Mark Zuckerberg's address at Stanford University

By Sophie

在這篇演講中，扎克伯格談了社交網(wǎng)站Facebook的創(chuàng)業(yè)過程，包括產(chǎn)品開發(fā)、團隊建設、創(chuàng)收、企業(yè)管理技巧、新領域投資以及未來的挑戰(zhàn)等。

I actually studied Psychology in Harvard,not Computer Science.Although a little bit of Computer Science.I've been programming since I was ten.And I think that it kind of like went,reach the point where went into my intuition.

I’m not really thinking that much about it consciously.So that was pretty good.And then,I mean when I started thinking about all the people issues and doing Psychology,it was like being in the university interacting with a lot of people,that it kind of occurred to me that this would be something I was interested in,and like I knew how to do it so I just do it.I mean,it took me like a couple of weeks here,even less,putting together the site.

And I remember that by the time I was done putting together the site.I had no idea how successful it could be.And I was actually thinking that after day in and day out.I had a different idea that I wanted to do,and I was going to scrap it after that.So I'm happy that I didn't do that.So I think that it’s more like how you spend your time doing stuff type of question,or type of answer than like something that I’ve learned specifically from college.

I made a ton of random things when I was at Harvard and most of them no one ever saw.A lot of them just weren’t meant for other people to see.And there are thing that I made for myself because I thought they’ll be cool.I used to make stuff like the natural language interface to play my MP3.Or a thing I made before,this was a Hot or Not program out of everyone’s IDs at Harvard,that almost got me kicked out.

So I don’t know,I actually spent a long time making random stuff.I think that that definitely made it.So by the time it came,more like,by the time I came to meet this random project.I was pretty well tuned toward making that,you know in terms of managing this whole process,nothing.Like I have no idea what I’m doing，you know.

When you're running a site,and you're four people,are on the kitchen,your operating expenses are relatively low.So traditionally,what we’ve done was we had a vey small sales force,and we sell some ads.You might see them on the site,and we just keep our operating expenses low so far and by doing that，we've been able to stay cash flow positive for basically the entire existence of the company.

After we took some money from this guy,we decided that it is OK to go a few months in casual negative.While(Jim)Barely any cash flow.Yeah,at a point,you know.Losing like a$100,000 a month,not millions.So，and then...Now we're back.And we do a lot of page views.I think that it's not something that you really think about,because you probably just think about this as a Stanford site.

But everyday,we do more than 200 million page views.I think recently were up to 230 million.By the end of,like two weeks from now or so.We're going to pass Google in page views.That’s a lot,right?You don’t normally think about it in that kind of application.Like，I think,you know,I was pretty surprised when I heard that.

But when you have that many page views and that many people spending much time on the site.You can monetize it pretty easily just by putting matters on there.The revenue,we’re generating over million dollars a month in revenue,and well more,and that way covers our expenses,and we’re not even doing anything cool yet.

We're designing stuff,we look not necessarily just about what any given users going to experience,but what’s going to be better for the whole coummunity and the whole product.And I mean,it's gonna be a lot of trade-off going all over the place in a product.Probably the most that you see every day is that you can’t see the profiles of peo ple at other schools.That’s a really a major trade-off in the application.

For those of you who aren’t familiar with this,we spilt up the user base by what school they go to and we make it so that people at a given school can only see the profiles and contact information of people at their school.And the reason for this was mostly to,because we realized that the people around you,at your school,are the people who you want to look at mostly anyway.And if we made the space too broad and let anyone see your information.Then that probably fine.I mean,look up some people,but you also probably won’t put up your cellphone.

More than a third of people on Facebook have their cellphone up there,and that's something useful for the application.So in designing it,this was a trade-off that we made.I kind of thought of what was wrong.I said,well,what would be more useful?Would it be better for people to be able to see everyone and maybe not feel like this is a secure environment in which they can share their interests and what they thought and what they care about?Or would be better that more information and more expression was available,but to a smaller audience which is probably the relevant audience for any person?So，there are a lot of decisions like that,that are getting made and a lot of them are gut level.

So I mean,we tried to be as academic about this possible in trying to think rigorously through the different results that we’ll get if we go to different directions.But a lot of it is just like you define your objectives,what you’re going for.In this case,to optimize for the best of the whole community and the whole user base and over the long term,and that’s important too,long term but short term.And then,you just kind of operate and do what you think will be best along that line.

