prevent duplicated items

I don't see any reason that we should have duplicated items--this is like a database without a primary key. A suggestion for the simplest solution: provide a check box in general settings saying : do not allow duplicated item download. that's it.
when one tries (unknowningly) download a duplicated item, there should be a pop up warning "are you sure you want to download duplicated items"?
  • This has been a much-discussed topic on this forum.

    What is a duplicate? Identifying and management of potential duplicates not straightforward.

    Is an existing record without volume, issue, pagination metadata duplicated by a more recent version with that information? Is an existing record from a publisher that provides author names with last-name and initials duplicated by a record from a database that provides full author names? Etc.

    Your pop-up suggestion would be a boon for some but a manace for others. The value of a pop-up might differ for those downloading articles item-by-item vs downloading multiple items. For example, a PubMed import of many items can happen while other work is done. However, with a pop-up system the download must be monitored.

    These "duplicates" should be merged in a way that the merged product is more complete than either of the original records.
  • the pop up is turned on when a user enables it. when it's on, it does the monitoring. say you run the title against existing database before download and if it's the same, a pop up warning appears. I always know what i am downloading apparently. but I don't always know if the article is in already zotero or not. so when I see the pop up, I just stop download.
    pop up can be turned off. when a user does not enable it, you simply bypassing the monitoring system, which is exactly what we have now. if anyone downloads multiple, just turn it off? right?
  • I don't think the desirability of (optional) duplicate prevention is particularly controversial.

    There's an old add-on that still kind of works on Firefox, and a discussion thread on the development list that describes what would be required to make this broadly workable, which is not insignificant. It's obviously up for grabs for anyone who wants to continue development on this -- I'd be surprised if it's in the card in the near term from Zotero core devs.
Sign In or Register to comment.