Featured Graphic. Mapping the Geoweb: A Geography of Twitter
Featured Graphic. Mapping the Geoweb: A Geography of Twitter
复制标题
DOI：
10.1068/a45349
复制
发表时间：
2013-01
期刊：
影响因子：
--
通讯作者：
Mark Graham;Monica Stephens;Scott A. Hale
中科院分区：
文献类型：
--
作者：
Mark Graham;Monica Stephens;Scott A. Hale
在线社交媒体已经成为许多互联网用户日常生活中不可或缺的一部分，目前全球有数亿人在使用这些服务。伴随着使用量的增长，公司、政府机构和学者希望研究和绘制人们使用Twitter、Flickr和Facebook等服务留下的数据痕迹。在线用户留下的数据阴影和信息踪迹揭示了社会、经济和政治过程和做法。尤其是Twitter，由于其相对开放的网络，研究人员可以访问通过该平台发布的几乎任何信息，因此Twitter被反复用作社交数据的储存库。然而，尽管有许多研究(学术界内外)借鉴了Twitter的数据，但很少有学者致力于研究Twitter的地理位置。作为第一步，我们决定收集2012年3月5日至3月13日期间发送的所有地理参考推文。必须指出的是，地理引用的推文占所有推文的不到1%，而且人们在哪里以及如何对其内容进行地理引用可能存在显著的地理偏差(Graham等人，2012a)。然后，我们从该数据集中随机抽取了20%的样本：给出了大约450万条推文，我们在国家层面上加入了这些推文。我们的图表将这些数据显示为空间感知的树状地图。每个区块的大小代表来自该国家的推文数量，阴影显示了地理编码的推文数量占该国家互联网人口的比例(即，它让我们了解互联网用户创建地理编码的Twitter内容的可能性)。该图表揭示了内容地理位置上的大量不平等。然而，尽管许多其他在线平台的特点是不同的数字劳动分工，其中全球北方是主要的内容生产者和主体(Graham等人，2011年)，但Twitter展示的是明显不同的地理位置。通过Twitter产生的信息最多的六个国家是：(1)美国、(2)巴西、(3)印度尼西亚、(4)英国、(5)墨西哥和(6)马来西亚。有趣的是，该名单上只有两个国家位于全球北部，是传统的编码知识生产中心。通过绘制推文在世界各地的分布图，很明显，Twitter允许比大多数其他平台和媒体更广泛的参与。换句话说，它可能会允许信息生产和共享的“民主化”，因为它的进入门槛很低，而且对移动设备的适应性很强。同样，信息传播的障碍，如审查制度，也可以从发自中国(全球网民人数最多的国家)的一小部分推文中看出。然而，为了更好地了解该平台上的内容地理位置，更多的研究无疑是必要的。我们在这篇文章中的样本是有限的，更重要的是，我们只能可视化通过该平台的地理参考推文的数量。随着虚拟层面和地点的扩大对日常生活越来越重要(Dodge and Kitchin 2005；Graham et al，2012b)，理解信息的地理位置将变得更加重要。这张地图提供了一个起点。
Online social media has become an integral part of daily life for many Internet users and there are now hundreds of millions utilising these services around the world. Concomitant with this growth of usage is a desire by companies, government agencies, and academics to study and map the data trails left by people using services like Twitter, Flickr, and Facebook. The data shadows and information trails left by users online reveal social, economic, and political processes and practices. Twitter, in particular, is repeatedly used as a repository of social data because of its relatively open network that allows researchers access to almost any information published through the platform. Yet, despite the many studies (both inside and outside of academia) that draw on data from Twitter, there is little scholarship devoted to the geography of Twitter. As a fi rst step, we decided to collect all georeferenced tweets sent between 5 March and 13 March 2012. It is important to point out that georeferenced tweets comprise fewer than 1% of all tweets and it is possible that signifi cant geographic biases exist in where and how people georeference their content (Graham et al, 2012a). We then took a random 20% sample of that dataset: giving us approximately 4.5 million tweets that we spatially joined at the country level. Our graphic illustrates these data as a spatially aware treemap. The size of each block represents the number of tweets emanating from that country and the shading shows the number of geocoded tweets as a proportion of that country’s Internet population (ie, it gives us a sense of how likely Internet users are to create geocoded Twitter content). The graphic reveals a large amount of inequality in the geography of content. However, while many other online platforms are characterised by distinct digital divisions of labour in which the Global North is a predominant producer and subject of content (Graham et al, 2011), Twitter displays signifi cantly different geographies. The six largest countries in terms of information production through Twitter are: (1) the United States, (2) Brazil, (3) Indonesia, (4) the United Kingdom, (5) Mexico, and (6) Malaysia. It is interesting to note that only two of the countries on that list are in the Global North and are traditional hubs of the production of codifi ed knowledge. By mapping the distribution of tweets in the world it becomes apparent that Twitter is allowing for broader participation than is possible in most other platforms and media. In other words, it might be allowing for a ‘democratisation’ of information production and sharing because of its low barriers to entry and adaptability to mobile devices. Similarly barriers to the dissemination of information, such as censorship, are also visible through the small proportion of tweets originating in China (home to the largest population of Internet users in the world). However, more research is undoubtedly necessary to better understand the geography of content on the platform. Our sample in this post is limited, and even more importantly allows us to visualise only the quantity of georeferenced tweets that pass through the platform. As virtual layers and augmentations of place increasingly matter to everyday life (Dodge and Kitchin 2005; Graham et al, 2012b), it will become more important to understand the geographies of information. This map offers a starting point.