This article outlines the findings of a new study that explores the portrayal of disability within a sample of the primary-age children's literature most readily available to UK schools. The kind of literature to which children are exposed is likely to influence their general perceptions of social life. How disability is handled by authors is therefore important from the standpoint of disability equality. Findings suggest that whilst there are some good examples of inclusive literature 'out there', discriminatory language and/or negative stereotypes about disability continue to be present in a range of more contemporary children's books. Clearly, more still needs to be done to ensure that schools and teachers are provided with information relating to the best examples of inclusion literature and efforts must continue to be made to inform authors, publishers and illustrators about how to approach the issue of disability.

Real-world missions require robots to detect objects in complex and changing environments. While deep learning methods for object detection are able to achieve a high level of performance, they can be unreliable when operating in environments that deviate from training conditions. However, by applying novelty detection techniques, we aim to build an architecture aware of when it cannot make reliable classifications, as well as identifying novel features/data. In this work, we have proposed and evaluated a system that assesses the competence of trained Convolutional Neural Networks (CNNs). This is achieved using three complementary introspection methods: (1) a Convolutional Variational Auto-Encoder (VAE), (2) a latent space Density-adjusted Distance Measure (DDM), and (3) a Spearman's Rank Correlation (SRC) based approach. Finally these approaches are combined through a weighted sum, with weightings derived by maximising the correct attribution of novelty in an adversarial 'meta-game'. Our experiments were conducted on real-world data from three datasets spread across two different domains: a planetary and an industrial setting. Results show that the proposed introspection methods are able to detect misclassifications and unknown classes indicative of novel features/data in both domains with up to 67% precision. Meanwhile classification results were either maintained or improved as a result.

For a large part of Western music history we are forced to interpret in the absence of signs. The appearance in the ninth century of a system of signs to represent music thus not only comes as something of a relief but also raises certain questions. How would the signs have been understood? How would something with no immediate history have been comprehended? Recent answers to such questions have placed notational signs within the context of oral history, positing a degree of continuity and interaction across oral and literate domains. Much insight has been gained through this awareness of oral issues, and it is not intended to challenge claims made in this area.

No abstract