互联网与文化simonweckert.com
Google Maps 恶作剧实验
Google Maps Hacks
2020 — 2022，它一次次回到首页
黑点是上首页（≥ 50 分）的那次投稿，空心点是没冲上去的投稿，点越大分数越高。
导读
这是 Simon Weckert 的 Google Maps Hacks 项目页面：他把许多智能手机装在手推车里在街上行走，让 Google Maps 把空旷街道误判为严重拥堵，从而展示众包地图数据可被操纵。项目以行为艺术的方式暴露导航应用依赖手机位置推断路况的脆弱性，评论则延伸到雪天绕行、火车数据干扰和算法推荐路线带来的现实影响。
为什么经久不衰：它用极低成本的行为艺术揭示日常依赖的导航数据如何被伪造，也让人重新思考算法给出的“最快路线”是否真的可信。
HN 评论精选
手机信令数据会把高速列车误判为车流，揭示众包交通数据的局限。
我曾在一家提供交通 API 服务的地图和导航公司工作。它利用匿名手机数据预测交通模式。我听说过一个故事：高峰时段，高速公路上每隔五分钟拥堵就会神奇消失又出现。后来发现，高速公路车道之间有一条铁轨，上面坐满高速通勤者，他们的数据抵消了静止汽车通勤者的数据。
nudgeee查看英文原文
I used to work for a mapping & navigation company that offered a traffic API service. It worked by using anonymized cell phone data to predict traffic patterns. I once heard a story that during peak hour, every five minutes or so the jams on a highway would magically disappear then reappear. After some head scratching, turns out there is a train track inbetween the lanes of the highway full of high speed commuters that would cancel out the stationary car commuters.
拥堵绕行建议可能把人引向更危险、路况更差的道路。
类似问题会发生在全国或山区异常大雪时。主干道变红，因为车开得很慢——路上有 50 厘米雪。于是地图把你重定向到次要道路，那里没人，雪还更多。
wh1t3n01s3查看英文原文
A similar problem happens during unusual big snowfalls in the country or in the mountains. The primary roads turn red because the cars are going really slow cz there is 50cm of snow. So maps redirects you to secondary roads, where there is nobody and even more snow.
导航产品仍缺少用户控制、多模式比较和多点路线优化。
导航应用大多停留在“足够好”的平庸状态。除了别人指出的“红但无交通”问题，我发现逐向导航在我很熟悉的地方开车时非常烦人，而且没法说“这片区域我知道怎么走”或“暂停语音导航 10 分钟”。此外：我不能在同一张地图上比较多种交通方式，比如开车、步行和公交；没法优化减少左转，尤其是转入繁忙街道；多目的地路线优化也不可用。比如我要去商场、杂货店和银行，按什么顺序和路线走总时间最短。
AlexandrB查看英文原文
Navigation apps have mostly settled into a comfortable state of "good enough" mediocrity. In addition to the "red but no traffic" problems highlighted by others I find turn by turn directions are incredibly annoying while driving in areas I'm intimately familiar with and there's no way to say "I know what I'm doing when I'm in this area" or "pause giving me voice directions for 10 minutes". Additionally: * I can't compare multiple modes of transportation on the same map. E.g. driving vs. walking vs. transit. * There's no way to optimize for minimizing left turns, especially onto busy streets. * Multi-destination route optimization is not available. E.g. I need to go to the mall, the grocery store, and the bank, what's the sequence of destinations and route that minimizes travel time.[1] Edit: [1] I realize this is describing the travelling salesman problem, but for small (<=4) n it should not be too difficult while still being useful in practice.
算法优化最快路线可能把用户引经负面观感街区，影响地产展示。
在假交通拥堵项目首次实施前后，我朋友在做一个即将挂牌出售的公寓开发项目。他们计划办开放参观。房产本身很好并成功交付，但有个问题。房产靠近本地高速公路很方便，又不至于受噪音、交通和尾气的主要影响。Google Maps 会可靠地带人上这条高速并从同一个出口下。从出口，本地人会走一条穿过住宅区的安静好走的路。但 Maps 会带你走另一条路，绕到一条侧街，那里有公共住房项目，常出现露天毒品市场，也频繁发生枪击死亡。推测 Maps 这样指路是因为另一条直接路线是大多数人走的，所以这个绕行路线车少、略快。这条街离公寓项目有些距离，但仍会让人对社区产生某种印象，开发商觉得不公平。我把链接的文章发给他们希望试试，但据我所知他们从未试过。另外，我也不确定一个人推着装满手机的车走在这条街上会受到什么欢迎。
eckmLJE查看英文原文
Around the time this fake traffic jam project was first performed, a friend of mine was working on a condo development that was just about ready to list for sale. They planned an open house. The property itself was very nice and successfully delivered. However, there was a problem. The property was very convenient to a local highway without being so close that it experienced noise/traffic/primary exposure to exhaust, etc. Google Maps would reliably take people on this highway and to the same exit. From that exit, locals would go one way that was a nice, quiet drive through some residential parts of town. But Maps would take you a different way, a detour down a side street with a public housing project that often presented an open air drug market, and was the site of not infrequent shooting deaths. Presumably Maps sent you that way because the other direct way was the way most people took, and so this detour had less traffic and was marginally faster. This street was some distance from the condo project, but it would still give people a certain impression of the neighborhood that the developer felt was unfair. I sent them the linked writeup hoping they would try it, but they never did as far as I know. Also, I'm not sure how well-received someone walking down this street with a bunch of phones in a wagon would be.
历次讨论
同类还值得读
如今电脑上几乎所有操作在感知上都比 1983 年更慢（2017）
Almost everything on computers is perceptually slower than it was in 1983 (2017)
2017 年的推文串认为，尽管计算机硬件比 1983 年快得多，日常操作的感知速度却常常更慢：启动、界面响应和网络应用被层层抽象、动画、后台检查与网络依赖拖累。
互联网与文化上首页 5 次 · 2017–2020twitter.com网站肥胖危机
The Website Obesity Crisis
Maciej Cegłowski（idlewords）的经典演讲文稿《网站肥胖症危机》，批评现代网页越来越臃肿：动辄数 MB 的 JavaScript、追踪脚本和广告，把本可以轻…
互联网与文化上首页 5 次 · 2016–2021idlewords.com编辑精选亚马逊那本标价 23,698,655.93 美元的苍蝇书
Amazon’s $23,698,655.93 book about flies
博文讲的是亚马逊上一本关于果蝇的二手书被标到 23,698,655.93 美元。
互联网与文化上首页 5 次 · 2011–2022michaeleisen.org浏览器 User-Agent 字符串的历史
History of the browser user-agent string (2008)
2008 年的经典文章用戏谑的“创世纪”口吻，讲述浏览器 User-Agent 字符串如何从简单的标识演变成层层叠加的伪装：浏览器为了绕过按 UA 分发的网站检测，纷纷把自己伪装…
互联网与文化上首页 5 次 · 2012–2022webaim.org编辑精选最后的问题——艾萨克·阿西莫夫
The Last Question - Isaac Asimov
阿西莫夫最著名的短篇科幻小说《最后的问题》的在线全文。
互联网与文化上首页 5 次 · 2009–2019multivax.com编辑精选