以下是我对风险投资行业的三点认识：

    1.一直到五、六年前，风险投资模式基本还与上世纪80年代和90年代差不多。人变了，投资额增加了，但总体架构和策略没变。

    2.如今，有很多不同的新模式。有些专注于向创业者提供各种融资服务（市场营销、人力资源等等）。有些采用众筹模式融资。有些强调能更快做出决定，可能比“传统”风投熨一件蓝衬衣还要快。

    3.如果从创造高回报（这也是风投人真正的职责所在）的角度来考察，我们对于哪种模式更好并没有概念。

    这些在萨拉•莱西上周末题为《经典风投保卫战》（A rare defense of venture capital classic）的文章中均有阐述。文章旨在反驳这样的观念：越新越好，或者说风投公司必须构建“多平台”以便获得日后成功。事实上，她认为，当今风投的本质特征与过去几十年并没有什么差异——无论它们的模式被视为是创新模式、还是传统模式。

    不错，相比早年热投电子产品组装线和果园时期，风投行业已发生了改变。相比50年前，如今风投公司介入投资的时期更晚，获得的股份也更少。在演示活动(Demo Day)、科技博客和AngelList盛行的当下，交易信息也不再那么专有，变得更加公开。而且，我们发现，很多人做风投真的很差劲...

    但表现出色的风投不是这种做法，他们已搭上新营销趋势的顺风车。它们这么做，因为它们是不错的风投，就像风险投资之父阿瑟•洛克、硅谷风险投资之父唐•瓦伦丁和知名投资人汤姆•珀金斯是出色的风投人一样。他们承担风险。他们辅导创业者。他们尊重创业者的计划，就算它背离他们自己的计划。而最重要的是，他们拥有数百万的资金来投资到每家公司身上。

    对于莱西此文的回应基本上可以预见。“经典”风投人士公开感谢他，而有些新一代风投人士指责她。

    但大多数回应都搞错了一点，莱西没有真正站在哪一边。相反，她正努力引导创业者走过风投自我提升的进阶迷途。

    更重要的是，现在如果站在某一边，非常愚蠢。正如我在前面所写，没有足够的数据支持哪种模式最优。请注意，四年前，安德森-霍罗维茨风险投资公司（Andreessen Horowitz）才筹集了第一只基金。不管是AngelList，还是戴夫•麦库的500 Startups都是到了2010年才出现的。可作参考的是，马萨诸塞州养老金系统只披露五年及五年以上的风投基金回报率，因为它发现早期业绩的统计意义不大（由于J曲线效应，这是普遍看法）。

    Here are three things I believe about the venture capital industry:

    1. Until five or six years ago, the venture capital model was largely unchanged from what it had been in the 1980s and 1990s. Different people and bigger dollars, but the same general structure and strategy.

    2. Today there are a variety of new models. Some focus on providing a variety of non-finance services to entrepreneurs (marketing, HR, etc.). Some borrow from the crowdfunding model. Some emphasize their ability to make decisions in less time than it takes a "traditional" VC to iron his blue shirt.

    3. We have no idea which model is actually better, from the perspective of generating high returns (which is the actual job of venture capitalists).

    This is all in the context of a weekend post by Sarah Lacy, titled "A rare defense of venture capital classic." It sought to tamp down on the reflexive notion that newer is better, or that VC firms must become "platforms" in order to achieve future success. Instead, she argues that the fundamental characteristics of today's VCs are the same as they were in past decades -- no matter whether their model is considered innovative or traditionalist:

    Yes, things have changed about venture capital since those halcyon days of silicon assembly lines and fruit orchards. Typical venture firms invest later and get far smaller stakes than they did 50 years ago. Deal flow is far less proprietary in an age of demo days, tech blogs, and AngelList. And we've learned that many people are just awful at the job of venture capital...

    But VCs who perform well aren't doing it, because they are jumping on the bandwagon of new marketing trends. They are doing it because they are good VCs the way Arthur Rock, Don Valentine, and Tom Perkins were good VCs. They take risk. They coach entrepreneurs. The respect an entrepreneur's plan, even if it deviates from their own. And most of all — they have millions to invest in each company.

    The reaction to Lacy's post has been fairly predictable. "Classic" venture capitalists have publicly thanked her, while some next-gen VCs have taken her to task.

    What most of the responses missed, however, was that Lacy wasn't really taking sides. Instead, she was seeking to guide entrepreneurs through the disorienting haze of VC self-promotion.

    More importantly, taking sides right now is a fool's errand. As I wrote above, there is not yet enough data to support a conclusion about which type of model works best. Remember, it's only been four years since Andreessen Horowitz raised its first fund. Neither AngelList nor Dave McClure's 500 Startups showed up until 2010. For context, the Massachusetts state pension system only discloses returns for VC funds that are at least five years-old, because it finds earlier results to be statistically irrelevant (a common sentiment, due to the so-called J-curve).