Emiliano wrote a Seektable export translator recently at my request:

https://github.com/retorquere/zotero-seektable-export

This works well to export data to a CSV table which can then be imported to Seektable:

http://www.seektable.com

From there you can do pivot tables and other searches surrounding your tags and other fields.

If you are looking to do something more advanced, as a private project Emiliano also wrote a sync utility to output an entire Zotero library to Postgres SQL. I can share further details if you are interested.

I am planning to soon release a beta version of a website which will incorporate Zotero, MySQL, Postgre SQL, and other software to organize the literatue of medicine. If you are interested in participating to extend that to the humanities, contact me (my login name is my email) and I would be glad to discuss this further.

Thanks for your quick responses and suggestions. My apologies for my dilatory response, but I have been researching Zotero, the better to reply.

Regarding the web-based database software, I should have stated that I do not like to use the cloud or web-based software for security reasons. I prefer to keep everything on my desktop. That goes for the database software I am looking for as well.

I am looking for a desktop database program to keep track of my Zotero database. I am still thinking how I should structure my Zotero database. I will need, for example, to keep track of the tags I create and whatever data I record in the related field.

I am presently grappling with two major issues. Firstly, how to structure my Zotero database by note items rather than by citations. My impression is that Zotero is oriented more towards recording citations that have notes rather than notes that have bibliographical references. Perhaps I am conceptually unable to make the leap from index cards to digital databases.

For a fuller explanation of what I mean, see the addendum to this response.

Secondly, if I should structure my Zotero database with tags or with sub collections or some combination thereof. My impression is that they are something of
alternatives with the possibility that sub collections could quickly become very unwieldy.

I have extensively Googled these issues without too much success. The notes versus citation issue was extensively discussed over a decade ago in the Zotero forum. I imagine that since then the ability of Zotero to handle notes has greatly improved.

I do not want to reinvent the wheel, so I am thinking of putting a post on the forum asking how members have structured their Zotero databases around notes rather than citations.

As to my description of MS Access as for "writing programs to routinely manage complex database". I was trying to say that Access seems oriented towards writing programs in visual basic to manage and analyze databases. I found it cumbersome and time consuming compared to Dbase IV for a one-off analysis of a database.

Thanks again for your suggestions and comments. I very much appreciate them, as I will anything you might have to say in response to my reply.

NOTES vs. CITATIONS

I am, for example, planning to write various pieces on the voluminous works of Karl Marx. One article will be on his view of post capitalist society. Another will address his various accounts of class. Still another will explain that his belief that capitalists exploited proletarians did not rest on his theory of value.

I do not think it makes sense to structure my Zotero database by the various books that Marx wrote (Capital, Grundrisse, Civil Wars in France, etc.). Instead, I want to structure my Marx data by either subject (‘class’, ‘value’, ‘surplus value’, ‘landlord class’, ‘servant class’, “creative labor”, “Milton” etc.) or, perhaps the articles I intend to write (thought I am skeptical of this approach).

Fortunately, a historian long ago suggested how to structure Zotero by subject rather than citation. (See: https://www.chronicle.com/blogs/profhacker/taking-better-notes-in-zotero/36561).

My understanding of his suggestion is to rely exclusively on notes and have only one citation collection. All the individual item notes from a book or article would be related to the item citation through the ‘related’ field of both the item note and the item citation.

More concretely, I would have a single bibliography as a collection (named ‘bibliography’). The entry that would appear in the related field in both the item citations in the bibliography collection and the note item would use the ‘short title’ form of referencing a source e.g. Jones [1975], Jones, A [1975] (if there are two articles/books etc. by authors names Jones in 1975). (See: https://www.kcl.ac.uk/artshums/depts/french/pdf/style.pdf).

I would, for example, have a citation item for Capital, Volume I, Karl Marx, 1859 etc. In the related field of this citation item would be, Marx, Capital [1859]. In a collection labelled 'Marx' I would have item notes with first lines reading, say, Surplus Value p. 85, Capitalists qua capitalists do not work pp. 44-45, Value equals price (for purposes of exposition) p. 98. In the related fields of each of these item notes would be Marx, Capital [1859].

