Interesting that most are in favour of a centralized authority. I would have thought the likely bias would have been towards people who favoured a decentralized system. - Cameron Neylon

I think it is a highly complex issue for all sorts of reasons, and people just aren't aware of all the ramifications (including me, though I did not take the survey). This is one reason why nobody has implemented this system across the board, so far. (Further comments at blog post.) - Maxine

The number of people taking the survey was small and therefore probably highly selected. Keeping that in mind I would conclude: Digital Author Identifier is a good name, most people want to use this just for unique identification and use another tool for authentication, there should be one centralized tool, and this service should not be restricted to authors, but cover all areas of scholarly communication - Martin Fenner

I suspect that many people have difficulty envisaging how a decentralised system could have any authority. - Neil Saunders

@Neil I reckon you are right and also that CrossRef has good brand recognition as a trusted authority (the question as to who should run it was after the one about centralize/decentralize wasn't it?). Presumably that is also a contributing factor towards picking on DAI as a preferred name. I'd still prefer Digital Researcher Identifier or Digital Contributor Identifier, and agree that it should definitely not be restricted to authors. That would severely limit the potential benefits in my view. - Cameron Neylon

From everything I have seen in discussions (e.g. here on FriendFeed or in the interview with Geoffrey Bilder from CrossRef) and also the survey, I don't think anybody wants to limit the Digital Author Identifier to just paper authors. Aren't bloggers, commenters, database contributors, etc. also authors? - Martin Fenner

Possibly but I'm not sure that reviewers of papers or grants are. And that is where is see the biggest future benefits. Mostly I'm worried about the name focussing on authors reducing the imaginative application of new ideas. Just by making it "contributor" you raise the question "what type of contributions?" and I think that could in itself be a powerful motivator towards new ideas and new applications. - Cameron Neylon

It's a little like what happens when someone writes "she" in a sentence about a non-specific person. It still often has the power to make people think and question their assumptions. - Cameron Neylon

Whether to be centralized or distributed is absolutely critical, but should be purely a social question; not a technical one. If, rather than inventing yet another new identifier, you start instead with the notion of using URIs as the basis for contributor identity, things tend to fall in place for not foreclosing the possibility of a robust distributed system, while still benefiting a centralized one. - Bruce D'Arcus