Although there are increasing incentives and pressures for researchers to share code (even for projects that are not essentially computational), practices vary widely and standards are mostly non-existent. The practice of reviewing code then falls to researchers and research groups before publication.