢森亦提出，建筑的內(nèi)部結(jié)構(gòu)形成了不同類型的使用者之間的界面，最明顯的是“住戶”和“訪客”。住戶指有權(quán)擁有和使用建筑各項(xiàng)功能的居住者，而訪客是指獲住戶許可方能訪問(wèn)建筑的用戶。其次，住戶與訪客之間的界面還被描述為兩類用戶所分別享有的建筑各部分的句法屬性，和兩類用戶相遇空間的屬性。在該情形中，將建筑劃分為分布式和非分布式子系統(tǒng)就顯得尤為重要。分布式系統(tǒng)包含了可利用流線中的環(huán)路經(jīng)由其他路徑到達(dá)的空間。非分布式子系統(tǒng)包含了從分布式系統(tǒng)分支出去的只能按照流線序列順次到達(dá)的空間。研究表明，在很多類型建筑中，住戶與訪客之間的界面反映為分布式與非分布式子系統(tǒng)之間的區(qū)分與聯(lián)系。

分析表明，社會(huì)范疇的空間定義建立在不同句法條件的內(nèi)在特性基礎(chǔ)上。創(chuàng)建不同條件并建立相互關(guān)聯(lián)，是建筑學(xué)的基本目的之一。因此，我們可從如下角度思考界面：界面形成于某種模式下不同的句法條件或可供性的系統(tǒng)性創(chuàng)造和關(guān)聯(lián)。雖然不同的句法條件和社會(huì)類別之間的聯(lián)系可能是空間的社會(huì)邏輯的根本，但我認(rèn)為這并非界面的本質(zhì)所在。

最近的學(xué)界討論結(jié)果與一般意義上的界面定義相契合。希列爾（2001）曾提出，大多數(shù)城市可以區(qū)分出兩個(gè)子系統(tǒng)：即主要街道網(wǎng)絡(luò)（將城市連為一體的整體網(wǎng)絡(luò)）和分布在主要街區(qū)中的普通街道形成的局部網(wǎng)絡(luò)。一條主要街道通過(guò)廣角相交方式與其他主要街道相連，提供了更長(zhǎng)的線性運(yùn)動(dòng)或視野；而局部街道提供了較短線性運(yùn)動(dòng)和視野，它們之間通常以直角方式相交，雖然提供了多種路徑選擇，但這些路徑在運(yùn)動(dòng)和視線的方向上幾近相同。從本質(zhì)上看，希列爾將街道網(wǎng)格看作局部和整體兩種組織規(guī)模之間的界面。我本人也提出過(guò)類似的觀點(diǎn)，建筑物能被讀懂是基于主要空間和一般空間的系統(tǒng)性設(shè)置所形成的界面；主要空間指具有擴(kuò)展視野和多方向連接的全景空間，一般空間指感知能力被限定在室內(nèi)特定范圍內(nèi)的空間（Peponis, 2012）。走廊、庭院、中庭和大廳是我們可能用來(lái)設(shè)置界面的措施。

以下章節(jié)從研究、設(shè)計(jì)實(shí)踐和工作坊教學(xué)的角度展開(kāi)有關(guān)界面例子的討論。

互動(dòng)與合作的界面和機(jī)構(gòu)生產(chǎn)力： Thoughtform Inc.（匹茲堡）

ThoughtForm Inc. 是一家位于匹茲堡的傳播設(shè)計(jì)公司，其辦公場(chǎng)所由Archideas事務(wù)所的Michael Fazio于2002年設(shè)計(jì)建造。該項(xiàng)目旨在加強(qiáng)內(nèi)部合作、促進(jìn)員工互動(dòng)、提高創(chuàng)造力和生產(chǎn)力。項(xiàng)目利用了Steelcase公司愿意提供定制辦公設(shè)備的機(jī)會(huì)。2005年，Steelcase與佐治亞理工學(xué)院的空間句法研究團(tuán)隊(duì)協(xié)作，共同對(duì)設(shè)計(jì)的有效性進(jìn)行研究，并發(fā)表了相關(guān)的研究報(bào)告（Peponis、Bafna、Bajaj等，2007年）。與該公司老的辦公場(chǎng)所相比，新辦公場(chǎng)所更有助于員工在項(xiàng)目構(gòu)思和概念發(fā)展階段提高工作的效率和效果。

辦公樓平面為矩形，長(zhǎng)寬為20 x 74米，占地面積約1,500平方米。主要通道為貫穿整個(gè)辦公區(qū)樓面長(zhǎng)度的雙邊開(kāi)放式走廊。沿辦公樓的長(zhǎng)邊邊緣另設(shè)有通道連接各工作區(qū)。從入口大廳走進(jìn)，迎面而來(lái)的是多功能開(kāi)放式公共休息室，可用于舉辦正式和非正式會(huì)議或咖啡休息。緊挨著休息室的是四間項(xiàng)目室，是項(xiàng)目團(tuán)隊(duì)的大本營(yíng)。由于采用折疊墻隔板設(shè)計(jì)，項(xiàng)目室可向休息室關(guān)閉或開(kāi)放。主通道的兩端分別設(shè)有小型的封閉式會(huì)議室。圖書(shū)室內(nèi)設(shè)有額外座椅，供員工休息使用。員工工作區(qū)僅占樓面面積的50%，大幅低于老辦場(chǎng)所的70%（圖1）。

我們進(jìn)行研究時(shí)，ThoughtForm Inc.擁有約50名員工。問(wèn)卷調(diào)查表明，員工十分滿意工作場(chǎng)所的布局，在需要時(shí)有場(chǎng)所可召開(kāi)會(huì)議及展開(kāi)團(tuán)隊(duì)工作。社會(huì)學(xué)計(jì)量分析表明，在每名員工與公司余下65%的員工保持互動(dòng)的情況下，員工之間的整體接觸密度不會(huì)發(fā)生大的變化。然而，員工每天或每周在工作間隙發(fā)生互動(dòng)的頻率和比例增加。頻繁的接觸在整個(gè)辦公室內(nèi)形成一個(gè)關(guān)系網(wǎng)絡(luò)，使得大多數(shù)員工不僅與工位附近的同事產(chǎn)生互動(dòng)，同時(shí)也與其他同事產(chǎn)生互動(dòng)（圖1.4）。此外，新辦公樓內(nèi)工作區(qū)的中心性和與其所對(duì)應(yīng)的互動(dòng)網(wǎng)絡(luò)中員工的中心性之間的相關(guān)度大大增加。因此，就空間布局而言，員工的互動(dòng)模式變得更加明顯易懂。

新工作場(chǎng)所的設(shè)計(jì)增強(qiáng)了員工的共同意識(shí)，這是項(xiàng)目構(gòu)思和概念發(fā)展效率提升的原因所在。例如，面對(duì)面交流的比例提高，意味著當(dāng)中一人或兩人需穿過(guò)中間休息室和鄰近項(xiàng)目室。因此，多數(shù)互動(dòng)，不論其目的和初衷，都意外地增加了尋求和提供建議、專業(yè)知識(shí)或其他貢獻(xiàn)的可能性。新的布局以非正式的方式加深了員工對(duì)組織運(yùn)作的了解，從而提升了創(chuàng)造性和生產(chǎn)力。

