Information-centric networks (ICNs) have emerged as a prominent future Internet architecture. They work on the principle that, in today???s society, a network should be optimised towards the delivery of information, rather than the transit of messages between pre-determined end hosts. At the same time, another prominent research direction has been the delay-tolerant networking initiative, which argues that networks must explicitly support the tolerance of high delays and disruptions in end-toend paths. We believe that this observation is pervasive across a number of applications and network technologies and should therefore be strongly considered in any future ICN designs. This paper explores the potential of combining these concepts into an information-centric delay-tolerant network (ICDTN). Through a qualitative and quantitative investigation, we present arguments in favour of the design, as well as important research challenges.