Syncing only part of my library (noob question)

wgw

AFAIK the z server syncs my entire local library.

Is it possible to sync by collections only? I.e. prevent syncing of all my library and limit it to select collections?

For example, I have a collection "polysémie" and a collection "programming/javascript". Could I ask the z server to download anything on the z-server to my local library and to upload only those things from those two collections?

My problem is that I have a mass of things which I don't want to be out there in the world (even privately stored on Zotero). But other things that I do want out there.

I understand that as a member of a group, only the group collection is synced. Can I sync as a group member only without syncing my entire library?

In short, I don't see how I can have fine control (collection level) over what is synced in my library.

(Sorry for what is no doubt a noob question.)

dstillman

Is it possible to sync by collections only?

No, this isn't currently possible. The best you can do right now is create separate groups for those collections instead.

wgw

Here is the scenario I am facing:

1) I have never synced my library, but I have a few items on the server that are not on my local machine, as well as a large local library; I don't want the whole local library to go up on the server.
2) I create a group and I am a member of that group (necessarily?).
3) I then hit sync (manual) on my local zotero. (Or maybe the sync can be done specifically for the group only.)
4a) My whole library does not go up! (I hope! Can I cancel if it starts to go up?)
4b) Any references in my group that are on the server come down.
4c) Any references in my local group collection will go up.

Done

I think the best idea is for me to set up on another machine with a small local library to see how sync works. Or perhaps another way would be to "detach" my main library (= rename it temporarily?) and use one that I can experiment with to see how sync works.

Thanks for your help (and software!)

dstillman

Sorry, what I said above didn't even make sense. Having separate groups won't help you if your concern is having data on the Zotero server at all, because there's currently no way to selectively sync data. All you can do is sync everything and choose which libraries are public.

dstillman

If you're using Zotero for Firefox, you could set up another Firefox profile with a separate Zotero library that you didn't sync, or use separate Zotero data directories in Zotero Standalone and Zotero for Firefox (if you tell the one you install second not to share the other's data directory, as it would by default). That's certainly less convenient than having everything in the same interface, though.

wgw

Thanks for the tip. Using Firefox profiles is a good, simple solution for me. I can work as normal, then export the collection and import it into the second profile. (I only need metadata.) From there, a sync to the server, and then export-import in the reverse sense to the main Firefox profile.

This will affect plugins, but again, that isn't a problem for my use-case.

Another way to go would be to write a (python) routine that would do a collection-specific sync between local profile databases. But that is something that can wait -- I don't mind doing it by hand.

Thanks for resolving this issue.

sdspieg

Still no change on this Dan? We are once again confronting this issue. We have the data for multiple projects on our storage plan. Multiple people work with multiple projects. But Firefox doesn't like big libraries - it becomes quite slow and unwieldy (because of the many attachments that are also stored there). So we are now starting a new project, and would like to have a few of our collaborators work with JUST the new collections. It really WOULD be nice for us paying users to have a selective sync option...

adamsmith

not arguing about the usefulness of selective sync, but, Stephan - for your purposes, couldn't you just put the new project in a separate group and have people just join that group? The group could be owned by the same paid account, so you wouldn't incur additional costs.

dstillman

Right, the concern above was about not wanting to sync some data to the server at all.

sdspieg

Sebastian, That's what we do of course. And we have loads of groups. But whenever a new person then installs Zotero and syncs to our paid account, he gets all the other groups too. Or am I missing sthg here?

adamsmith

why do you have a single account for all users? Remember group storage only counts towards the account of the owner. So everyone else can have a free account of their own and just join the groups they need.

sdspieg

Because most of our groups are bigger than 200 Meg or whatever the limit is... We are talking at least a few 100s of records per group and almost all with pdf etc.

adamsmith

nono, you're still misunderstanding how payment works. Any files in any group _only_ count towards the storage quota of the group owner. You can have a group with 10GB of files, owned by one account with unlimited storage and have 15 members in the group, all with access (online and locally) to the files, none of them paying.

sdspieg

Ok, I'll try that. Seems to me they told me they still ran into trouble that way. But we'll try tomorrow! Thanks.

adamsmith

if that's the case that'd be a Zotero bug, though, and it should (and would) get fixed.

Tparrettx1

Its not possible and probably never going to be possible with their current business model. ALthough its very useful, Zotero is freeware, and has been designed to maximize the cloud library so that you have to pay them for extra space. IF they didn't do this, then people could just import into their standalone and minimize their cloud libraries. It is a major problem with large libraries, libraries you want to comparmentalize, or libraries you want to keep private. To try to export or import with attachements from the cloud library is probably made difficult for the same reason, as Zotero appears to obfuscate this unless you construct the library totally as standalone. I wish there was another way to pay for this, so it was more convenient, but that is their business model and unfortuantely it makes things very inconvenient for some types of users who might be more than willing to pay by means of a different business model. Zotero is the best out there for convenience of downloading references and attachements (pdf of complete article etc) by simply websurfing, so its much better than the competition especially in this regard so its just ashamed these kinds of inconveniences exist.

aurimas

Its not possible and never going to be possible. ALthough its very useful, Zotero is freeware, and has been designed to maximize the external library so that you have to pay them for extra space. IF they didn't do this, then people could just import into their standalone. It is a major problem with large libraries to try to export or import, as Zotero appears to obfuscate this unless you construct the library totally as standalone. I wish there was another way to pay for this, so it was more convenient, but that is there business model and unfortuantely it makes things very inconvenient for some types of users.

Yeah, no. If Zotero wanted to squeeze all the money out of you for storage, they would not allow local-only file storage, turning off file sync for personal and/or group libraries, attaching links to local files that can be synced via whatever third-party service you want, and allow using third-party WebDAV services. They would also charge for storage of metadata. Whatever limitations are currently in place are for technical reasons, not financial.

Tparrettx1

No I think they do not try to squeeze every cent, and there tools are extremely useful and ahead of the curve. If you have a large standalone library, you may have articles you downloaded to zotero online, but would like them in your standalone library. If you sync totally, all those articles in your standalone will go out into the cloud, where for various reasons, one may not want. If you have a huge library, it will also fail unless you pay. I've tried to make an rdf+folder(with attachments) to circumvent this, but it seems this option is not available in the online zotero (although it is in the standalone zotero). I've interpreted this as a business rather than technical issue, as it is understandable Zotero would want us to maximize cloud storage space for their profit, and attachements (namely pdf file attachments) use up most of the space. If I am wrong and this is possible, please let us know. This was a workaround I tried for partial syncing (to import a rdf+attachement library by trying to generate this with an online library), but this functionality only appears to exist for the standalone version of zotero. The export option in the online version does not appear to have this (along with many other) functions found in the standalone version.

aurimas

Sorry, I don't really understand how your issue relates to this thread (I'm also not entirely sure I understand your issue in general). I'd suggest starting a new thread and explaining in detail what feature you are looking for. Best I can tell, what you're talking about is not related to this thread.