sure, but what I'm actually arguing for (in this project) - as a way to keep admin work low - is to do a similarity search among documents at time of document import, then the admin can choose out of the 5 most similar documents if one is the same, and if not that they can always say no I know this document is the same as document ID X, or just import it as a new document.
Otherwise the admins are going to need to do work about figuring out what version they are dealing with and give the document the right 'name' or right id. either way I'd prefer automating that for them as much as possible.
but my coworker, who already implemented some stuff before I came on to the project, expects that the admins will put documents into a watched folder on their system, and documents will be named as follows releasedate_correction_titleofdocument.pdf - titleofdocument could be the system document id, but I'm assuming that requires way too much admin work and has chances of being faulty - but it's only an assumption. So I'm hoping others can tell me - yeah I think that is way too problematic, or Bryan you're overthinking it, it won't really be a problem.
