Strategic rebels: a spatial econometric approach to rebel fighting durations in civil wars

ABSTRACT

Recent research on multi-actor civil wars highlights that rebel organizations condition their conflict behavior on that of other rebel organizations, with competition and free-riding constituting the core theoretical mechanisms. We provide a new actor-centric approach to explicitly model strategic interdependence in multi-actor civil wars. We argue that rebel organizations have incentives to remain mobilized until the end of a conflict to maintain their power to negotiate, power to spoil, power to enforce, and power to protect. This induces strategic complements that dominate duration dynamics in multi-actor conflicts. Based on a network game-theoretic model, we derive a spatial econometric framework that allows for a direct test of strategic interdependence. We find that the estimated duration interdependence is positive but partially offset in secessionist conflicts where the public goods nature of the incompatibility also induces strategic substitution effects.     

Accountability in Public Management Networks: An Analysis of Cost Monitoring in Chicago Municipal Networks

Public management networks (PMNs) continue to proliferate as governance becomes more complex, and so does the need to hold them accountable. Agency and collaboration theories provide a solid foundation, but additional constructs are necessary to adapt to this dynamic era. Building from Bovens et al. (2008), the authors offer an ecological framework of third-party accountability that incorporates a number of prominent accountability theories under the same construct. Employing a mixed-methods research design that examines eight local PMNs in the Chicago region, the study finds that an exemplar accountability mechanism—additional cost monitoring—occurs more often in situations predicted by the framework. Additionally, qualitative interviews of network participants suggest that cost structures can differ by network type, which is verified quantitatively.     

“互联网条款”对于新类型网络服务的适用问题——从“通知删除”到“通知加采取必要措施”

"通知删除"规则率先确立于《信息网络传播权保护条例》,《侵权责任法》"互联网条款"将其扩展为"通知加采取必要措施",并使其广泛适用于一般网络侵权领域。《电子商务法》知识产权保护条款又有相应的规定。随着互联网技术和商业模式迅速发展,新类型网络服务层出不穷,在适用"通知删除"等规则中不断产生新问题和新争议。除另有特别规定外,新类型网络服务应当适用《侵权责任法》第36条规定的"通知加采取必要措施",其中"必要措施"的界定具有开放性和灵活性,"必要措施"的妥当性取决于与特定网络服务的匹配性,可以根据网络服务提供者的具体情形,以利益衡量的方式确定相应的具体措施。个案中应当根据新类型网络服务的特性以及产业发展需求等情况,合理权衡特定网络服务与"必要措施"的适应性,妥善确定应当采取的"必要措施"。     

网络信息安全犯罪的定量评价困境和突围路径——大数据背景下网络信息量化标准的反思和重构

刑法对网络安全的保护立场，正由网络运行安全转向网络信息安全。刑事立法更新不断增设网络信息安全专属罪名的同时，刑事司法的定量评价体系却未能一体构建，成为网络安全刑法保护的薄弱环节。当前网络信息安全犯罪的定量评价体系缺乏明确的逻辑主线，同时适用错位的“传统法益侵害程度”量化标准和模糊的“数据规模”新型量化标准，无法满足网络信息安全专属罪名司法适用的准确性和统一性需求，导致了司法适用的困境。消弭规范和技术之间的差异，以“信息规模”为定量评价核心的基础上，建构信息价值的分层评价模型，并将“组信息”作为基础的数据规模计量单位，是大数据时代背景下，突破网络信息安全犯罪司法定量评价困境的合理路径。     

电信网络诈骗犯罪侦查管辖制度的反思与调整

为了解决电信网络诈骗犯罪与传统侦查管辖之间的结构性冲突,刑事司法在横向上发展出“多地共管”“人案关联”的“大侦查管辖”格局,在纵向上建立起侦查管辖决定审判管辖(包括起诉管辖)的“侦查管辖中心”模式。但大侦查管辖格局导致不同地域公安机关的横向管辖冲突,侦查管辖中心模式又引发公检法机关的纵向管辖争议。考虑到公安机关条块分割的侦查体制与电信网络诈骗犯罪打击治理的现实困境,大侦查管辖仍是刑事司法当前相对合理的选择,但应优化地域管辖与指定管辖制度,减少不同地方公安机关之间的横向管辖冲突;同时,刑事司法需要把“以侦查管辖为中心”调整为“以侦查管辖为导向”,承认侦查管辖的指引功能,但应完善公检法机关的纵向管辖争议协调机制。     

