Can simplicity help?

Jean-Yves Blaise (1) , Iwona Dudek (1)

1 MAP - Modèles et simulation pour l'architecture, l'urbanisme et le paysage

Analysts are today given (or develop) a wide range of powerful visual reasoning tools, of leading-edge platforms where services like high interactivity, high computing capacity, structured knowledge, processing of complex reasoning tasks on massive data, are duly implemented. Classic XVIIIth or XIXth century dataviz solutions may then appear as respectable, but deprecated - far too simple to match today's challenges. In this paper we try to show that the very simplicity of these solutions can still be of help in problem-solving situations. We first revisit some great classics in order to uncover patterns and exceptions inside a data set consisting of incomplete historical evidence on groups of stalls that used to be located on the market square in Cracow. We show that, once reinterpreted and re-implemented, classic visualisations like Charles De Fourcroy's tableau poléométrique or Munehisa Homma's candlestick chart do help the analyst to re-read the data, and to augment his level of knowledge. We then introduce an online, free to use implementation of these classics that we expect to weigh through a sort-of crowdsourcing approach to which extent they can act as relevant and generic visual formalisms. Finally, we underline the potential contribution of this research in terms of methodology, and discuss in what simplicity proves here helpful.

Mots clés: Historical sciences; Methodology; Spatio-temporal data; Information visualization; Visual formalisms

Domaines: Sciences de l'information et de la communication; Méthodes et statistiques; Informatique; Education; Histoire; Architecture, aménagement de l'espace

Jean-Yves Blaise, Iwona Dudek. Can simplicity help?. ACM International Conference Proceedings Series, 2014, i-KNOW 2014 - Knowledge Visualization, paper 17 (8p). ⟨10.1145/2637748.2638414⟩. ⟨halshs-01074206⟩