Concomitant with the publication of this Special Issue of Neuroinformatics, a substantially updated version of the DIADEM web site has been released at http://diademchall enge.org. This web site was originally designed to host the challenge for automating the digital reconstruction of axonal and dendritic morphology (hence the DIADEM acronym). This post-competition version features additional content for continued use as the access point for DIADEM-related material. From the very beginning, one of the spirits of DIADEM has been to share data and resources with the neuroscience research community at large. The resources available from or linked to the DIADEM website constitute a substantial scientific legacy of the 2009/2010 competition. The new content includes finalist algorithms, image stack data, gold standard reconstructions, an updated DIADEM metric, and a retrospective on the competition in text and images. The website continuing intent is to facilitate development of automated reconstruction algorithms. The DIADEM Data Sets include image stacks, manually reconstructed digital tracings (the “gold standards”), and metadata. The six extensively curated, diverse data sets can be used to train, test and aid in the development or tuning of existing automated reconstruction algorithms. The previously posted image stacks (training and qualifier sets for the DIADEM competition) are still available, and are augmented with the addition of the Final Round image stacks. The Final Round sets include a previously unreleased type of data, a visual cortical pyramidal cell, used as a surprise set in the competition. The DIADEM metric (including source code) also has a new release with a wider set of user options and extended documentation. These options provide flexibility for use on data beyond the DIADEM data sets and for alternative scoring schemes. Integration of the metric with automated reconstruction development has been made easier with more output options and methods to make results accessible to an interfacing program. These and other changes make the DIADEM metric viable in both algorithm development and evaluation. Since it was designed as a machine-derived surrogate to humans for quantifying the differences between two reconstructions of the same neuronal structure, the DIADEM metric can be used as a reliable tool to evaluate reconstruction quality of either newly developed or expansions of existing automated reconstruction algorithms. Most importantly, the DIADEM finalists made their algorithms freely available on the website. These include code, executables, and documentation, in addition to links