????上世紀(jì)80年代初，微軟（Microsoft）發(fā)布MSDOS操作系統(tǒng)的時(shí)候不過是華盛頓州雷德蒙德的一家創(chuàng)業(yè)公司。但該公司從一開始就堅(jiān)信其新產(chǎn)品非比尋常，MSDOS的設(shè)計(jì)允許其輕易兼容于不同的硬件，降低了潛在用戶進(jìn)入的成本門檻。而且微軟鼓勵(lì)參與者根據(jù)特定環(huán)境調(diào)整MSDOS，這意味著隨著時(shí)間流逝，這個(gè)系統(tǒng)可以不斷得到改進(jìn)。

????不過，與其他任何優(yōu)秀平臺(tái)一樣，MSDOS的價(jià)值在于其網(wǎng)絡(luò)效應(yīng)。盡管這個(gè)系統(tǒng)發(fā)布之初的用戶基礎(chǔ)不是很龐大，但微軟很快就與IBM和英特爾（Intel）等科技巨子談妥了合作關(guān)系，刺激了增長(zhǎng)預(yù)期，推動(dòng)更多先行者成為其用戶。通過這種方式，這個(gè)平臺(tái)迅速獲得了臨界質(zhì)量，實(shí)現(xiàn)了網(wǎng)絡(luò)效應(yīng)。這種狀態(tài)非常神奇，它意味著隨著加入的成員日益增加，網(wǎng)絡(luò)之于每個(gè)參與者的價(jià)值也隨之提高。而隨著這個(gè)網(wǎng)絡(luò)不斷升值，它就會(huì)進(jìn)一步吸引更多的人參與進(jìn)來。間接成本往往會(huì)拖累、甚至限制傳統(tǒng)商業(yè)網(wǎng)絡(luò)的發(fā)展，但由于微軟依賴不斷增長(zhǎng)的經(jīng)濟(jì)效益來吸引、招攬參與者，從而顯著降低了這種間接成本。

????或許，比微軟的網(wǎng)絡(luò)規(guī)模本身更重要的是，個(gè)體與MSDOS平臺(tái)互動(dòng)并對(duì)其進(jìn)行調(diào)整的方式。從某些方面來看，通用平臺(tái)使參與者置身于公平的競(jìng)爭(zhēng)環(huán)境之上，使他們有動(dòng)力去以改善性能的方式“保護(hù)自己的地盤”。隨著參與者的數(shù)量持續(xù)增加，競(jìng)爭(zhēng)壓力隨之加劇，贏得成功的回報(bào)也不斷增長(zhǎng)，永遠(yuǎn)維持圍繞標(biāo)準(zhǔn)平臺(tái)不斷創(chuàng)新的良性循環(huán)。

????在許多情況下，不同參與者建立了協(xié)作，共同來改善網(wǎng)絡(luò)，而另一些參與者則選擇孤軍奮戰(zhàn)。無需微軟介入調(diào)停，參與者之間的這些互動(dòng)自發(fā)演變成了復(fù)雜的合作網(wǎng)絡(luò)，互動(dòng)不僅存在于個(gè)體與個(gè)體之間，還存在于不同參與者組成的集團(tuán)之間。各種實(shí)驗(yàn)因此如雨后春筍般地涌現(xiàn)，許多前途廣闊的點(diǎn)子很快被其他參與者采納，而失敗之舉也成了眾人借鑒的教材。如此一來，每個(gè)參與者以及作為一個(gè)整體的生態(tài)系統(tǒng)學(xué)習(xí)速度都比自己?jiǎn)为?dú)摸索要快得多。

????微軟首個(gè)平臺(tái)的發(fā)布似乎已經(jīng)是遠(yuǎn)古時(shí)代的事了，但它已然彰顯了我們所謂的“網(wǎng)絡(luò)生態(tài)系統(tǒng)”的巨大潛力。我們?cè)谧匀唤绨l(fā)現(xiàn)的網(wǎng)絡(luò)足以使人心生敬畏，而且無從解釋。比如，一個(gè)結(jié)構(gòu)精密的蜘蛛網(wǎng)可從其中心開始，不斷蔓延到似乎不可思議的距離。就算風(fēng)狂雨驟，臨近的樹葉飄零、枝條墜地，這張網(wǎng)依然安然無恙。因此我們不難理解，為什么很多領(lǐng)域都借用了網(wǎng)絡(luò)這個(gè)兼具力量與精妙的象征符號(hào)（沒錯(cuò)，萬維網(wǎng)也是其中一例）。

