作者：David Uberti
編譯：朱弘川
原文網址：http://www.cjr.org/innovations/in_at_least_one_respect.php
The social network’s response to Gizmodo’s story on its “trending” section was telling in this regard. On Tuesday morning, VP of Search Tom Stocky wrote a post denying the allegations of political bias, adding that Facebook’s team of news curators merely shepherd topics already identified by its algorithm. The supposed impartiality of this algorithm acts as a smokescreen to the other charges. Wrote Stocky:
基於這個前提，對於「熱門話題」所引起的風波，臉書搜尋部門副總Tom Stocky否認了有關外界對臉書政治立場偏頗的指控，強調工作團隊僅負責監看演算法的機制，彷彿藉由維護演算法的中立性來回應外界的指控。Stocky這麼說：
We have in place strict guidelines for our trending topic reviewers as they audit topics surfaced algorithmically: reviewers are required to accept topics that reflect real world events, and are instructed to disregard junk or duplicate topics, hoaxes, or subjects with insufficient sources…
審核「熱門話題」的工作人員遵照嚴格的指導方針，除了接受反映現實世界的文章外，還能無視高重覆性文章，或是來源不可信的主題……
We do not insert stories artificially into trending topics, and do not instruct our reviewers to do so. Our guidelines do permit reviewers to take steps to make topics more coherent….
我們不會隨意加入任何文章到熱門話題，也不會要求員工這麼做。我們只允許讓版面更符合一致性……
We will also keep looking into any questions about Trending Topics to ensure that people are matched with the stories that are predicted to be the most interesting to them, and to be sure that our methods are as neutral and effective as possible.
我們將研究「熱門話題」所引發的問題，確保人們能接觸到最符合自身興趣的文章，也儘可能使我們保持「中立和有效」的做法。
To be fair to Facebook, Gizmodo’s report is based on interviews with anonymous and disgruntled former employees. (The same outlet reported last week that Facebook wouldn’t even invite these contractors to company happy hours. The horror!)
的確，Gizmodo網站的報導，是來自於匿名和心生不滿的前員工專訪（其實上星期還傳出臉書不願意邀請這些雇員前來同樂的新聞，怪怪）。
But a few key words and phrases from Stocky’s statement stand out, as Zeynep Tufekci, a University of North Carolina professor who often writes about technology and society, highlighted on Twitter Tuesday. What makes a “hoax” or a source “insufficient”? What, exactly, is “neutral and effective”?
但我們仍可以從Stocky的聲明裡發現一些端倪，就像長期關注科技和社會議題的北卡大學（University of North Carolina）教授Tufekci在推特上說的：這場風波起因是什麼？是消息來源的「不可信」嗎？臉書所謂的「中立又有效」的做法，究竟指得是什麼？
————————————-
Zeynep Tufekci
I don’t mean "neutral" and "effective" don’t work for Facebook. It doesn’t work for human affairs. Welcome to your business model, Facebook.
我不認為「中立或有效」作法對臉書沒幫助，但它不適用在人際事務上。坦誠面對你的行銷模式吧，臉書！
Zeynep Tufekci
I am not saying why isn’t Facebook neutral, surfacing non-hoax news only from sufficient sources. I’m saying that will always be contested.
臉書的中立與否，不是我討論的重點，畢竟可信的新聞，還是得仰賴可靠的消息來源。我要強調的是爭議仍會不斷發生。
Zeynep Tufekci
The point isn’t the past was some unbiased paradise and the future is horrible, but we’ve entered a new regime with new concentrated powers.
我不認為過去沒發生過類似的問題，只是我們面臨的是一個權力日趨集中的社會。
Zeynep Tufekci
Facebook feed algorithm sold as "what you want" but it structures the experience. FB trending is sold as algorithm but is FB’s preferences
臉書的演算法像是提供了我們想要的，但也控制了消費者的體驗。臉書推銷的其實是它的喜好……。
——————————————-
Of course, such questions speak to a larger issue: While Facebook has become the public’s primary conduit for digital content, its business imperative is to maximize engagement, not objectivity. The algorithms designed to do that are human-made and therefore biased by nature. But we can only guess as to how they’re constructed.
當然，一個更大的問題等著我們：臉書，儼然已成為公眾獲取數位內容最普遍的管道，其商業考量必須是最大化的公眾參與，而不是客觀與否的問題。演算法出自於人類之手，也當然會有偏見，這是再自然也不過的事情。但我們只能猜測，它們是如何被創造的。
The obvious danger of the situation is that free societies have little knowledge of the systems funneling information into their newsfeeds. The sad irony is that the news organizations with the wherewithal to find out are the very same outfits that increasingly depend on Facebook for their survival.
自由社會對於這套系統如何過濾資訊仍一知半解，是目前最大的危機。悲哀的是，新聞機構苦思求生之計的同時，已擺脫不了對臉書的日漸依賴。
