唐納德·諾曼（Donald Arthur Norman，1935年12月25日－）為美國認知心理學(xué)家、計算機工程師、工業(yè)設(shè)計家，認知科學(xué)學(xué)會的發(fā)起人之一，關(guān)注人類社會學(xué)、行為學(xué)的研究。1999年，他被Upside雜志提名為世界100精英之一。Norman博士出版了大量的書籍和研究報告。他的作品有13本之多，并被翻譯成12種語言。其中最有名的要數(shù)《設(shè)計心理學(xué)》、 《情感化設(shè)計》 以及2009年出版的 《未來產(chǎn)品的設(shè)計》。
Don Norman聲稱他的目標在生活中做出了顯著的差異，但這樣做有樂趣。他是商人（在蘋果公司，惠普執(zhí)行副總裁和啟動）和學(xué)術(shù)（哈佛大學(xué)，加州大學(xué)圣迭戈分校，西北大學(xué)，KAIST）。由于諾曼尼爾森集團的創(chuàng)始人之一，他擔(dān)任公司董事會，并幫助企業(yè)使產(chǎn)品更愉快，稱心，和盈利。他是一個IDEO的研究院和國家工程學(xué)院的成員。
第一次受到優(yōu)秀產(chǎn)品設(shè)計的震撼，令我終身難忘。那時我剛加入Apple，還在熟悉業(yè)務(wù)。工業(yè)設(shè)計團隊的一名成員來訪，他給我看了一個產(chǎn)品提案的仿制模型。“哇哦，”我說，“我想要一個！這啥東西呀？”
那次經(jīng)歷讓我感受到了設(shè)計的力量：我還不知道那是什么東西，它就令我為之激動和狂熱！這種讓人不禁叫出“哇哦”的設(shè)計，只來自有創(chuàng)意的設(shè)計師。這是很主觀的、很個人的事情。你瞧瞧，工程師可不喜歡聽這個——沒辦法量化？那就不重要！如此一來，消滅設(shè)計師的趨勢就出現(xiàn)了。我們工程師只靠測試就可以走向成功，簡單得很，誰還需要設(shè)計師！有震撼力、俘獲人心的設(shè)計所帶來的激動心情，被認為是無關(guān)緊要的。更糟糕的是，設(shè)計的本質(zhì)正在被忽視和踐踏，設(shè)計正在面臨危機。
兩種創(chuàng)新：漸進式改善和新概念
談到設(shè)計，和幾乎所有的創(chuàng)新，至少有兩種截然不同的實踐形式。其一是漸進式改善（incremental improvement）。這意味著，一家企業(yè)在產(chǎn)品制造過程中的單位成本隨著對產(chǎn)品持續(xù)、漸進的改善而逐漸降低。由此形成的穩(wěn)定的漸進式創(chuàng)新鏈條，有助于運營、部件的供貨以及供應(yīng)鏈管理。持續(xù)對產(chǎn)品設(shè)計進行修補：調(diào)整界面、追加新功能、在各處做小修訂等等。在既有平臺的基礎(chǔ)上，采用不同的功能搭配組合，簡單做些微修改，就可做到每年都發(fā)布新產(chǎn)品。既減少部分功能，以便推出低端產(chǎn)品線，也可以對部分功能進行強化或追加全新的功能。采用漸進式改善，基本的平臺底子總是不變的。漸進式設(shè)計和創(chuàng)新不如開創(chuàng)新概念、新想法來的有魅力，但比后者常見得多，也重要得多。這樣的創(chuàng)新都是小創(chuàng)新，但其中大部分都非常成功。此即所謂的企業(yè)“搖錢樹（cash cows）”：這樣的一條產(chǎn)品線，只需追加很少的開發(fā)成本，就能實現(xiàn)常年獲利頗豐。
第二種形式的設(shè)計，則是教授“突破性產(chǎn)品創(chuàng)新”時所談的那種設(shè)計，廣泛見之于設(shè)計、工程和MBA課程當(dāng)中。這種設(shè)計即發(fā)明新概念、定義新產(chǎn)品、開創(chuàng)新商機，是創(chuàng)新中有趣的那一部分，因而也是大部分設(shè)計師、發(fā)明家希望盤踞的領(lǐng)地。然而這種設(shè)計的風(fēng)險是很大的：大部分創(chuàng)新會失敗。成功的創(chuàng)新可能經(jīng)歷數(shù)十年才會被廣泛接受——所以創(chuàng)新者并不一定就是獲益者。
