Although technologists constantly praise computers as ‘universal machines’, and while it is true that they are remarkably flexible and powerful, this does not mean that we should use them for every kind of historical effort or production. We can truly begin to think about history in the second decade of the web, and to postulate new, more appropriate and ideally more enlightening forms of online history when we begin to look beyond the distribution of documents (including essays, articles, exhibits, news and messages) and consider instead the collection, interrelation and exploration of those documents. Much of the problem here is conceptual: from the beginning of the web historians have largely discussed nouns such as web pages and websites, rather than verbs such as searching, sorting, gathering and communicating.