虽然没什么卵用，但还是要写，不然怎么毕业。
看资料的时候，从 Information-Centric Networking: Seeing the Forest for the Trees 跳到了 Finding a needle in Haystack: Facebook’s photo storage
译文（http://www.importnew.com/3292.html）
做科研，导师也不懂，糊里糊涂过了一年。沮丧太多，也是自己不努力。
按理说，NDN 也是大牛提出来的，缓存对访问的提升效果有巨大的局限性这事儿，大牛不应该不知道啊。哎……某些科研真的就是大牛挖坑，渣渣灌水吧。
Anyway，跳转阅读到了经典论文，还是开心的。粘几句 Information-Centric Networking: Seeing the Forest for the Trees里形象生动说出我新生的神吐槽吧（自己真的是弱爆了，想吐槽都说不明白）。
Because there is no common framework, the focus is often on low-level mechanisms. As a result, many of the papers accentuate the differences between their design and others2 , leaving it as an exercise for the reader to construct the ICN forest out of these individually-proposed trees.
Changing the overall network architecture in order to tame the exponentially growing world of content with the logarithmic sword of caching seems a classical example of taking a knife to a gunfight: it may make for a great story, but it won’t end well.
So we end this paper by posing two fundamental questions: what benefits do we think ICN designs offer, and are ICN designs the best way to achieve those benefits? The field has focused on the former (though somewhat uncritically, as suggested by our discussion of caching), but we think addressing the latter is equally important.