我希望探討界面設(shè)置對(duì)非正式學(xué)習(xí)的影響，這對(duì)知識(shí)型工作而言至關(guān)重要。第一個(gè)、也是最顯而易見(jiàn)的界面是實(shí)際或潛在的走動(dòng)或互動(dòng)。希列爾和佩恩（1991年）已經(jīng)注意到該界面的重要性。Thoughtform的所有會(huì)議室均與主通道相連或被其貫穿。此外，每個(gè)工作團(tuán)隊(duì)所在的工作區(qū)均配有移動(dòng)式衛(wèi)星資源中轉(zhuǎn)設(shè)備，可邀請(qǐng)其他同事召開(kāi)簡(jiǎn)短會(huì)議。這件定制的家具配有書(shū)架，可存放從主圖書(shū)室借閱、或由團(tuán)隊(duì)成員整合或編制的項(xiàng)目關(guān)鍵文件。這件家具的頂部高度不超過(guò)手肘高度，員工可將其用作資料閱讀桌、咖啡或筆記本擺放桌。所有中轉(zhuǎn)設(shè)備均擺放在主通道視線可及之處，其中半數(shù)以上緊挨著主通道?？梢哉f(shuō)，空間布局對(duì)互動(dòng)的支持無(wú)處不在。（圖1.5）。

第二個(gè)界面普遍存在于員工走動(dòng)和可視化信息之間。點(diǎn)子、草圖、圖片或便簽、海報(bào)和便利貼的痕跡雖然隨處可見(jiàn)，但并非辦公場(chǎng)所內(nèi)的所有位置都能輕易看見(jiàn)。辦公室內(nèi)配備軟質(zhì)黑板和標(biāo)記板，供會(huì)議室、團(tuán)隊(duì)工作區(qū)和個(gè)人工作位使用。新辦公場(chǎng)所還留有題寫會(huì)議主題詞的空間，供會(huì)議召開(kāi)時(shí)使用。

第二個(gè)普遍存在的界面位于運(yùn)動(dòng)和視覺(jué)信息顯示之間。點(diǎn)子、草圖、圖片或便簽、海報(bào)和便利貼的痕跡雖然隨處可見(jiàn)，但并非辦公場(chǎng)所內(nèi)的所有位置都能看清楚內(nèi)容。辦公室內(nèi)配備軟質(zhì)黑板和標(biāo)記板，供會(huì)議室、團(tuán)隊(duì)工作區(qū)和個(gè)人工位使用。新辦公場(chǎng)所還留有題寫會(huì)議主題詞的空間，供會(huì)議召開(kāi)時(shí)使用。

上述兩個(gè)界面的存在，使每名員工與其他員工得以維持一種動(dòng)態(tài)關(guān)系，而不論其是否身處同一項(xiàng)目組，也不論其是否有著類似的專業(yè)知識(shí)背景和經(jīng)驗(yàn)。辦公室的人際網(wǎng)絡(luò)無(wú)形中扮演著共有知識(shí)的集體管理者和生產(chǎn)者的角色。然而，空間內(nèi)的相遇和視覺(jué)交流成為人際網(wǎng)絡(luò)有效運(yùn)作的媒介。

在此案例中，界面設(shè)計(jì)并非單純?cè)O(shè)計(jì)一條筆直穿過(guò)辦公場(chǎng)所的中心走道，該辦公場(chǎng)所中走道與樓層邊緣的距離最多不超過(guò)10米。界面設(shè)計(jì)涉及其他方面，包括集體辦公區(qū)內(nèi)工位的設(shè)置，會(huì)議區(qū)的預(yù)留，以及定制化家具和設(shè)備的陳列。若用空間句法語(yǔ)言，則可將其概述為引導(dǎo)目標(biāo)對(duì)象沿清晰的整合中心運(yùn)動(dòng)，而該中心總體上具備知識(shí)型工作所需的互動(dòng)、學(xué)習(xí)和協(xié)作功能。將布局、家具和設(shè)備設(shè)計(jì)作為機(jī)構(gòu)及其文化的整體部分，是句法的整合度能發(fā)揮積極作用的原因所在。

機(jī)構(gòu)的個(gè)性和功能可供性：雅典比雷艾夫斯街貝納基博物館

2001年，Maria Kokkinou和Andreas Kourkoulas兩位建筑師被選中負(fù)責(zé)設(shè)計(jì)位于雅典比雷艾夫斯大街138號(hào)的貝納基博物館新館。兩位建筑師都曾在倫敦向比爾?希列爾求學(xué)。在該項(xiàng)目中我擔(dān)任兩位建筑師的設(shè)計(jì)顧問(wèn)。比雷艾夫斯大街的兩端分別連接著雅典老城和比雷艾夫斯港，街道沿著雅典長(zhǎng)墻的北墻延伸。長(zhǎng)墻是公元前5世紀(jì)保護(hù)雅典城和比雷埃夫斯港之間通道的重要工事。自19世紀(jì)起，比雷艾夫斯大街沿線主要為工業(yè)區(qū)和低收入住宅區(qū)。最近數(shù)十年間，逐步的城市更新改造為比雷艾夫斯大街重新賦予了文化生活功能。2004年建成的貝納基博物館新館是更新改造過(guò)程中的里程碑項(xiàng)目。

自1929年建成至今，貝納基博物館的永久館藏一直以希臘文化藏品，橫跨古希臘、拜占庭帝國(guó)和現(xiàn)代希臘等多個(gè)時(shí)期。博物館的老館位于雅典城中心的憲法廣場(chǎng)附近。新館位于比雷艾夫斯大街，計(jì)劃作為博物館的衛(wèi)星式分館使用，舉辦臨時(shí)性委托、策劃或巡回作品展。新館原為一處庭院式布局的汽車修理廠房，經(jīng)改造和擴(kuò)建后建成。入口設(shè)置在側(cè)面的小街上，充分利用了建筑后退紅線較大的優(yōu)勢(shì)。

新館在設(shè)計(jì)時(shí)保留了庭院，以其為博物館的核心，并且保留了從側(cè)面小街進(jìn)出的入口方式以控制比雷艾夫斯大街噪音和車流的影響。新館處于兩個(gè)風(fēng)格截然不同的邊界包圍之中：外墻面為紅色石墻，僅策略性設(shè)有幾個(gè)條形窗。庭院內(nèi)立面中的三面為軟質(zhì)表面，實(shí)為向上傾斜的巨型“屏風(fēng)”。屏風(fēng)由特別設(shè)計(jì)的水平百葉組成，百葉的材料為大綠柄桑木，角度可調(diào)，用來(lái)控制陽(yáng)光入射。新館上面的三層在 “屏風(fēng)”與建筑內(nèi)邊之間設(shè)有通廊。外挑的通廊和“屏風(fēng)”在底層沿庭院周邊形成光影。庭院內(nèi)立面的第四面與入口相對(duì)，由夾在內(nèi)墻和金屬框架玻璃幕墻之間的之字形坡道形成。坡道用于展現(xiàn)參觀者的行走軌跡，是博物館的一大特色。雖然坡道使主要立面的形象鮮活，但“屏風(fēng)”暗示了其他三個(gè)立面后潛藏的風(fēng)景。底層庭院由鵝卵石鋪就，暢通無(wú)阻。站在比雷艾夫斯大街，可以從外墻的條形窗、博物館前部的商店和咖啡廳窺見(jiàn)館內(nèi)的景象。大尺度的空間開(kāi)敞使得庭院得以向后延伸至博物館展覽區(qū)。因此，盡管以坡道為主精心設(shè)計(jì)的正立面形象主導(dǎo)了入口方向的軸線景觀，但橫軸上空間的層疊展示了與之不同的富有變化的景象（圖2）。

