It now becomes clear that most of these slash tags are not consciously input by scholars, but are automatically embedded by WorldCat, GoogleScholar, etc. So if you get data from these source, please try to pay attention to what kinds of tags are being embedded in your data. They effectively undermine the power of Zotero tags.

If one checks the General Settings tab in the Zotero Preferences dialog then vital switches are found under 'Miscellaneous' and 'Groups'.

If, under 'Miscellenous', the 'Automatically tag item..' option is deselected, then Zotero will not create tags based upon a page scrape. Deselecting all items under 'Groups' should prevent whatever notes etc, contained in a personal library are not transferred elsewhere.

WJG

Thanks Will. I would like to strongly recommend that people turn off the automatic tagging setting as described. Not only are the tags from GoogleScholar produced with several in each field--they are also quite often produced from a generalist perspective, and not really appropriate to a database like this for specialists of Buddhism.  -- Chuck

The "added by" information in zotero seems to be misleading and only registers the last person modifying an entry; the reference to the user who has originally uploaded the entry seems to disappear. Is there no function in zotero to see the "history" of an entry?

Christoph: I've been bothered by the same thing. But AFAICT there is nothing in the basic Settings for the group regarding this (actually, the options are extremely limited). I'm going to raise this issue on the user forum. If you are interested, please follow the discussion there. -- Chuck

In regard to Christoph's note above, I have started a thread on the Zotero forum under "Feature Requests" with the title "Edit history / retaining of <resp> information." Let's see if anything happens. -- Chuck

Having done a fair amount of keyword tagging and seeing how it works, I have now been tending toward inputting tags without diacritics, with the reasoning that:

(1) When people want to use tags via the tag input box, they can't input with diacritics;

(2) Since they are just tags, and not actual data, it is not necessary for them to reflect the precision one would need for publication;

(3) Diacritics in the texts in the blbliography are in any case not uniform. For example, we find Yogâcāra, Yogācāra, and in articles in the JIABS, even  Yogâcâra. Not even to mention incorrectly input, or partially input diacritics. There are all kinds of other examples.

Other opinions?

A month's experience with this shows that the matter of the proliferation of spurious, redundant, or trivial tags is something that will require continuous attention. To deal with this, I have just written a script that will check for keywords that only appear once, and which remain in that status for three weeks running. When such tags are identified, I will try to merge them with a related term, or up into a more general categorical level. -- Chuck