在開頭我提到的那個Apple產(chǎn)品模型事例中，設(shè)計師就是在發(fā)明新概念。相較之下，Google和Amazon實踐的就是漸進式改善。這是兩種不同的實踐活動。和大部分創(chuàng)新一樣，那個Apple產(chǎn)品最終失敗了。為什么會失敗呢？我過一會兒再來說明。
兩種形式的設(shè)計都是必要的。圍繞“數(shù)據(jù)驅(qū)動（data-driven）”型設(shè)計的爭論是有誤導(dǎo)性的，因為其無非是用一種設(shè)計的優(yōu)勢來否定另一種設(shè)計的重要性。對于改善既有產(chǎn)品而言，數(shù)據(jù)驅(qū)動型設(shè)計確實行之有效。然而，產(chǎn)品本身又是從哪兒來的呢？當(dāng)然是來自某個有創(chuàng)意的腦袋瓜。測試有助于強化一個既有想法，前提是需要有創(chuàng)意的設(shè)計師和發(fā)明家來給出這個想法。
為什么測試很重要卻又不完善
數(shù)據(jù)驅(qū)動型設(shè)計正好比一種知名的優(yōu)化算法——“爬山（hill-climbing）”法。設(shè)想你身在一座不熟悉的山丘上，一片漆黑伸手不見五指。如果你看不見，要如何爬到山頂呢？你可以測試自己周圍的地形，哪個方向是最陡且往上的，就向哪個方向邁一步。重復(fù)探尋，直到你周圍任一方向都往下行為止。
但如果這片地區(qū)有很多山丘怎么辦呢？如何能知道你是否處于整片山丘的最高處呢？答案是：你不能知道。此即“local maximum（局部最大值）”問題：你無法判定你是在最高的山丘頂上（即全局最大值，global maximum）上，還是在一個小山丘頂上。
在數(shù)學(xué)空間中，計算機可嘗試從空間中多個不同的部分同時施行“爬山”算法，并選取所有嘗試結(jié)果中的最大值，從而避免“局部最大值”問題。這種做法仍然無法保證能取到真正的最大值，但能避免被局限在單一的局部最大值上。這種策略對設(shè)計師而言鮮能湊效。確定一個起點就已經(jīng)很不容易了，更不用說確定多個不同的起點。如此一來，通過測試來進行改進的設(shè)計只可能達到一個局部上限。測試永遠不可能告訴我們，是否存在好得多的方案（也許另一個山丘要高得多）。
于是就需要有創(chuàng)意的人來參與。當(dāng)這個人重新構(gòu)造問題，認識到之前探索的局限性，突破就會出現(xiàn)。設(shè)計和發(fā)明需要創(chuàng)意的一面。漸進式的設(shè)計無法做到這一點。
偉大創(chuàng)新的障礙
激動人心的創(chuàng)新所具備的一些根本特征，使創(chuàng)新本身不適合通過測試來進行決斷。人們對新穎設(shè)計有抵觸情緒，采取的態(tài)度會趨于保守。做事情的新技術(shù)、新方法往往要歷經(jīng)數(shù)十甚至上百年才會被接受。與此不同的是，各種基于測試的設(shè)計方式都假設(shè)，做出一個改動之后，能夠立刻測試、得到反饋，并立刻決定改動后是否比改動前更好。
我們沒有辦法判別激進的新想法最終是否能成功。我們還需要偉大的領(lǐng)頭者和勇氣。歷史告訴我們，有許多人面對一次又一次的拒絕和抵觸，堅持了很長時間，其想法才終獲接受。這些成功者經(jīng)常指出，在產(chǎn)品獲得成功后，人們就無法想象以前沒有這個產(chǎn)品的時候是怎么過的了。歷史也告訴我們，有許多人堅持過，最終也未獲得成功。對激進的新想法持懷疑態(tài)度并不為過。
一個初成的想法不被接受，因素很多：可能是因為技術(shù)還不成熟，可能是因為還有很多東西有待優(yōu)化，可能是因為受眾群體還沒有做好接受它的準備，也可能是因為這是個糟糕的想法。判定其中的主導(dǎo)因素是很困難的——是在確立想法很久之后，才會得到的后見之明。
一個激動人心的想法，從想法形成并初步實現(xiàn)，到最終認定其在市場中的成功或失敗，歷時長久。 有些人想以證據(jù)作為標準，對新發(fā)展方向進行定奪，卻被這漫長的時間差所擊敗。 更好的方案 即使曾經(jīng)被提出過 ，也可能會被自動化測試否決掉——這并不是因為它不好，而是因為它等不了數(shù)十年的時間來獲得認可。只看測試結(jié)果的人注定會錯過巨大的回報。