Angelos Delivorias和Irini Geroulanou先后擔(dān)任貝納基博物館館長(zhǎng)。Angelos Delivorias于2014年離任，此后，由Irini Geroulanou負(fù)責(zé)新館的修建和運(yùn)營(yíng)工作。在兩位富有創(chuàng)見(jiàn)的館長(zhǎng)領(lǐng)導(dǎo)下，博物館發(fā)展成為雅典的文化地標(biāo)。博物館被視為不同藝術(shù)和文化社群的集會(huì)地和放大器，以及積極參與藝術(shù)公眾化的機(jī)構(gòu)（Delivorias,2004）。庭院對(duì)博物館實(shí)現(xiàn)這一創(chuàng)想發(fā)揮了極其重要的作用。庭院中不僅舉辦裝置作品展、雕塑展，而且還舉辦各類音樂(lè)和戲劇表演、社交和集會(huì)活動(dòng)。庭院在舉辦文化活動(dòng)時(shí)所受的歡迎程度及其功能的多樣性超出了設(shè)計(jì)團(tuán)隊(duì)和館方的預(yù)期。在設(shè)計(jì)過(guò)程中，客戶及其顧問(wèn)曾一度提議將庭院改為綠化。在主體建筑完成后，設(shè)計(jì)團(tuán)隊(duì)在實(shí)施館長(zhǎng)的構(gòu)想過(guò)程中發(fā)現(xiàn)，可以將庭院設(shè)計(jì)為獨(dú)具特色的多功能舞臺(tái)。簡(jiǎn)言之，庭院設(shè)計(jì)提供了豐富的功能，有助于發(fā)展、豐富和厘清館方最初的用意（圖3）。

下列設(shè)計(jì)舉措強(qiáng)化了庭院作為舞臺(tái)的特點(diǎn)。首先，盡管博物館底層的視線聯(lián)系較廣，將博物館入口置于側(cè)面小街則從認(rèn)知和情感上切斷了庭院與比雷艾夫斯大街的聯(lián)系。其次是庭院立面材料的質(zhì)感與區(qū)分。第三，庭院四周的圍護(hù)結(jié)構(gòu)暗示著場(chǎng)館內(nèi)其他人的出現(xiàn)。身處博物館之中，參觀者可體會(huì)到相互之間的看與被看、打招呼與被打招呼。第四，正對(duì)入口軸線看到的參觀者在坡道上的行進(jìn)如同庭院與博物館之間形成的舞臺(tái)上的前景運(yùn)動(dòng)，與比雷艾夫斯大街上的人車流呈直角關(guān)系。第五，傾斜的“屏風(fēng)”拓寬了縱向視野，將整個(gè)庭院的風(fēng)光盡收眼底。在貝納基博物館，視覺(jué)形態(tài)起到強(qiáng)化空間的組織和活力的意象。庭院的空間感覺(jué)吸引著參觀者探索潛在的可供性和空間關(guān)系的在場(chǎng)體驗(yàn)，并強(qiáng)調(diào)了對(duì)于棲居的渴望。

本質(zhì)上，設(shè)計(jì)確定了一種組構(gòu)方式，通過(guò)可達(dá)與可見(jiàn)、運(yùn)動(dòng)的連續(xù)性與視線能否穿越滲透性或透明性邊界之間富有表現(xiàn)力的界面，調(diào)和了外部街道與內(nèi)部庭院、內(nèi)部庭院與建筑之間的雙重功能性關(guān)系。知覺(jué)空間的結(jié)構(gòu)與入口、流線序列的精心層疊，使得各種可能的功能在庭院內(nèi)被不斷發(fā)現(xiàn)和實(shí)施。通過(guò)這一方式，建筑為機(jī)構(gòu)賦予了部分的個(gè)性特征。

建筑類型的轉(zhuǎn)變：亞特蘭大佐治亞理工大學(xué)，零售業(yè)的未來(lái)設(shè)計(jì)工作坊

零售業(yè)正面臨著重組以便應(yīng)對(duì)科技、經(jīng)濟(jì)和文化的發(fā)展變化。高效可靠的線上賣家和物流服務(wù)的出現(xiàn)，令人質(zhì)疑實(shí)體商店是否有必要存在。專家主張推行集實(shí)體店、網(wǎng)上購(gòu)物及移動(dòng)應(yīng)用程序購(gòu)物于一體的全渠道零售，使顧客能隨時(shí)隨地、隨心所欲地購(gòu)買商品。我們正面臨Mitchell在1999年描述的空間和文化體驗(yàn)上的深刻變化。在2017年春獲得NCR的研究資助后，我主持了一項(xiàng)本科生參加的設(shè)計(jì)工作坊，探索除便利性外，逛商店對(duì)消費(fèi)者來(lái)說(shuō)還有哪些重要的可取之處。服務(wù)模式不給力的實(shí)體零售店受線上賣家的沖擊最為明顯。因此，我們想知道什么才是實(shí)體零售店需重新定義的固有功能。

商店有助于消費(fèi)者發(fā)現(xiàn)新事物。當(dāng)我們?cè)跁?shū)店無(wú)意間發(fā)現(xiàn)一本感興趣的新書(shū)時(shí)，或在超市找到一款新口味奶酪時(shí)，我們對(duì)該商店供貨范圍的認(rèn)知得到擴(kuò)充。因此，商店不僅僅是直接搜索特定商品的場(chǎng)所，還是開(kāi)放式搜索的場(chǎng)所。開(kāi)放式搜索有助于新產(chǎn)品和新想法的傳播。

商店有教育功能。例如，在樂(lè)器店，顧客可以直接感受外形相似的樂(lè)器之間的細(xì)微區(qū)別。傳統(tǒng)屠夫會(huì)根據(jù)顧客的烹飪需求向其推薦肉品。藥劑師會(huì)對(duì)輕傷治療提供意見(jiàn)。商店的存在，為一般性需求與特定實(shí)用技術(shù)的對(duì)接提供了可能。同時(shí)，商店有助于我們明確購(gòu)買商品的想法、習(xí)慣和行為。隨著我們?cè)絹?lái)越關(guān)注與健康、環(huán)境和社會(huì)生產(chǎn)條件有關(guān)的各種產(chǎn)品之間的關(guān)系，商店的教育功能將越來(lái)越重要。

