I think there is an option within the ORCID framework to include contact information? Or conversely one could infer it from a publication track. Find the most recent paper with that person on it and grab associated email address? Actually a rather interesting set of issues here... - Cameron Neylon

True that you can do traces through previous publication track if necessary, but where the information is limited, this can be an issue - for example forward tracking of unlisted/untracked data (where the researcher themselves has failed to curate - currently an issue). There are a couple of advantages to a deinstitutionalised email address though that I was thinking up after making the post: 1. more transparent peer review (because most people can't be bothered to look up the author if he/she is unknown to them ) - this only works if if this is the only address information that is used; and similarly 2. less bias against diy research for similar reasons (although it may be necessary to include a validation mechanism here to get the initial ID as a passport to publication). Disadvantages might exist where affiliation is a necessary part of the information required: eg Vested interests, same institution. - Anna Croft