Thanks for commenting back!

Yeah, please be assured, you and other readers, that the CUCO project was just a starting point for this reflexion, and that I'm not pointing any fingers. As said in the article, my own database still shares some of the mentioned issues. I perfectly understand the owners authorization issue (even if I'm still amazed that's still an issue 24 years later to be honest).

I understood that CUCO is kind of a big Excel file, so that sounds file regarding access to raw data (aside anonymization). Regarding the "interpretation" issue, you mean that CUCO has a way to store different conclusions on a case? Sounds great, but is it a matter of including these in some free text comment, or in a structured way which would allow the user to search for explanationA OR explanationB? One may also wonder what a case filtering for "unidentified" cases should yield: case marked as unidentified/without any explanation or case that only some investigators marked as unidentified? Here again that's a search feature issue that apply on all databases I guess.

Aside this, I agree with all the psychological factors (and more!) you mention and, as you confessed, I probably share some myself. But my point was to list other potential factors that we could (more rationally/easily?) work on.

Actually the hidden conclusion of this article is referring to the previous one: standardizing, aggregating or merging and doomed to fail at some point, because new databases will come again and again (and because the original merged databases may still evolve after merge!). The only visable solution to me is to *map* existing data, thus allowing a *live* view/mapping of external data.