信譽(yù)是商店興旺的基礎(chǔ)。這對(duì)提供定制化服務(wù)的商店，如裁縫店，或提供個(gè)性化建議的商店，如專業(yè)書(shū)店而言，尤其重要。店主或店員可扮演專家和顧問(wèn)的角色。

商店的某些附屬功能同樣重要。商店提供了偶遇的場(chǎng)所，猶如一間社會(huì)劇場(chǎng)，人們之間互相看與被看。在發(fā)展視覺(jué)傳播語(yǔ)言方面，商店和博物館的作用同等重要；在物質(zhì)文化的傳播方面，商店的作用更大。即便未能達(dá)成交易，顧客和商店品牌之間的心理聯(lián)系也會(huì)得到加強(qiáng)。

工作坊以團(tuán)隊(duì)形式開(kāi)展工作。兩個(gè)小組針對(duì)商品陳列、存儲(chǔ)和客服之間的界面重組，以相似的方式發(fā)展了不同的組合，旨在解決大多數(shù)百貨商店普遍存在的功能混亂問(wèn)題。

Collin Garnett、Tim Peterson和 Christina Delurgio組成的小組沿著基地的兩個(gè)主要邊界設(shè)計(jì)了細(xì)長(zhǎng)的L形展廊式空間，用于商品陳列。L形陳列空間限定的范圍內(nèi)，設(shè)計(jì)了相互連接的三維矩陣式吊艙，作為導(dǎo)購(gòu)員和顧客的交流區(qū)。吊艙設(shè)計(jì)為創(chuàng)意空間，顧客與店員在這里交流，根據(jù)交流的結(jié)果對(duì)某家公司的產(chǎn)品或某種產(chǎn)品進(jìn)行定制化、修改、組合或開(kāi)發(fā)。吊艙式空間位于底層公共開(kāi)放空間之上，既可充當(dāng)舉辦各類表演和活動(dòng)的通道和獨(dú)立的公共場(chǎng)所，也可以借助周圍的咖啡廳、酒吧和餐廳體驗(yàn)日常城市生活。貨物儲(chǔ)存于地下室，由電梯送至吊艙式空間（圖4）。

Danny Griffin、Maria Pastorelli和Skylar Royal小組設(shè)計(jì)了畫廊式商品展示區(qū)，沿商場(chǎng)正前方非對(duì)稱布局的公共空間逐層向上分布。小組成員對(duì)展廊空間進(jìn)行了仔細(xì)研究，可以讓顧客在其中使用移動(dòng)設(shè)備訂購(gòu)商品。在底層，兩側(cè)的小型商鋪豐富了公共空間的內(nèi)容。在上層空間，商品展廊后方設(shè)有貨物存放區(qū)，用于存放和配送貨物。貨物存放區(qū)和商品展示區(qū)之間設(shè)有厚墻容納商品自動(dòng)傳送。在頂層，顧客手拎裝有選購(gòu)產(chǎn)品的購(gòu)物籃，對(duì)產(chǎn)品進(jìn)行試穿，并與專業(yè)的員工討論組合搭配、定制或改進(jìn)。為向顧客提供方便，頂層設(shè)有試衣間/咨詢室、配有酒吧的大廳，顧客可在此一覽城市美景。有通道可通往位于貨物存放區(qū)上方的創(chuàng)意工坊車間和實(shí)驗(yàn)室（圖5）。

通過(guò)對(duì)百貨商店的界面進(jìn)行重組，兩支設(shè)計(jì)團(tuán)隊(duì)從三個(gè)維度建立了視覺(jué)聯(lián)系，加強(qiáng)了看與被看見(jiàn)之間的關(guān)系，強(qiáng)化了商場(chǎng)的社交場(chǎng)合感。構(gòu)成商店前景的商品展示和客商交流充分利用了商品存儲(chǔ)和檢索的新技術(shù)、同時(shí)也支持客戶將選中商品放入虛擬購(gòu)物籃內(nèi)。

兩個(gè)設(shè)計(jì)方案的表現(xiàn)成果包括建筑表現(xiàn)圖、展示主要界面定義與組成的軸測(cè)分解圖、和描述購(gòu)物體驗(yàn)序列的故事板。這些圖闡述了界定建筑類型的界面重構(gòu)所涉及的幾何、動(dòng)線、視線和功能概念。上述項(xiàng)目?jī)H代表了商業(yè)建筑類型演化的一種可能性路線，而決非定論。然而，相關(guān)的設(shè)計(jì)成果表明，創(chuàng)造某種可能性的抽象空間設(shè)計(jì)與形成啟發(fā)式的界面概念之間存在密切關(guān)系。

界面的定義

在上述三例中，界面與空間關(guān)系存在聯(lián)系，界面或存在于子系統(tǒng)之間而非單個(gè)要素之間，或在某種模式之中不斷重復(fù)：在Thoughtform表現(xiàn)為員工互動(dòng)與交通流線的關(guān)系；在貝納基博物館表現(xiàn)為庭院內(nèi)部及其周邊空間的可見(jiàn)與可達(dá)關(guān)系；在學(xué)生設(shè)計(jì)工作坊項(xiàng)目表現(xiàn)為商品陳列、存放和商客互動(dòng)區(qū)之間的關(guān)系。因此，界面涉及多類元素。界面的言外之意暗示了某種概念化，即根據(jù)某種共性將某些元素歸為一類。上述三個(gè)案例的共同之處是提供了某些可供性，這些可供性在功能上的重要意義與某種特定的句法條件相關(guān)。

我本人對(duì)界面提出如下定義：界面指影響整體空間結(jié)構(gòu)的各類要素之間的復(fù)合關(guān)系。功能的可供性及其相關(guān)的句法條件界定了要素的類別。組構(gòu)的概念指一組關(guān)系基于所有其他關(guān)系的影響，界面的概念則要求關(guān)系分類的有序組織。系統(tǒng)的組構(gòu)屬性是形成界面的素材。與此同時(shí)，界面表述了更高層級(jí)的布局原則，不論這些原則來(lái)自常識(shí)，或源于對(duì)設(shè)計(jì)意圖的反思。

界面的種類多種多樣，如社交互動(dòng)界面、功能分區(qū)界面等。貝納基博物館案例的意義重大，它表明知覺(jué)模式與運(yùn)動(dòng)可供性之間的界面在特定環(huán)境中具有潛在功能發(fā)生器的重要意義。

任何特定的設(shè)計(jì)或某種類型的設(shè)計(jì)，界面都非常重要，它是建筑類型、功能意圖、設(shè)計(jì)構(gòu)思和概念形成的產(chǎn)物。然而，空間句法理論表明，只有當(dāng)什么是重要界面的開(kāi)放，構(gòu)建界面的幾何形式開(kāi)放，才可能有本文定義的界面理論。正如文中引用的希列爾（2002年）和佩恩（2012年）的著作所述，建筑空間中最根本的界面在于局部和整體兩種尺度關(guān)系的組織。這是空間能夠被從人和文化的角度理解的根本所在。

