I'm starting this discussion because I have seen many problems regarding duplications in this website, and it seems like even though it is possible for the RG team to solve these, I believe because of some minor technicalities, these problems are not still solved.
Two of the main problems that I've seen people encountering a lot are:
1) Duplications: This problem is very common and I have seen many questions regarding this. If you have two identical papers in the website, it is promised that RG will automatically merge the duplicate. But in many cases, it does not happen.
Many have problems when two of the uploaded papers already have a separate DOI. I have contacted the RG team regarding this problem, and they've told me that we cannot merge two identical papers when both of them have a DOI. For example, Imagine you upload your preprint, and you also see that the paper is added to RG via Arxiv. In this case, you have to remove one of them, losing reads, recommendations, citations and more..., only because the website is enable of merging two papers with two different DOI.
2) Preprints and papers do not merge. Now this is a very common problem. Again the only answer that RG team provide you with is to delete your preprint.
These problems just make people not upload their preprints in RG. I've seen many people aware of this problem, and just recommending not uploading anything in RG, because it will just create duplicates. I have seen people saying we have so many duplicates in RG that we don't care cleaning it.
And even if they want to clean these duplicates, most of the times the only thing that they can do is to delete one of versions of your paper. Now if you want to remove your uploaded file after a while (Just because it is not gonna merge) why bother uploading it in the first place?
It's a serious problem and I believe RG team should plan something serious on it. Preprints and the published papers should merge, in a single page containing all the informations regarding both in the same place. And two identical papers with different DOI should merge, at least visually, if it is not applicable technically.
That's it. If anyone has any comment, or have any more of these problems and want to add them, or do agree on these fact, please comment your problems here.
I hope RG team try fixing these problems.
Best regards