Network positioning systems provide an important service to large-scale P2P systems, potentially enabling c lients to achieve higher performance, reduce cross-ISP traffic and improve the robustness of the system to failures. Because tr aces representative of this environment are generally unavaila ble, and thereis noplatform suitedfor experimentationat the appro priate scale, network positioning systems have been commonly impl ementedandevaluatedinsimulationandonresearchtestbeds .The performance of network positioningremainsan open questio nfor large deployments at the edges of the network. This paper evaluates how four key classes of network positioning systems fare when deployed at scale and measured in P2P systems where they are used. Using 2 billion network measurements gathered from more than 43,000 IP addresses probing over 8 million other IPs worldwide, we show that network positioning exhibits noticeably worse performanc e than previously reported in studies conducted on research testb eds. To explain this result, we identify several key properties o f this environment that call into question fundamental assumptio ns driving network positioning research.