界面基本理論的發(fā)展，特定類型或功能模式的界面模型的發(fā)展，豐富了關(guān)于空間的社會(huì)與文化邏輯的解釋性理論及其規(guī)范化命題的發(fā)展。界面的概念也架起了空間句法理論與設(shè)計(jì)意圖或設(shè)計(jì)構(gòu)想之間橋梁。

鳴謝

在本文的寫作過(guò)程中與比爾?希列爾的交流讓我受益匪淺。他發(fā)表的觀點(diǎn)毫無(wú)疑問(wèn)是本文討論的核心所在，他的意見(jiàn)讓本文的觀點(diǎn)得以成形。

Introduction: Space Syntax and the idea of interface

Our common understanding of space is dual. We understand space as pure extension or volume in which things exist. In everyday life, we assume it to be homogeneous in fundamental ways: material objects will have the same shape and dimensions in any two locations, other things, such as temperature,being equal. In addition, however, we understand space in terms of relationships. Things in space are distant, proximate or adjacent, in front, behind, above,below, beneath, beyond, on, under, or beside other things. Space can be subdivided and reconnected by an arrangement of boundaries. Things can be inside,between, across, or outside relative to other things or enclosures. In his earliest work on the foundations of geometry, Bertrand Russell (1897) drew a distinction between empty space as the possibility of relations and the order of space, the actual relationships that arise when material things are present. With the presence of objects, boundaries or surfaces, extension is never unrestricted. Hence, inhabited built space has a syntax,a particular way in which one location can come into view or be accessed starting from another.

‘Space syntax’ is a descriptive theory of the perceptual, relational and functional affordances of inhabited space that are relevant to its cognitive and social intelligibility. This definition takes into account the evolution of the field. Originally, space syntax was proposed as the theory of the fundamental organizing principles that govern built space across societies and cultures (Hillier, Leaman, Stansall and Bedford, 1976).Subsequently space syntax was introduced as a set of quantitative techniques applied to the comparative analysis of built space (Hillier, Hanson, Peponis, et al.,1983) in order to account for its social performance.In the Social Logic of Space (Hillier and Hanson, 1984)those two ideas were brought into a unified argument:quantitative analysis captures the observable effects of the principles of organization that generate built space.

The definition I offer puts the emphasis on the representations of spatial relationships that define space syntax methods: line-maps (Peponis, Wineman,Bafna et al., 1998; Turner, Penn, Hillier, 2005), various kinds of convex partitions (Peponis, Wineman, Rashid,et al. 1997), visibility polygons (Benedikt, 1979; Batty,2001), visibility relationships (Turner, Doxa, O’Sullivan et al., 2001). Syntactic measures, including closeness centrality or integration, betweenness centrality or choice and reach, are defined on the basis of such underlying representations of spatial relationships. This is true whether distances are calculated according to direction changes (Hillier and Hanson, 1984; Peponis, Bafna,Zhang, 2008), boundaries traversed, (Hillier and Hanson,1984; Peponis, Wineman, Rashid, et al. 1997), path angular rotation (Turner, 2007) or metric path length. The representations and measures of space syntax describe the structures of space that define possible movement,access, view, co-presence or encounter.

The term ‘space syntax’ alludes to the fact that by virtue of such affordances, space acts like a language,a medium in which we can design relationships to accommodate desirable ways of common or individual life. The definition of configuration as relationships that take into account other relationships (Hillier, 1996;Peponis, Bafna, Dahabreh et al., 2015) underscores the fact that individual spatial relationships interact with and qualify each other in the context of patterns. For example, the same connection between two rooms can be asymmetric, if access to one room depends upon first traversing the other; or it can be symmetric,if each room can be independently accessed from the outside (Hillier, 1996). The measures of space syntax capture the structural properties of patterns wherein each relation is potentially affected by all others. The fundamental thesis is that the human functions of built space are built on this foundation of intense and intelligible relationality.

We can now distinguish between space syntax, as a theory of the language of space, and what space syntax is used for; namely to understand and design sociospatial systems. What are the concepts that may help us to understand the statements made in the language of space syntax? In response to this question I focus on the idea of interface.

Hillier and Hanson (1984) use the word ‘interface’in two contexts. In the analysis of settlements, they describe the relationship between the open and closed parts of the system. ‘Interface maps’ represent the connection between each discrete building and the publicly accessible convex space that the building is linked to, as well as the interconnections of convex spaces into a continuous system of movement and copresence. By implication, interface maps capture the relationship between inhabitants and strangers: the people that have rights of access to, or ownership of,at least one of the buildings and the people that only have rights of access to the common open space.

Hillier and Hanson also propose that the internal structure of buildings constructs interfaces between different categories of users, most notably between‘inhabitants’ and ‘visitors’. Inhabitants are the occupants who have authority and power over the functions of the building and visitors are the users whose access to the building is subject to regulation by the inhabitants. The interface between inhabitants and visitors is then described in terms of syntactic properties of the parts of the building that are severally available to each category of users and the properties of the spaces in which the two categories meet. In this context,the partition of buildings into distributed and nondistributed subsystems acquires particular significance.The distributed system comprises the spaces that can be reached in alternative ways taking advantage of the presence of circulation loops. The non-distributed subsystem comprises the spaces that can only be reached by circulation sequences branching off the distributed system. It is suggested that in many kinds of buildings the interface between inhabitants and visitors is mapped onto the distinction and relationship between the distributed and non-distributed subsystems.

The analysis shows that the spatial definition of social categories is built upon the intrinsic properties of distinct syntactic conditions. The creation and interrelation of distinct conditions is a fundamental purpose of architecture. We may, therefore, think of the idea of interface as follows: An interface exists when distinct syntactic conditions or affordances are systematically created and related within a pattern.The association between distinct syntactic conditions and social categories may be fundamental to the social logic of space but it is not, I suggest, intrinsic to the idea of interface itself.

More recent arguments fit within the generalized idea of interface. Hillier (2001) has proposed that in most cities a clear distinction can be drawn between two subsystems: the primary streets which form a global network that connects the city as a whole; and the local networks of ordinary streets in the areas between the primary streets. Primary streets afford longer straight lines of movement and sight extended by means of wide-angle intersections to other such lines. Local streets provide shorter straight lines of movement and sight and usually intersect at right angles offering choices between alternative but equivalent directions of movement and sight. In essence, Hillier looks at street grids as interfaces between two scales of organization,local and global. Similarly, I have suggested that buildings are intelligible based on the systematic creation of an interface between prominent spaces that provide an overview of extended areas and of multiple connections and normal spaces which confine perception to a limited part of the interior (Peponis,2012). Corridors, courtyards, atria and halls are devices that we use to make such interface possible.

In the following sections I discuss examples of interfaces from the perspectives of applied research,professional design and studio teaching.

Organizational productivity and the interfaces of interaction and collaboration: Thoughtform Inc.,Pittsburgh.