當(dāng)然，有很多合理的商業(yè)考慮能夠解釋，為什么忽略有可能更好的方案是明智的。畢竟，如果受眾沒有做好接受新想法的準備，這個新想法一開始就是會在市場中失敗。短期看來確實如此。但若要想在未來獲得成功，最佳的方案是先發(fā)展新想法并將其商業(yè)化，投入市場以獲取經(jīng)驗，并不斷地進行優(yōu)化，發(fā)展客戶基礎(chǔ)。同時，公司還要做好準備，應(yīng)對現(xiàn)有方案之不測。既要保持把現(xiàn)有的做好，還要準備隨時迎接新的。如果公司沒能洞察到新趨勢，其競爭對手就會迎頭趕上，接手市場。這些競爭對手往往是被現(xiàn)有公司忽略的小創(chuàng)業(yè)團隊。之所以被忽略，是因為這些新來者的所作所為還不太為市場所接受，無論如何都不像是老公司現(xiàn)有業(yè)務(wù)的有力挑戰(zhàn)者。請參見“創(chuàng)新者的困境（The Innovator's Dilemma） ”，以了解這種公司的運營困境。
用于屏幕驅(qū)動（screen-driven）型設(shè)備和電子游戲的勢控（gestural）界面和多點觸控界面，正是 兩個久經(jīng)蹉跎才成功的創(chuàng)新例子。 它們難道不是杰出的創(chuàng)新嗎？當(dāng)然是。它們難道不杰出嗎？當(dāng)然杰出。但是它們新嗎？絕對不新！多點觸控設(shè)備在研究實驗室里等待了近30年，才首次迎來大規(guī)模量產(chǎn)的成功產(chǎn)品。20年前我就見過勢控界面演示。新想法要花上相當(dāng)可觀的時間，才會在市場上獲得成功。過快地把想法商業(yè)化，往往以失敗（以及大筆的資金損失）而告終。
當(dāng)年那位給我看模型的Apple設(shè)計師同事也未能幸免。他給我看的是一臺為小學(xué)生設(shè)計的便攜設(shè)備，其外形設(shè)計不同于我之前所見的任何東西。那真是絕妙的設(shè)計——即便是在我這通常很挑剔的眼里，其設(shè)計也完美切合了其用途和受眾?？上У氖牵罱K產(chǎn)品成了Apple公司部門間內(nèi)訌的犧牲品。盡管產(chǎn)品最終被投放到了市場中，但部門間的不合導(dǎo)致了糟糕的實施、糟糕的產(chǎn)品支持和糟糕的市場推廣，破壞了產(chǎn)品的整體性。
公司抵觸完全地創(chuàng)新，也有根有據(jù)。在不能確定贏利潛力的情況下開發(fā)新產(chǎn)品線，代價是很高的。而且現(xiàn)有產(chǎn)品的責(zé)任部門也會擔(dān)心新產(chǎn)品打壓了現(xiàn)有產(chǎn)品的銷售（這叫做“同類相食”）。這些擔(dān)憂一般都是合理的。這種形勢也屬經(jīng)典案例，即有益于公司的好事情對現(xiàn)有產(chǎn)品部門來說卻是壞事情，因為那意味著現(xiàn)有產(chǎn)品部門職員得到升遷和獎勵的機會不容樂觀。如此想來，公司會抵觸創(chuàng)新也就不足為奇了。統(tǒng)計數(shù)據(jù)清楚地表明，盡管極少數(shù)創(chuàng)新取得了非凡的成功，但絕大部分創(chuàng)新都失敗了并付出慘重代價。無論公司的新聞稿和年度報告里怎么說，公司都會猶豫甚至抵觸創(chuàng)新，這都不足為奇，因為持保守態(tài)度是明智的。
展望未來
數(shù)據(jù)驅(qū)動的自動化流程會慢慢侵占如今人類設(shè)計師所掌握的地盤。諸如基因算法、知識密集型系統(tǒng)等等這些依靠計算機生成創(chuàng)意的新方法會開始接管設(shè)計的創(chuàng)意空間。醫(yī)療診斷或工程設(shè)計等其他領(lǐng)域也正在發(fā)生相同的變化。
我們將面對更多無需設(shè)計師的設(shè)計，但主要只限于在對既有概念的強化、精化和優(yōu)化方面。即使到了以后，神經(jīng)網(wǎng)絡(luò)、基因算法，抑或其他某種尚未被發(fā)現(xiàn)的方法都能被用來開發(fā)新的、有創(chuàng)意的人工系統(tǒng)了，任何新概念也還是須要面對同樣的困難，經(jīng)歷漫長的接受周期，？？人類在心理上的、社會上的和政治上的復(fù)雜需求。要做到這一點，我們需要有創(chuàng)意的設(shè)計師、有創(chuàng)意的商業(yè)人士和有冒險精神的人來突破極限。會有新想法遭到抵觸。許多偉大的創(chuàng)新將以更多巨大的失敗為代價。