I have yet to fully explore the possibilities of the Zotero ‘related’ field to see how feasible this is.

Hi,

Thanks for your help and advice. I very much appreciate it.

I've started using Apache Open Office Base. So far it seems to be dong the trick.

As for my plan to structure my Zotero database solely with notes (outlined) above), I've moved away from that. I'm now thinking of a combination of collections, sub collections and item notes, mostly relying on citations within collections, with these collections being organized by the articles I plan to write.

What I am wrestling with is a) not getting too many sub collections b) the tag hierarchy problem which has been much discussed on this forum

Thanks again for your suggestions and comments

Symphytum can't do the job?
https://github.com/giowck/symphytum#symphytum

I don't know. I am a basic user of personal databases.

You absolutely do not want to edit the Zotero database using tools that are not Zotero BTW. Zotero has expectations in place for how the database changes and you could well end up with a database Zotero could not safely load again.

Thanks for all your comments and suggestions. I very much appreciate them.

I first tried Apache Open Office and then switched to LibreOffice Base when I learned that Apache Open Office had suspended further development. Unfortunately, Base did not seem to have all the menu features that I needed that MS Access did. I don't want to write scripts.

I've looked at quite a few other database products including Symphytum which got a poor review here: https://www.macworld.com/article/2060829/symphytum-1-1-review-bento-like-database-app-a-work-in-progress.html.

I ended up where I didn't want to: back to MS Access. The worst thing was it kept crashing on me. I had to reinstall MS Office and then do a Windows 10 reinstall that left my programs/apps and files unaffected. Even after that, Access kept crashing when I invoked certain wizards. After searching through the Windows event viewer and running scan now Dism etc. I finally seem to have Access working. Wish me luck!

I now have a database to keep track of my tags.

The last thing I need is a software program that will display my collections and sub collections (I'm still having trouble conceptualizing/visualizing this) as tree diagrams.
Genealogy/family tree software might be adaptable. MS PowerPoint allows for tree-like organization structures, but they are limited to the size of a slide. I could probably customize something in MS Excel. A purpose built software would still be preferable.

I'd be interested to know how other Zotero users keep a balance between sub folders and tags. I can see it would be very easy to run wild with sub collections and soon find yourself with a collection that is ten deep and twenty wide at its deepest.

I've perused quite a few discussions of the collections/sub collections vs tags issue (e.g. https://forums.zotero.org/discussion/79/hierarchical-tags/, https://www.reddit.com/r/AskAcademia/comments/8mhdc9/how_to_organise_my_zotero_collections_or/). The problem with deep tree structures seems to be writing the searches to generate the Various Zotero outputs. The advantage of deep tree structures is they are easy to organize and visualize. Tags on the other hand seem to rely a lot more on memory.

Anyway, any thoughts or advice on this issue would be welcome.

Thanks again for your thoughts and suggestions.

Final comment: yes it is probably a very good idea not to edit your Zotero database in another database program and then import it back into Zotero. After forty years of using computers I know one thing: computers are very 'picky'.

1) Not with any existing tools, no. Might be scriptable, but probably not easily.
2) Also not with any existing tools. There are workarounds and again, probably scriptable solutions, but nothing that would be easy to use.

One additional comment about Symphytum. You do not need to build or preconfigure the database. Simply import the .csv file you exported from Zotero. Symphytum will use the same field names and field contents

A lot depends upon the required performance and the number of simultaneous users that there will be. If you want a top-of-the-range database which supports potentially thousands of users then have a look at MariaDB (links below). For small databases and single use though, the obvious choice is SQLite3. SQLite3 doesn't have a separate DB engine, instead it uses a library to manipulate a single data file. Currently there are known to be over 1,000,000,000,000 instances of SQLite3 running and you are almost certainly running it without knowing! It is slower that MariaDB and does not have some of the exotic features of the likes of Oracle or MariaDB.

I've checked the documentation for both and if you are still running Windows then it is fully supported, as of course is Mac, Linux and Android.

* https://mariadb.com/kb/en/
* https://www.sqlite.org/index.html
* https://www.sqlite.org/about.html