The workplace of ThoughtForm Inc., a communications design company located in Pittsburgh, was created by Michael Fazio, of Archideas, in 2002. The aim was to enhance internal collaboration and interaction and accelerate creativity and productivity. The project took advantage of the willingness of Steelcase to supplement catalogue office furniture with custom designed pieces. The effectiveness of the design was studied by Steelcase in 2005 in collaboration with the space-syntax research team at Georgia Tech and a report was published (Peponis, Bafna, Bajaj et al.,2007). The new premises allowed the ideation and concept development phases of projects to proceed more effectively and efficiently as compared to the old premises.

The layout occupies a rectangular office floor of about 20 x 74m and an area of about 1500m2. The main circulation is along a double loaded open corridor traversing the entire length of the floor. Alternative connections between work areas are made along the edge. Facing the entrance hall, a multi-purpose open lounge accommodates formal and informal meetings or coffee breaks. Adjacent to the lounge, four project rooms function as home bases for project teams.With folding partitions, these rooms can be closed or opened towards the lounge. Smaller enclosed meeting rooms are located at the two ends of the main circulation. Additional seating for relaxation is available in the library. Individual workstations cover only 50% of the floor, a drastic reduction compared to 70% in the old layout (FIGURE 1).

Questionnaires indicated that employees, who were about 50 at the time of the study, were very satisfied with the range of work settings that were available to them, and with the ability to have meetings and work in groups as needed. Sociometric analysis showed that the overall density of contacts between employees did not change much, with each employee interacting with about 65% of all other employees. However, the proportion of interaction that occurs more frequently at the daily or weekly time intervals increased. Frequent contacts form a network that covers the entire floor,with most individuals interacting not only with people in the vicinity of their workstation but also with people across the floor (FIGURE 1. 4). Furthermore, correlations between the centrality of workstations in the layout and the centrality of the corresponding individuals within the network of interaction increased dramatically in the new premises. Thus, the pattern of interaction became more intelligible in terms of the organization of the layout.

The explanation for the increase of effectiveness during project ideation and concept development resided in the way in which the design promoted co-awareness.For example, a great proportion of reported face-to-face interactions imply that one or both parties had to walk past the central lounge and adjacent project rooms.Thus, most interactions, regardless of their purpose and origination, increased the likelihood that advice,expertise, or other contributions could be sought and offered serendipitously. The layout enhanced creativity and productivity by supporting an almost ubiquitous process of informal learning about the functioning of the organization at any point in time.

I would like to discuss informal learning as a function of the way in which the layout organizes interfaces which are critical to knowledge work. The first and most obvious interface is between movement and interaction, actual or potential; the significance of this interface has been noted by Hillier and Penn (1991).At Thoughtform, all meeting rooms are linked to the main circulation or traversed by it. In addition, each cluster of workstations associated with a work team is equipped with a satellite resource hub on wheels which also invites brief meetings. This custom-designed piece of furniture provides shelves for storing documents critical to the project, whether borrowed from the main library, or assembled or produced by the team. The top surface is at elbow height to afford spreading material for review, or resting a cup of coffee or a notepad.Hubs are visible from the main circulation and more than half of them are immediately adjacent to it. Thus,interaction is supported across the layout (FIGUREE 1.5).

The second interface, as ubiquitous, is between movement and the display of visual information. Traces of ideas, diagrams, images or notes, posters and post-it notes are visible — but not necessarily readable — from almost any position. Pin-up and marker boards are designed to be carried between meeting rooms, team areas and individual stations. Surfaces of inscription are usually present wherever a brief or longer meeting can take place.

These two interfaces ensure that each individual employee sustains evolving relationships to many others, whether they working on the same project or not, whether they are endowed with similar expertise and experience or not. The network of people are the collective custodians and producers of shared knowledge. However, encounters and visual communication in space are the medium in which the network functions productively.In this example, the design of interfaces is not a simple matter of running a straight central corridor though a work environment with is never more than 10 m deep from the edge of the corridor to the perimeter of the building. Much else is involved, ranging from the arrangement of workstations into clustered group areas and the provision of meeting areas, to the presence of customized furniture and equipment. From the point of view of space syntax, one can summarize the condition as the channeling of movement along a legible integration core which is systematically invested with affordances for interaction, learning and collaborative knowledge work.The positive effects of syntactic integration result from the design of layout, furniture and equipment as integral parts of the organization and its culture.

Programmatic affordances and institutional identity:the Benaki Museum at Piraeus Street, Athens.

In 2001, Maria Kokkinou and Andreas Kourkoulas, both of whom had studied with Bill Hillier in London, were selected to design the new Benaki Museum at 138 Piraeus Street, in Athens. I worked as their architectural consultant. Piraeus Street is the historic link between Athens and the port of Piraeus. It traces the north wall of the Long Walls that defined and protected the connection in the 5th century BC. Since the 19th century, it has been associated with industry and lower income housing. In recent decades, a process of gradual transformation brings the street back into the cultural life of the city. The new Benaki Museum, completed in 2004, stands as a landmark of this process.

Since its creation in 1929, the Benaki Museum has been associated with its permanent collection of Greek culture, spanning from ancient through byzantine to the present; this is on display in a historic building at the center of Athens, near Syntagma Square. The new premises at Piraeus Street were programmed as a satellite museum that would host temporary exhibitions, whether commissioned, curated or travelling. The building was designed as a conversion and extension of pre-existing industrial structures arranged around a courtyard and mostly used for car repairs. Access was provided from a side street, taking advantage of a wide set back.

The design retained the courtyard as the fulcrum of the museum, and preserved access from the side street to control for noise and traffic on Piraeus Street. The new building is contained between two radically different boundaries: The external hard face is covered in red stone and is only pierced by a few strategically placed strip windows. The internal soft face, around three sides of the courtyard, is defined by an upwardly slanted screen, composed of custom designed horizontal iroko wood louvers that rotate to regulate incident sunlight. Access balconies run between the screen and the edge of the building on all three upper floors. On the ground floor, the projecting balconies and screen cast a shade along the perimeter. The fourth side of the courtyard, across the entrance, is defined by a dogleg ramp ascending between an inner wall and an outer glass face protected by a steel mesh. The ramp is intended to showcase the movement of visitors as an essential feature of the museum. While the ramp animates the dominant internal elevation, the screen implies potential presence along the other faces. The floor of the courtyard is completely unobstructed and covered in cobblestones. From Piraeus Street, the inner space can be glimpsed from the strip window, through the museum shop and café at the front of the building.Large openings allow the extension of the courtyard plane into the exhibition space at the back. Thus, while the dominant impression along the entrance axis is one of deliberate frontality dominated by the ramp, the cross axis suggests variable potential views through layers of space (FIGURE 2).

The museum has become a cultural landmark in Athens as a result of the innovative vision of Angelos Delivorias, who directed it until 2014, and Irini Geroulanou who has was in charge of the creation and operation of the building at Piraeus street. The museum was conceived as a meeting point and an amplifier of diverse communities of art and culture,an institution that actively engages the formation of its public (Delivorias, 2004). As part of this vision, the courtyard played a particularly significant role. It has hosted not only installations and sculpture exhibitions,but also all kinds of music and theatrical performances,events and gatherings. The popularity and flexibility of the courtyard as a cultural venue exceeded the expectations of the design team and the client organization alike. At some point in the design process,the client and some of their consultants advocated that the courtyard be turned into a planted green oasis.The idea that the courtyard could function as a flexible stage of unique character was discovered in the process of implementation of the general curatorial vision,after the building was finished. In short, the design of the courtyard provided rich affordances that helped develop, enrich and clarify the original programmatic intentions of the administration (FIGURE 3).

