I’d somehow missed this— but CCC is creating a service to help identify researchers and their affiliations. This is a problem that is evergreen, in that I’ve seen various implementations going back a number of years, from biomedexperts (if I recall correctly, the founders went on to launch Uber research, and that then became Dimensions).

Publons was kind of a route to this— kind of.

There was some really good work from the network science community on probabilistic matching of researchers. Thomson Reuters introduced a researcher ID. We have ORCID. http://prophy.ai looks good. I’ve been impressed recently with https://allref.io and the approach they are taking to build LLM inference on top of a knowledge graph.

Here is a question— how do you make it easy to use a system like this for non-technical folks to use in a dumb workflow?

Perhaps you need to have submission vendors integrate it, and maybe integrations into Word and Google Docs. Or maybe you have an agent that sits on your computer with you when you are working through manuscripts and it kind of reads them for you, checking against the authenticity of the authors. It could schedule a meeting with you and talk you through what it has found, it could sit in on editorial discussions, much like we now have automatic tools that take our notes for us, but this one is more active in the conversation. That reminds me of the vision of Google Wave, which allowed the ability to create “bots” that interacted with your content as you created it.

There are some core problems then that we need to address:

academic research, data management, publishing technology, researcher identification, workflow optimization, innovation in publication, information technology.