Some feed back from a grateful and excited user

[HughP]

I saw this plugin and said: "What a clever idea!" I proceeded to install and use the plugin on my Zotero library of over 3,000 items.

I first tried the plugin with one item not so bad. Then I tried it with the other 2,999 and it took about 1/2 hour based on the logs of materials added. – That was a bit unexpected, and I also felt a bit foolish for trying so many. There didn't seem to be a way to escape the script or to know what the consequences would be if I force quite/escaped.

I found the link in the left click menu button was an intuitive place for reaching the service as that is also where the zotfile service is located. But I did also look for a button in the UI. Is there one? there didn't seems to be anything in the application menu (I'm on Linux, Zotero standalone).

What was unexpected and I still don't know what to think about is all the new folders created under the library. There is something nice about being able to access the related citations in this way, but I think these should by default all be in a common sub-folder rather than directly under "My Library". I have a lot of collections and being able to hide these with a single click would be really helpful. Zotero does not allow the user to multi-select these library folders and move them, so now I have several hundred folders to manually move one at a time... this is not really all that encouraging. Also, I don't usually relate to citations by their DOI, so if these folders were titled by the name/title of the entity being referenced that would be much more usable.

Related resources: It seems (and please correct me if I am wrong) that when a DOI match is found a new record is created and then Zotero steps in and fills in the rest of the metadata. So, for items that are already in my Zotero database I now have two records. But I also now have a bunch of records which I might not have had before.

It is also unclear to me if when I am in Zotero's merge items view if items which are merged will appear in both collections still .... as a new item seems to be created for each resource citing it. That is if I have two citations each citing a single paper which I don't have in my current zotero lib, this plugin will download two copies of the metadata, one for each citation's reference.

The references in the newly created folders are to the new imports not the ones which already existed in my library. It seems that I can manually relate resources but this is not automated? isn't this the point of #9 I think automating this makes this plugin functional. Without connecting new imports to the user's existing entries the plugin seems to create a lot of noise in the user's library.

Also, again in the UI in the related entities tab everything is referenced as DOI, Titles are much more usable to end users, though DOIs are obviously not un-helpful, just helpful to a smaller degree.

Thank you for a lot of interesting and relevant work. Keep it up. I look for forward to seeing how this project evolves.

[zuphilip]

Thank you for your feedback. This is very much appreciated!

The whole plugin is very much Alpha/Beta version. Thus, my usual test scenario contains abut 5 items and I don't think I have ever tested it with more than a dozens items. Therefore, it is good to know that it also works with several thousands of items. I am sorry that it took you hour(s) and could not be stopped. An easy thing is to show a warning before continue with a lot of items, see #13.

There is a preference pane for the plugin if you click on the Tools -> Add-ons -> Zotero Open Citations -> Options. Thereby one can configure the number of retrieved items.

The newly created folders should be subfolders of the currently connected one and not the overall My Library, see also my screencast in the README. Do you say this is never working for you? Maybe we should discuss this in a separate issue (please open one with any additional info you have).

Yeah, I agree that the name of the collection is maybe not very useful (but it is very accurate:). What exactly do you suggest to use instead?

It seems (and please correct me if I am wrong) that when a DOI match is found a new record is created and then Zotero steps in and fills in the rest of the metadata. So, for items that are already in my Zotero database I now have two records.

Yes, that is correct. For the newest N=3 (but configurable) citations the DOI is taken and searched in Crossref and the received metadata are saved as a new item. This will potentially create duplicates of items which are already in your library. Further discussed in #9.

Also, again in the UI in the related entities tab everything is referenced as DOI, Titles are much more usable to end users, though DOIs are obviously not un-helpful, just helpful to a smaller degree.

Are you referring to the note created with the list of all citations found? I agree that the title and other metadata would very much be preferred here. However, the COCI service from OpenCitations does not provide this information. One could ask them, whether they want to change that. Alternatively, I also hope to enrich the data with the information already in the library #9 or received by Crossref #14.

[HughP]

I saw #13, I agree. About Preferences: see #15 The reason the new folders were under My Library was because that must be selected in order to select all files. I see #9 and agree that connecting to the existing entity would be preferred. I see #14 and wonder if something like the following might work...

select Zotero Entry
query COCI API
get response, take response, search current Zotero Lib for result,
if positive create [temporary object] based on Metadata in Zotero Lib, Use this [temporary object] to write the link in the note including the title of the link, and the "citations of" folder name,
if no positive result, search crossRef, if positive response from CrosRef create the new real object and then pull in data from new object into [temporary object] and write the link in the note including the title of the link, and the "citations of" folder name,
else if fail on Zotero, and CrossRef, go to next selected file

[zuphilip]

The suggested steps sounds fine for me in general, but we will see better once we tackle #9.

The reason the new folders were under My Library was because that must be selected in order to select all files.

Okay, but how did you then select the items without any collection selected?

Any suggestion for the name of the folder? BTW I am not sure how easy it is to update the same subfolder, so maybe we should at least let the DOI as part of the name in it.

[HughP]

For the folder name, I recommend the title of the paper plus the last 4 digits of the DOI. It puts the really recognizable thing for users first, and then the unique ID at the end which might not show any way.

[HughP]

Here is how I selected all the times in my library. Select the top part on the left and then in the center window select all, then right click and query COCI...