The following design moves contribute to the character of the courtyard as a stage. First, the cognitive and affective separation from Piraeus Street, achieved by offsetting access even as visual links are abundant on the ground floor. Second, the distinct material feel of the courtyard. Third, the fact that the building envelop around the courtyard implies and reveals human presence; visitors become aware of the reciprocity of seeing and being seen, addressing others and being addressed by others, in all of its parts. Fourth, the frontal foregrounding of movement as proscenium between courtyard and building along the axis of entry,at right angles to the traffic on Piraeus Street. Fifth, the slanted screen that pushes views upward and brings the whole volume into the purview of visual attention.At the Benaki museum visual form functions as an image of the organization and animation of space. The spatial feel of the courtyard invites the exploration of latent affordances and the embodied experience of spatial relationships; it foregrounds a desire to inhabit.

In essence, the design establishes a configuration where the double instrumental relations between street exterior and courtyard interior as well as courtyard and building interior are mediated by expressive interfaces between access and view, the continuity of movement and the ability to look across permeable or transparent boundaries. The structures of perceptual space that are deliberately overlaid upon the sequence of entry and access generate the great variety of programmatic possibilities that are continuously being discovered and enacted in the courtyard. In this way, architecture becomes part of the identity-in-action of the institution.

Transformations of a building type: a design studio on the future of retail, Georgia Tech, Atlanta.

Retail is being reorganized in response to changes in technology, economy and culture. The efficiency and reliability of web-based vendors and delivery services challenges the necessity for shops altogether. Experts argue for omni-channel retail: the integration of physical shops, web-pages and mobile applications so that customers can buy what, when and where they choose. We are dealing with a profound transformation of spatial and cultural experience of the sort that was described by Mitchell (1999). In the Spring of 2017,with funding and research-based inputs from NCR,I run an undergraduate studio, to explore what, in addition to convenience, can still be fundamental and desirable about visiting shops. The retailers that are most challenged by internet vendors are those that offer an impoverished model of service. Thus, we asked what are the intrinsic functions of shops that must be freshly reasserted.

Shops help us discover new things. When we find an unanticipated book on a subject of interest while browsing at a bookshop, or when we find a variety of cheese that we have not previously tasted in the supermarket, our sense of what is available expands.Thus, shops are domains not only of directed search,where the aim is to find what you came to buy, but also of open search. Open search helps put new products and ideas in circulation.

Shops educate. At the music shop, the subtle differences between seemingly similar musical instruments can be directly felt. The traditional butcher opines on the suitability of a meat cut to a cooking intention. A pharmacist advises regarding the treatment of minor injuries. Shops help bridge between generic needs and desires and specific practical knowhow. They help define what aspirations, habits and practices we buy-into as we purchase a product. As our sensitivities regarding the relationships of products to health, to environment and to the social conditions of production increase, so the educational function of shops may assume greater significance.

Shops thrive on relations of trust. This becomes evident with the customization of a product, as with tailoring,or the provision of personalized recommendations,as with the specialized bookshop. The shopkeeper or salesperson can be an expert and advisor.

Some incidental functions also matter. Shops can provide a sense of occasion, a social theater where we go to see and be seen by others. They can be as important as museums in getting us to develop our visual communication languages and more important as transmitters of material culture. Even when a purchase does not occur, the psychological connection to the shop brand is reinforced.

Studio participants worked in teams. Two of the teams developed different compositions in order to reorganize the interface between display, inventory storage and customer service in similar ways. The aim was to address the prevailing confusion and intermingling of these functions in most department stores.

Collin Garnett, Tim Peterson, and Christina Delurgio created a thin L-shaped gallery-like space for display along the two most prominent edges of the site. Social interaction occurs in an interlinked three-dimensional matrix of pods, within the volume defined by the display edge. The pods are designed as spaces of creation, in which the products of a given company or of a given kind are customized, altered, combined or developed according to the dialogue between the customer and store specialists. The pods are raised over an open public space at ground level. This provides access as well an independent public venue to host performances and events, or invite ordinary urban life, taking advantage of surrounding cafes, bars and restaurants. Inventory is stored underground and reaches the pods through service elevators (FIGURE 4).

Danny Griffin, Maria Pastorelli and Skylar Royal created a gallery-like zone of display, ascending over a public space scooped asymmetrically at the front of the building. As they peruse the galleries, customers are able to order products by hand-held devices. At the ground level, the public space is flanked by small shops that directly contribute to its life. On the upper floors,a storage zone behind the display galleries handles inventory and distribution. A poché wall between storage and display accommodates automatic product movement. At the top floor, customers pick up a basket that contains the products they selected. They try them on as well as discuss combinations, customization, or development with specialized staff. To support this process, the upper floor provides generous changing/consultation lounges as well as a large hall with bar,taking advantage of the spectacular views over the city. There are connections to the workshops and development laboratories of a creative studio situated over the storage zone (FIGURE 5).

In so re-ordering the department store interfaces, the design teams created visual links in three dimensions,intensifying relationships of seeing and being seen, and amplifying the sense of social occasion throughout the store. The foregrounding of display and interaction took advantage of new technologies for inventory storage and retrieval as well as of new technologies that allow customers to select products into a virtual basket.

The figures that describe the two projects bring together architectural presentation drawings, an exploded axonometric diagram that shows the definition and composition of the primary interfaces,and story boards that describe the sequences of customer experience. They point to the multiple kinds of conceptualization of geometry, movement, view and program that are involved with any intentional restructuring of the interfaces that define building types. The projects represent only one possible path for the evolution of store typology, and are, therefore,open to debate. However, the intellectual operations evident in the drawings suggest a close relationship between the creation of an abstract space of design possibility and the motivating conceptualization of the relevant interfaces.

A definition of interface

In all three examples the interfaces under consideration are associated with spatial relationships that either hold between sub-complexes, rather than individual elements, or are repeated within a pattern: relationships between spaces of interaction and movement at Thoughtform; the interplay between visibility and access in and around the Benaki courtyard; the relationship between display, storage and interaction areas in the studio projects. Thus, an interface involves classes of elements. By implication, the idea of interface implies conceptualization, the grouping of elements according to some common characteristic. In all cases the common characteristic is some affordance held to be programmatically important and associated with a syntactic condition.

I propose the following definition of interface: An interface is a complex relation between classes of elements that holds over a spatial configuration as a whole. The classes of elements are defined in terms of programmatic affordances and associated syntactic conditions. While the idea of configuration points to the effects of each relation on all others, the idea of interface requires that relations be ordered in classes. The configurational properties of the system are the material interfaces are made of. At the same, time interfaces express higher order principles of arrangement,whether these are embedded in common know-how or arise from reflexive design intentionality.