网络问政的政府回应逻辑：公开承诺压力与选择性回应——基于66个政府网站领导信箱的实验设计

立足我国制度环境与行政体制，建构了网络问政领导信箱回应运行理论框架，提出影响回应绩效五个影响因素，包括行政系统区域特点与领导者特质两个内部因素，外部因素包括政治压力、舆论压力及诉求压力三个维度。采用德尔菲法建构了领导信箱回应绩效指标体系，并采用实验设计搜集66个领导信箱网络问政回应的有效样本。实证结果显示：政治压力中的公开承诺程度对回应绩效有正向效应，诉求压力中的诉求类别对回应绩效有显著影响，政府回应存在“选择性回应”问题。为了进一步提升政府网站领导信箱的回应绩效，可以完善内部回应流程、公开对外服务承诺与建立自下而上回应性问责机制。     

What Makes a Key Player in Interorganizational Social Policy Networks? The Case of Uruguay

ABSTRACT

Networks of organizations involved in public policy implementation require strong interaction, concerted action and high degrees of collaboration to be effective. However, little is known about how different types of organizations involved in implementation of multi-sectoral social policies interact in these networks. In this article the relationship between organizational characteristics and network position is explored, as well as how the intensity of collaborations can also determine organizations’ involvement in networks. The nature of funding (public/private) and the remit of activity of organizations are found to determine their influence and importance in social policy networks. Furthermore, the network position of the organizations also depends on the level of intensity of their interactions. These results can guide public administrators when developing and promoting networks to involve a particular type of actor and also policymakers as to which types of ties are more aligned with the implementation of a particular policy.     

The Cosponsorship Patterns of Reserved Seat Legislators

This study analyzes the bill cosponsorship behavior of reserved seat legislators in the Pakistan National Assembly, where special legislative seats are reserved for women and religious minorities. It offers a critical case because two theories on reserved seat legislators—the theory of electoral incentives and the theory of institutional weakness—lead to opposite predictions about the extent to which such legislators will engage in bill cosponsorship, with the former (latter) predicting the negative (positive) effect of reserved seats. I find that women and minorities in reserved seats initiate a greater number of bills and cosponsor bills with a larger number of peers than nonreserved male legislators. Using network analysis, I also show that reserved legislators tend to play a role in bridging different partisan blocs of legislators within the cosponsorship network. These empirical patterns provide support to the theory of institutional weakness, rather than to the theory of electoral incentives.     

构建社会主义和谐社会是一项复杂的社会系统工程

社会主义和谐社会是一个复杂的动态网络系统,而构建社会主义和谐社会则是一项复杂的社会系统工程。这项工程在运行机制和功能上都有其特殊的表现。不仅如此,它还内蕴着规律的复杂动态网络系统的交互作用。正确把握这些规律及其交互作用,对于推进社会主义和谐社会的建设至关重要。     

基于角膜图像的人体死亡时间推断模型初探

目的尸体角膜随死后时间延长发生的形态学变化是规律性较好的指标,常用来判断死亡时间(postmortem interval,PMI)。本文尝试用机器视觉代替人的肉眼主观判断,收集尸体样本以建立通过人体角膜图像推断PMI的模型。方法收集实际案例建立包含505例人体死后角膜图像的数据库,PMI范围为0.24h(约死后14min)至492h(约死后20.5d),大致分为三类(依次为:0~<6h、6~<20h、20h及以上)或二类(0~<15h、15h及以上);使用由华盛顿大学陈天奇博士提出的Xgboost模型分别进行二分类与三分类分析;使用多种卷积神经网络模型分别进行分类和回归学习,并通过比较最终选择了由微软研究院提出的ResNet模型进行分析。结果Xgboost在三分类时预测准确率依次为71.8%、40.7%、65.7%,二分类时为90%、48.5%。ResNet分类模型中,精准率、召回率在三分类时分别依次为:81%、75%,30%、50%,61%、71%,二分类时为:70%、92%,76%、38%。ResNet回归模型中,比较整个模型的预测结果,0~6h内的预测值与真实值较为接近,均值误差为0.5616,均方误差为0.5873,6h之后开始出现较大误差。结论分类和回归模型都在0~6h之内得到了很好的结果,说明在此时间段内,角膜图像噪声较低,可预测性强。