其實我在哈佛念的是心理學，不是計算機。但也學了點計算機，我從10歲就開始編程，我覺得編程已經(jīng)成了我本身的一種技能。

我在寫程序時不再會有意識地想該怎么做。這種感覺很好。當我在學心理學的 時候，開始思考人與人之間的關系，這就好像在大學里讀書，去結(jié)識很多的人，我覺得我會對這些很感興趣，我知道該怎么做，然后就做了。我只花了幾周甚至更少的時間建立了這個網(wǎng)站。

我記得當我建好網(wǎng)站后，我不知道這個網(wǎng)站會不會成功。我真的在想，等我有了新點子想去實踐時，我就把這個網(wǎng)站丟棄不管了。現(xiàn)在我很慶幸當初沒那么做。 所以我覺得這個問題的關鍵更在于你如何支配自己的時間去做各種事情，而不在于你我在大學里學到了什么具體的技能。

我在哈佛時編了很多各式各樣的程序，大多數(shù)沒有人見過，因為那都是為我自己編的，我為自己編程序，因為我覺得那很有意思。我為我的MP3做了一個自然語言界面，之前還用哈佛每個人的ID做過一個類似Hot or Not的網(wǎng)站，我曾因此差點被哈佛開除。

我真的花了很多時間編寫各種奇怪的程序。這對我后來做的事很有幫助。當這個奇怪的點子冒出來的時候，我已經(jīng)準備得比較充分，可以實現(xiàn)它了。但說到怎么規(guī)劃整個過程，我沒想過，我都是想到什么做什么。

當你經(jīng)營一個網(wǎng)站時，你們有四個人，圍坐在餐桌旁工作，你的運營成本是比較低的。之前我們一直只有一個很小的銷售團隊，主要靠廣告賺錢。你巳經(jīng)在網(wǎng)站上看到了，而且我們一直努力將成本控制在最低，這樣一來，我們就能保持現(xiàn)金流入大于現(xiàn)金流出，基本上公司創(chuàng)辦以來一直是這樣的。

所以當我們從這個家伙（吉姆布雷耶）那里得到一些錢后，我們決定允許幾個月讓現(xiàn)金流出現(xiàn)負值，可是吉姆基本上沒有什么現(xiàn)金流。對，就是有段時間，一個月?lián)p失10萬美元，不是幾百萬，不過現(xiàn)在好了，我們的頁面訪問量很大，大家可能不會仔細去想這個，因為你可能只把它當做一個斯坦福的一個網(wǎng)站。

可是每天，我們都有超過2億的頁面訪問量，最近升到了2.3億。這樣算來，兩周后，我們就能超過谷歌的訪問量。很多吧！以前你可能從沒想過這種網(wǎng)站會那么受歡迎，我第一次聽說時也很吃驚。

但是當你有那么多的訪問量，那么多的人花大量時間在你的網(wǎng)站上，你單靠投放廣告就能輕松賺錢，我們每個月都有超過100萬的營業(yè)收入，可能更多，這早就超過了我們的開支，關鍵的這是我們還沒使用任何出色措施的情況下。

當我們設計應用程序的時候，我們并不注重單一用戶的使用體驗，而更關注其是否有益于與整個社區(qū)和產(chǎn)品。這就需要在開發(fā)產(chǎn)品的各個環(huán)節(jié)進行取舍?？赡苣?注意到你不能瀏覽其他學校的人的信息。這就是一種取舍后的結(jié)果。

你們有些人對此不太了解，我們將用戶按照學校分類，所以只有來自相同學校的人才能互相看到對方的信息和聯(lián)系方式。之所以這樣做，是因為我們發(fā)現(xiàn)人們更樂于關注身邊的人的信息。如果我們把空間擴展到所有人都可以瀏覽你的信息，那可能也不錯，誰你都找得到，但是可能你就不愿意把你的手機號碼放上去了。

超過三成的Facebook用戶在注冊時登記過電話號碼，這對我們的應用程序是很重要的。所以在設計產(chǎn)品的時候就要做這種取舍。我覺得這樣做好像不對，一直在想什么會更有用？將用戶的信息對所有人開放是不是更好一些呢？但這卻會讓人們覺得在這樣的網(wǎng)絡中分享自己的所感所想不太安全?；蛘邇H將更多的用戶信息和狀態(tài)展示給少部分與該用戶有關的人更好一些呢？我要做很多類似的決定，而且這些決定是要靠直覺判斷的。

因此，我的意思是，我們一直努力以最學術的態(tài)度去謹慎地思考不同方式所能產(chǎn)生的不同結(jié)果。但多數(shù)時候，你得先確定你的目標，你想要的是什么。對我們來說，是尋求長期的整個社區(qū)及用戶群體的利益的最大化，注意是長期的，而不是短期的。然后就是為更好地達到這一目標而努力。