There are many different kinds of interface. For example, we have interfaces of interaction or interfaces of distinct functional zones. The example of the Benaki Museum is particularly poignant because it suggests that interfaces between perceptual patterns and movement affordances can, in particular contexts, assume significance as generators of programmatic potential.

The interfaces that are significant in any particular design, or family of designs,are a function of building type, programmatic intent, and design formulation and conceptualization. The theory of space syntax, however, suggests that while the universe of significant interfaces is open, and while the set of geometries by which interfaces are constructed is also open, a theory of interface, as defined here, is possible. Based on the cited work of Hillier (2002) and Peponis (2012), the fundamental interface in all built space is the interface between local and global scales of organization. This is at the root of the human and cultural intelligibility of space.

The development of the fundamental theory of interface and of type or programspecific models of interfaces enriches our specific explanatory theories and normative theses regarding the social and cultural logic of space. The idea of interface also bridges between the theory of space syntax and design intentionality or design formulation.

Acknowledgment

In the course of writing multiple drafts of this article I have benefited from an ongoing conversation with Bill Hillier, whose published ideas are, of course, at the core of the arguments presented, and whose comments have helped shape them.

[1] Batty M, 2001, "Exploring isovist fields: space and shape in architectural and urban morphology"Environment and Planning B: Planning and Design 28 123-150

[2] Benedikt M L, 1979, "To take hold of space: isovists and isovist fields" Environment and Planning (B)6 47-65

[3] Δεληβoρι?? A, 2004, "Mια σημαντικ? επ?τειo? και μια ευo?ωνη πρooπτικ?", in Oδ?? Πειραι?? 138.To ν?o κτ?ριo τoυ Moυσε?oυ Mπεν?κη (Benaki Museum, Athens) pp 5-7

[4] Hillier B, 1996 Space is the machine (Cambridge University Press, Cambridge)

[5] Hillier B, 2002, "A theory of the city as object: or, how spatial laws mediate the social construction of urban space" Urban Design International 7 153-179

[6] Hillier B, Leaman A, Stansall P, Bedford M, 1976, "Space syntax" Environment and Planning B:Planning and Design 3 147-185

[7] Hillier B, Hanson J, 1984 The social logic of space (Cambridge University Press, Cambridge)

[8] Hillier B, Hanson J, Peponis J, Hudson J, Burdett R, 1983, "Space syntax, a different urban perspective"Architects' Journal 178 47-63

[9] Hillier B, Penn A, 1991, "Visible Colleges: Structure and Randomness in the Place of Discovery"Science in Context 4 23-50

[10] Mitchell W J, 1999 City of bits (MIT Press, Cambridge, MA)

[11] Peponis J, 2012, "Building layouts as cognitive data: purview and purview interface" Cognitive Critique 6 11-52

[12] Peponis J, Bafna S, Bajaj R, Bromberg J, Congdon C, Rashid M, Warmels S, Zhang Y, Zimring C, 2007,"Designing space to support knowledge work" Environment and Behavior 39 815-840

[13] Peponis J, Bafna S, Zhang Z Y, 2008, "The connectivity of streets: reach and directional distance"Environment and Planning B: Planning and Design 35 881-901

[14] Peponis J, Bafna S, Dahabreh S M, Dogan F, 2015, "Configurational meaning and conceptual shifts in design" Journal of Architecture 20 215-243

[15] Peponis J, Wineman J, Rashid M, Kim S H, Bafna S, 1997, "On the description of shape and spatial configuration inside buildings: Convex partitions and their local properties" Environment and Planning B-Planning & Design 24 761-781

[16] Peponis J, Wineman J, Bafna S, Rashid M, Kim S H, 1998, "On the generation of linear representations of spatial configuration" Environment and Planning B-Planning & Design 25 559-576

[17] Russell B, 1897 An essay on the foundations of geometry (Cambridge University Press, Cambridge)

[18] Turner A, 2007, "From axial to road-centre lines: a new representation for space syntax and a new model of route choice for transport network analysis" Environment and Planning B: Planning and Design 34 539-555

[19] Turner A, Doxa M, O'Sullivan D, Penn A, 2001, "From isovists to visibility graphs: a methodology for the analysis of architectural space" Environment and Planning B: Planning and Design 28 103-121

[20] Turner A, Penn A, Hillier B, 2005, "An algorithmic definition of the axial map" Environment and Planning B: Planning and Design 32 425-444

Captions

Figure 1. Thougthform.

1.1: Thoughtform plan, 2005.

1.2: Lines analysis of all paths of movement. Syntactically central paths, which are more likely to attract movement, are shown in red.

1.3: Tiles analysis at knee level. The most directly accessible areas are shown in red.

1.4: Work related interaction network, daily time interval.

1.5: Areas programmed for holding formal or informal meetings.

(Plan by Archideas. All photographs and analytical drawings by the author.)

Figure 2. Benaki Museum, Piraeus Street.

2.1: Transverse section, facing towards the ramp.

2.2: Longitudinal section cut along the entrance axis.

2.3: Ground floor plan.

2.4: Southwest elevation showing the entrance.

2.5: Southeast elevation along Piraeus Street.

2.6: View of the courtyard from the south.

(Drawings by Kokkinou+Kourkoulas. Photograph by Erieta Attali.)

Figure 3. The courtyard of the Benaki Museum in use.

3.1: Sculpture installation, Thodoros, 2011.

3.2: Sculpture installation, Costas Varotsos, 2007.

3.3: Performance of the Oresteia by Iannis Xenakis, directed by Spyros Sakkas, 2005.

3.4: Performance of Antigone by Sophocles, directed by Natasa Triantafilli, 2013.

3.5: Large meeting held at the courtyard.

3.6: Gastronomy Days, a celebration of Greek cooking culture, 2013.

3.7: Landart, Antonis Pittas, 2013.

3.8: Exhibition, Costas Koulentianos, 2012.

(All photographs provided by the Museum.)

Figure 4. Store design proposal by Collin Garnett, Tim Peterson, and Christina Delurgio.

4.1: Approach from the urban plaza or the street.

4.2: Escalators and elevators take visitors up the display wall.

4.3: As they explore the displays, shoppers scan products with their phones and receive direction to pods.

4.4: The bridges to the pods offer spectacular views.

4.5: At the pod, shoppers consult with associates.

4.6: Associates summon products from the underground storage facility using a smart interface.

4.7: Shoppers test or try the products and discuss customization before purchase.

4.8: Purchases are completed using mobile POS systems.

Figure 5. Store design proposal by Danny Griffin, Maria Pastorelli, Skylar Royal

5.1: Approach to the urban plaza and street-level shops.

5.2: Escalators and elevators take visitors up through the display galleries.

5.3: As customers navigate the gallery, they digitally curate their selection.

5.4: While customers continue browsing, their selection is automatically retrieved from storage and collected into a personal pod.

5.5: At the top floor, customers enter lounges for comfortable and collaborative trial.

5.6: Personal pods arrive at the customer’s lounge via the delivery wall.

5.7: Lounges are linked to a large hall with views as well as a connection to the creative studios on the floor beneath.

5.8: Purchases are completed in the hall and automatically delivered to the customer on the ground floor.