無需設(shè)計師的設(shè)計？有些人討厭人類判斷的含糊性和不確定性，討厭人類不靠譜的過往表現(xiàn)和自相矛盾的論調(diào)。這些人會嘗試剝離設(shè)計中的人為因素，轉(zhuǎn)投數(shù)字和數(shù)據(jù)和懷抱，只因為數(shù)字和數(shù)據(jù)看起來似乎能提供確定性。還有一些人希望借助創(chuàng)意來得到巨大收獲，他們會遵循自己的原則來做。前者會帶來持續(xù)的小改進，顯著提高生產(chǎn)力并降低成本。后者會面對巨大的失敗，并迎接偶然發(fā)生的巨大成功——這些巨大成功會改變世界。
————以下是英文原文：
I will always remember my first introduction to the power of good product design. I was newly arrived at Apple, still learning the ways of business, when I was visited by a member of Apple's Industrial Design team. He showed me a foam mockup of a proposed product. "Wow," I said, "I want one! What is it?"
That experience brought home the power of design: I was excited and enthusiastic even before I knew what it was. This type of visceral "wow" response requires creative designers. It is subjective, personal. Uh oh, this is not what engineers like to hear. If you can't put a number to it, it's not important. As a result, there is a trend to eliminate designers. Who needs them when we can simply test our way to success? The excitement of powerful, captivating design is defined as irrelevant. Worse, the nature of design is in danger.
Don't believe me? Consider Google. In a well-publicized move, a senior designer at Google recently quit, stating that Google had no interest in or understanding of design. Google, it seems, relies primarily upon test results, not human skill or judgment. Want to know whether a design is effective? Try it out. Google can quickly submit samples to millions of people in well-controlled trials, pitting one design against another, selecting the winner based upon number of clicks, or sales, or whatever objective measure they wish. Which color of blue is best? Test. Item placement? Test. Web page layout? Test.
