Any leftover of the past can be considered a source. It might well be a document, and we often think of history as a textual discipline, based on the interpretation of written texts, but it might also be a building, a piece of art or an ephemeral object – a train ticket, say, or perhaps a pair of shoes. These are all 'sources' because they all provide us in different ways with information which can add to the sum of our knowledge of the past. Sources only become historical evidence, however, when they are interpreted by the historian to make sense of the past. The answers they provide will very much depend on the sorts of questions historians are asking. For example, a train ticket might be used to provide evidence of migration patterns or of the cost of living at a particular time, but also of broader cultural trends: for many years, for example, it was the practice to print a 'W' on a woman's ticket (this was when stations had women-only waiting rooms and trains had women-only carriages). As for a pair of shoes, it might provide the cultural historian with evidence of changing fashions and consumer tastes, or the social historian with evidence of class differences or production patterns. It all depends on what the historian wants to know. This is why it makes little sense to ask if something is 'good historical evidence', without knowing what evidence it's supposed to provide.