????我們?cè)賮硌芯恳幌律鷳B(tài)系統(tǒng)，也就是即商業(yè)參與者通過某種形式的合作組成的集群，我們發(fā)現(xiàn)這個(gè)了無新意的詞匯完美地凸顯了MSDOS和其他網(wǎng)絡(luò)的微妙之處。這些生態(tài)系統(tǒng)格外有趣的地方在于它的可拓展性——網(wǎng)絡(luò)擁有調(diào)動(dòng)數(shù)十萬乃至數(shù)百萬參與者的潛力。網(wǎng)絡(luò)的可拓展性不僅表現(xiàn)在人數(shù)方面，它們所能支撐的創(chuàng)新也比其他合作型生態(tài)系統(tǒng)范圍更廣。蜘蛛網(wǎng)的邊緣往往出現(xiàn)錯(cuò)綜復(fù)雜的圖案，與此相似，特定生態(tài)系統(tǒng)的不同區(qū)域也會(huì)出現(xiàn)獨(dú)特而令人興奮的創(chuàng)新。

?

????

????Microsoft was still a no-name startup based in Redmond, Wash. when it launched its operating system, MSDOS in the early 1980s. From the beginning, however, the company believed that this was not your average product launch. MSDOS's design allowed it to adapt easily to different hardware, reducing entry costs for potential users. And Microsoft encouraged participants to tailor MSDOS for particular environments, meaning that the product could actually improve over time.

????Like any good platform, however, MSDOS was only as valuable as its network. While it did not have a large user base in its early years, Microsoft (MSFT) soon negotiated relationships with tech giants like IBM (IBM) and Intel (INTC), fueling growth expectations and motivating more early adopters to sign on. In this way, the platform was able to quickly gain critical mass and achieve network effects -- that wonderful position when the value of the network increases for all participants as more members join. This growing value helped to attract even more participants. By relying on growing economic incentives to attract and engage participants, Microsoft significantly reduced the overhead costs that often slow, or even limit, the growth of more conventional business networks.

????Perhaps even more important than its network size was how individuals interacted with and tailored the MSDOS platform. In some respects, the common platform leveled the playing field among participants, creating incentives for them to "protect their turf" by improving their own performance. As the number of participants continued to expand, both competitive pressures as well as the spoils of success increased, perpetuating a cycle of continuous innovation around the standard platform.

????In many cases, different participants teamed up to make the network better, while others chose to work on their own. Without any intervention from Microsoft, these interactions among participants evolved into complex webs of collaboration, where the interactions were not just one-to-one, but among groups of different players. This resulted in an explosion of experimentation. Many of these promising ideas were quickly adopted by other participants, and those that failed served as collective learning material. In this way, each participant, and the ecosystem as a whole, learned much faster than they would have on their own.

????The first release of the Microsoft platform may seem like ancient history, but it illustrates the powerful potential of what we call "web ecosystems." In nature, the webs we discover can inspire awe and defy explanations. A well-constructed cobweb, for instance, can span a seemingly impossible distance from its central point and remain intact even in heavy rains and wind that send neighboring leaves and branches to the ground. It is no wonder, therefore, that this image of strength and subtlety has become an oft-referenced metaphor in so many other domains (yes, we're speaking of the World Wide Web, among others).

????As we researched ecosystems (gatherings of business participants engaged in some form of collaboration), we found that the well-worn term perfectly characterized the nuances of MSDOS and other webs. These ecosystems are particularly interesting because of their scalability -- webs have the potential to mobilize hundreds of thousands, even millions, of participants. More than just scalability in numbers, webs also support a much broader range of innovation than other kinds of collaborative ecosystems. Just as intricate patterns emerge on the edges of spiders' webs, so too can distinct and exciting innovations emerge from different pockets of a given ecosystem.