Yes it's a real problem. Manually updating every little detail added to any entry in different groups or libraries seriously undercuts the usefulness of zotero which s otherwise marvellous. How on earth can we remember if there is a duplicate to update? I can't except for the most commonly known items.
I don’t have anything to help you there other than to recommend that you check your data quality and make fixes on items before copying them between libraries and to try to adjust how you think about items in different libraries to not think of them as reflecting the same item, but rather different items that have no necessary connection with each other.
"to not think of them as reflecting the same item, but rather different items that have no necessary connection with each other." Unfortunately research is not link that - cross referencing between bibliogs is standard. Categories are rarely standalone.
We track items dragged between libraries and hope to offer some options for keeping the different items in sync, but unfortunately it's a pretty complex problem. If they were the same items, anyone in any library could make a change that applied to all libraries that had a given item. It'd be much harder to figure out and specify whether notes and tags on an item were shared to a given library. Annotations made to a file would affect all libraries. Etc.
So instead they're separate items, which has the drawbacks you mention but avoids a slew of other problems. We've talked a bit in the past about ways we could enable this sort of cross-library updating, though I don't have the discussions handy. If you have thoughts on how you'd like this to work for your particular workflow, feel free to share.
" If they were the same items, anyone in any library could make a change that applied to all libraries that had a given item."
But that;s EXACTLY what's needed. I enter a book or article in Bibliog A and in Bibliog B. I then find I have misspelled the publisher. Or I didn't know the pagination and later discover it. Or I add notes once I get to read it. That can also change its Tags. So evidently I need those same corrections and additions to show up wherever the item is listed.
"If they were the same items, anyone in any library could make a change that applied to all libraries that had a given item." If they are the SAME items then they need the same data on date, author, title, publisher, source, pagination, standardised across instances, as above.
If a different version is needed - maybe making private special notes for self, or for team colleagues, that could be done by permissions. If that is too complex then a string warning in the Help guidance,maybe even a popup dialogue, would prompt caution 'You are making changes to this entry which will also change it in other libraries. Proceed?'
You have to remember that yours is only one of many possible workflows people use with Zotero groups. For example, I use a separate group library with each different systematic literature review project I conduct. Each library is associated with its own tags, notes, attachments, and related items, as well as some unique metadata (e.g., I add the "annote" variable to Extra to customize citations for each review, and the "Library Catalog" field is important for tracking the source of items). With this use of groups, I absolutely do not want changes to any of those features of items made in one library to impact the versions of the items in another library. While correcting spelling errors or adding DOIs or page numbers across libraries would be nice, implementing a model where some, but not all, item changes are propagated across libraries (and accomodating different users' needs in this regard) is a complex task.
Dan is looking for information on different users' needs to develop some form of solution to meet both of our extreme set of preferences and everything in between.
I am grateful for your attention as this issue has forced me to use zotero in avery limited way.That's frustrating as I could do so much more.
In principle some users need cross referenced updates, and some need standalone updates. Why not make both available? So when creating a library/group I can opt to link it with other libraries/ groups to they update each other, or keep the current default standalone entries.
"While correcting spelling errors or adding DOIs or page numbers across libraries would be nice" This isn't just a pleasant bonus! The status of a database rests on a) its structural design, and b) accurate input. If corrections are not implemented then it's a garbage in garbage out scenario and the whole project is weakened. Mistakes, slips and omissions will be frequent because surely the whole point is that manual memory cannot manage 1,000s of entries.
I could easily make anything up to ten bibliogs on zotero. Their content would duplicate most core items, but each one would have specialist entries, and some core entries would not appear in some bibliogs. This would create a much finer grained service than the current crude categorisation I can make with one bibliog and its tags.
So I choose to run just one bibliog because the headache and time gobbling of manual updating would be so destructive. I make this a public service. The result is that I do not have the convenience of a bibliog that fits my own work.
Hello, just checking whether this issue has been resolved? I very much agree with morgain13. It is an extremely strong drawback that items are not updated across libraries, so strong that it almost completely breaks the whole use of Zotero for shared libraries.
It's planned, but it's probably not among the next couple of features to land, no. There are always lots of different features that different users consider essential.
Zotero devs have been very reluctant to give ETAs for any specific new feature unless it's basically done; sometimes they're surprisingly quick, sometimes they take several years longer than expected. So I think they'd certainly hope for within the next 2-3 years, but no one will make promises.
Upgrade Storage
How on earth can we remember if there is a duplicate to update? I can't except for the most commonly known items.
Unfortunately research is not link that - cross referencing between bibliogs is standard. Categories are rarely standalone.
So instead they're separate items, which has the drawbacks you mention but avoids a slew of other problems. We've talked a bit in the past about ways we could enable this sort of cross-library updating, though I don't have the discussions handy. If you have thoughts on how you'd like this to work for your particular workflow, feel free to share.
But that;s EXACTLY what's needed.
I enter a book or article in Bibliog A and in Bibliog B.
I then find I have misspelled the publisher. Or I didn't know the pagination and later discover it. Or I add notes once I get to read it. That can also change its Tags.
So evidently I need those same corrections and additions to show up wherever the item is listed.
"If they were the same items, anyone in any library could make a change that applied to all libraries that had a given item."
If they are the SAME items then they need the same data on date, author, title, publisher, source, pagination, standardised across instances, as above.
If a different version is needed - maybe making private special notes for self, or for team colleagues, that could be done by permissions. If that is too complex then a string warning in the Help guidance,maybe even a popup dialogue, would prompt caution 'You are making changes to this entry which will also change it in other libraries. Proceed?'
Perhaps I haven't understood the difficulty?
Dan is looking for information on different users' needs to develop some form of solution to meet both of our extreme set of preferences and everything in between.
In principle some users need cross referenced updates, and some need standalone updates.
Why not make both available? So when creating a library/group I can opt to link it with other libraries/ groups to they update each other, or keep the current default standalone entries.
"While correcting spelling errors or adding DOIs or page numbers across libraries would be nice" This isn't just a pleasant bonus!
The status of a database rests on a) its structural design, and b) accurate input.
If corrections are not implemented then it's a garbage in garbage out scenario and the whole project is weakened. Mistakes, slips and omissions will be frequent because surely the whole point is that manual memory cannot manage 1,000s of entries.
I could easily make anything up to ten bibliogs on zotero. Their content would duplicate most core items, but each one would have specialist entries, and some core entries would not appear in some bibliogs. This would create a much finer grained service than the current crude categorisation I can make with one bibliog and its tags.
So I choose to run just one bibliog because the headache and time gobbling of manual updating would be so destructive. I make this a public service. The result is that I do not have the convenience of a bibliog that fits my own work.