This procedure is hardly unique to Google. Amazon.com has long followed this practice. Years ago I was proudly informed that they no longer have debates about which design is best: they simply test them and use the data to decide. And this, of course, is the approach used by the human-centered iterative design approach: prototype, test, revise.
Is this the future of design? Certainly there are many who believe so. This is a hot topic on the talk and seminar circuit. After all, the proponents ask reasonably, who could object to making decisions based upon data?
Two Types of Innovation: Incremental Improvements and New Concepts
In design—and almost all innovation, for that matter—there are at least two distinct forms. One is incremental improvement. In the manufacturing of products, companies assume that unit costs will continually decrease through continual, incremental improvements. A steady chain of incremental innovation enhances operations, the sourcing of parts and supply-chain management. The product design is continually tinkered with, adjusting the interface, adding new features, changing small things here and there. New products are announced yearly that are simply small modifications to the existing platform by a different constellation of features. Sometimes features are removed to enable a new, low-cost line. Sometimes features are enhanced or added. In incremental improvement, the basic platform is unchanged. Incremental design and innovation is less glamorous than the development of new concepts and ideas, but it is both far more frequent and far more important. Most of these innovations are small, but most are quite successful. This is what companies call "their cash cow": a product line that requires very little new development cost while being profitable year after year.
The second form of design is what is generally taught in design, engineering and MBA courses on "breakthrough product innovation." Here is where new concepts get invented, new products defined, and new businesses formed. This is the fun part of innovation. As a result, it is the arena that most designers and inventors wish to inhabit. But the risks are great: most new innovations fail. Successful innovations can take decades to become accepted. As a result, the people who create the innovation are not necessarily the people who profit from it.
In my Apple example, the designers were devising a new conception. In the case of Google and Amazon, the companies are practicing incremental enhancement. They are two different activities. Note that the Apple product, like most new innovations, failed. Why? I return to this example later.
Both forms of innovation are necessary. The fight over data-driven design is misleading in that it uses the power of one method to deny the importance of the second. Data-driven design through testing is indeed effective at improving existing products. But where did the idea for the product come from in the first place? From someone's creative mind. Testing is effective at enhancing an idea, but creative designers and inventors are required to come up with the idea.
Why Testing Is Both Essential and Incomplete
Data-driven design is "hill-climbing," a well-known algorithm for optimization. Imagine standing in the dark in an unknown, hilly terrain. How do you get to the top of the hill when you can't see? Test the immediate surroundings to determine which direction goes up the most steeply and take a step that way. Repeat until every direction leads to a lower level.
But what if the terrain has many hills? How would you know whether you are on the highest? Answer: you can't know. This is called the "local maximum" problem: you can't tell if you are on highest hill (a global maximum) or just at the top of a small one.
When a computer does hill climbing on a mathematical space, it tries to avoid the problem of local maxima by initiating climbs from numerous, different parts of the space being explored, selecting the highest of the separate attempts. This doesn't guarantee the very highest peak, but it can avoid being stuck on a low-ranking one. This strategy is seldom available to a designer: it is difficult enough to come up with a single starting point, let alone multiple, different ones. So, refinement through testing in the world of design is usually only capable of reaching the local maximum. Is there a far better solution (that is, is there a different hill which yields far superior results)? Testing will never tell us.
Here is where creative people come in. Breakthroughs occur when a person restructures the problem, thereby recognizing that one is exploring the wrong space. This is the creative side of design and invention. Incremental enhancements will not get us there.
Barriers to Great Innovation
Dramatic new innovation has some fundamental characteristics that make it inappropriate for judgment through testing. People resist novelty. Behavior tends to be conservative. New technologies and new methods of doing things usually take decades to be accepted – sometimes multiple decades. But the testing methods all assume that one can make a change, try it out, and immediately determine if it is better than what is currently available.
There is no known way to tell if a radical new idea will eventually be successful. Here is where great leadership and courage is required. History tells us of many people who persevered for long periods in the face of repeated rejection before their idea was accepted, often to the point that after success, people could not imagine how they got along without it before. History also tells us of many people who persevered yet never were able to succeed. It is proper to be skeptical of radical new ideas.
