The Service-Oriented Computing (SOC) paradigm envisions a large, open and dynamic computing environment where anyone can publish his own services, for instance on the Web by using Web services (WS) as fundamental building blocks. In order to realize this vision, several challenges must still be addressed. In particular, consensus is growing that this “service revolution” won’t eventuate until we resolve trust-related issues. Indeed, the intrinsic openness and dynamism of the SOC vision makes crucial to locate useful services and recognize them as trustworthy. In this paper we critically review the field of trust-based Web service selection, proposing a structured classification of the current approaches and highlighting their main weaknesses. According to this analysis, we claim that a “soft” notion of trust lies behind such weaknesses. Finally, we address this limitation roughing out a “hard” trust based framework for WS publishing, selection and monitoring, namely Trust-By-Contract. We claim that the Trust-By-Contract rationale can be considered a first step toward trustworthy Web services.