The traditional concept of space in geography is based on the notion of distance. Where there is a spatial analysis, there is a distance measurement. However, the precondition for effective distance-based space is that the geographical systems have characteristic scales. For a scale-free geographical system, the spatial structure cannot be validly described with pure distance, and thus the distance-based space is ineffective for geographical modelling. In the real geographical world, scale-free patterns and processes are everywhere. We need new notion of geographical space. Using the ideas from fractals and scaling relations, I propose a dimension-based concept of space for scale-free geographical analysis. If a geographical phenomenon bears characteristic scales, we can model it using distance measurement; if a geographical phenomenon has no characteristic scale, we will describe it using fractal dimension, which is based on the scaling relations between distance variable and the corresponding measurements. In short, geographical space fall into two types: scaleful space and scale-free space. This study shows a new way of spatial modeling and quantitative analyses for the geographical systems without characteristic scale.
On the relation between structural diversity and geographical distance among languages: observations and computer simulations
最近出现的更大的类型学数据库，如Haspelmath et al.（2005），使语言学界更接近于拥有一个坚实的经验基础，来对史前迁移、深层谱系关系和区域语言互动模式做出实际的主张。尽管如此，仍然需要更多的数据，而且最重要的是需要解决一些方法问题。这些问题，正是本文研究的重点。这些问题将从两个角度来处理。第一种观点是经验主义的，观察主要来自Haspelmath等人（2005）的数据分析。第二个角度是计算的，其中模拟被用来测试不同的参数对结构语言多样性的发展的影响。结果表明，确实有一些希望，我们可能会得出新的经验见解语言史前的类型学数据。然而，我们并没有在本文中提出任何具体的经验主张，而是将我们的努力集中在方法论上
Unified framework for measuring segregation resolves how social and geographical space jointly shape connections
Imperial College London（帝国理工学院）；Massachusetts Institute of Technology（麻省理工学院）；Harvard T.H. Chan School of Public Health（哈佛大学陈曾熙公共卫生学院）；Indian Institute of Technology Delhi（印度德里理工学院）
The effects of spatial constraints on the evolution of weighted complex networks
出于对航空运输系统的实证分析，我们定义了一个网络模型，其中包括地理属性以及拓扑和权重（交通）属性。地理属性的引入是通过将网络约束在真实空间中来实现的。有趣的是，几何特征的引入引起了权重、连接模式和顶点的实际空间距离之间的非平凡相关性。该模型还恢复了介度关联函数中出现的异常波动，如Guimerà和Amaral首次观察到的[Eur. J. B { \bf 38}，381（2004）]。所呈现的结果表明，权重动态和空间约束之间的相互作用是理解真实世界加权网络形成的关键因素。