In the early years of an idea, it might not be accepted because the technology isn't ready, or because there is a lot more optimization still to be done, or because the audience isn't ready. or because it is a bad idea. It is difficult to determine which of those reasons dominates. The task only becomes easy in hindsight, long after it becomes established.
These long periods between formation and initial implementation of a novel idea and its eventual determination of success or failure in the marketplace is what defeats those who wish to use evidence as a decision criterion for following a new direction. Even if a superior way of doing something has been found, the automated test process will probably reject it, not because the idea is inferior, but because it cannot wait decades for the answer. Those who look only at test results will miss the large payoff.
Of course there are sound business reasons why ignoring potentially superior approaches might be a wise decision. After all, if the audience is not ready for the new approach, it would initially fail in the marketplace. That is true, in the short run. But to prosper in the future, the best approach would be to develop and commercialize the new idea to get marketplace experience, to begin the optimization process, and to develop the customer base. At the same time one is preparing the company for the day when the method takes off. Sure, keep doing the old, but get ready for the new. If the company fails to recognize the newly emerging method, its competitors will take over. Quite often these competitors will be a startup that existing companies ignored because what they were doing was not well accepted, and in any event did not appear to challenge the existing business: see "The innovator's dilemma."
Gestural, multi-touch interfaces for screen-driven devices and computer games are good examples. Are these a brilliant new innovation? Brilliant? Yes. New? Absolutely not. Multi-touch devices were in research labs for almost three decades before the first successful mass-produced products. I saw gestures demonstrated over two decades ago. New ideas take considerable time to reach success in the marketplace. If an idea is commercialized too soon, the result is usually failure (and a large loss of money).
This is precisely what the Apple designer of my opening paragraph had done. What I was shown was a portable computer designed for schoolchildren with a form factor unlike anything I had ever seen before. It was wonderful, and even to my normally critical eye, it looked like a perfect fit for the purpose and audience. Alas, the product got caught in a political fight between warring Apple divisions. Although it was eventually released into the marketplace, the fight crippled its integrity and it was badly executed, badly supported, and badly marketed.
The resistance of a company to new innovations is well founded. It is expensive to develop a new product line with unknown profitability. Moreover, existing product divisions will be concerned that the new product will disrupt existing sales (this is called "cannibalization"). These fears are often correct. This is a classic case of what is good for the company being bad for an existing division, which means bad for the promotion and reward opportunities for the existing division. Is it a wonder companies resist? The data clearly show that although a few new innovations are dramatically successful, most fail, often at great expense. It is no wonder that companies are hesitant – resistant – to innovation no matter what their press releases and annual reports claim. To be conservative is to be sensible.
The Future
Automated data-driven processes will slowly make more and more inroads into the space now occupied by human designers. New approaches to computer-generated creativity such as genetic algorithms, knowledge-intensive systems, and others will start taking over the creative aspect of design. This is happening in many other fields, whether it be medical diagnosis or engineering design.
We will get more design without designers, but primarily of the enhancement, refinement, and optimization of existing concepts. Even where new creative artificial systems are developed, whether by neural networks, genetic algorithms, or some yet undiscovered method, any new concept will still face the hurdle of overcoming the slow adoption rate of people and of overcoming the complex psychological, social, and political needs of people. To do this, we need creative designers, creative business people, and risk takers willing to push the boundaries. New ideas will be resisted. Great innovations will come at the cost of multiple great failures.
Design without designers? Those who dislike the ambiguity and uncertainty of human judgments, with its uncertain track record and contradictory statements will try to abolish the human element in favor of the certainty that numbers and data appear to offer. But those who want the big gains that creative judgment can produce will follow their own judgment. The first case will bring about the small, continual improvements that have contributed greatly to the increased productivity and lowering of costs of our technologies. The second case will be rewarded with great failures and occasional great success. But those great successes will transform the world.
