How to manually populate Zotero minimum information for RTF scan?

KrisMatrix

It seems that if I want to use RTFScan, the bibliography needs to be of the form:

{Author, "Title", Date}

The zotero GUI though seems to want creator and date. I noticed that the 'key' info grabbed via the zotero web plugin is not always consistent. I am going to assume that the creator is the author. How would I enter this information more easily. I sometimes struggle with populating this data manually because on the menu on the right (in include image), I am restricted to the keys that are already included. I want to enter the information for title, creator and date, so that the RTF scan work properly.

https://s3.amazonaws.com/zotero.org/images/forums/u14168690/fx62zct6u1v7s5kqhcg8.png

tim820

At a minimum, manually typed citation entries for RTF scan only require {creator, year}, as shown at the link below.
https://www.zotero.org/support/rtf_scan

Zotero's *automated* scannable cite entries include extra information to avoid ambiguity. You can use Edit\Copy As Scannable Cite or the keyboard shortcut (Ctrl-Shift-C) to get those unambiguous entries easily for pasting the selected item's citation into your document.

Any remaining ambiguity in manual entries is resolved when you run Tools\RTF Scan from within Zotero desktop, as the last step in processing. At some point in the future Zotero may get a unambiguous, displayed citekey (like that used by Endnote and the RTF/ODF-Scan plugin), which will make manual entries unambiguous too (which may allow scannable-cite documents to be processed entirely within the word processor plugins).

KrisMatrix

Thank you. I realize that I can fix things once I do RTF Scan. But this means that everytime I do an rtf scan, I need to redo the matching process right? That is what I am trying to avoid. Sometimes, the creator field is populated, but the year field is not when I use the web plug-in. So, a single key can have duplicates.
That is the sort of ambiguity that I want to fix as soon as I pull.

adamsmith

I still don't 100% understand, but especially if you're using non-academic sources (and especially especially for regular website), automatic import will often be incomplete and you have to manually add things like a creator or publication year -- is that what you're asking? More generally, RTF Scan works, but is not the most stable form of text-based writing. If you're serious about this workflow, looking at something like writing in Overleaf, RStudio, Fidus Writer, or just markdown in VS Code is probably a better idea.

tim820

OK, I see now by 'Zotero web plugin' you meant the Zotero Web Connector's ability to scrape an author and date from web pages.

I don't know what information Zotero has access to for scraping a web page's creator and date, nor therefore why those Zotero fields are often empty. My guess would be that the web page has not provided that information. I often find myself filling in a web page's missing author or date field manually, if the page author has put their name and a date at the top of the web page (which of course page authors don't always do).

Regarding less ambiguous citations, you can try the RTF/ODF-Scan plugin. It uses an ODT file instead of an RTF file. ODT files can now be read and written by Microsoft Word (LibreOffice was the traditional way of doing that).
http://zotero-odf-scan.github.io/zotero-odf-scan/

KrisMatrix

adamsmith and tim820. Sorry. I may have made my questions confusing. Let me clarify.

1. I am not asking about fixing the web plugin tool.
2. The web plugin successfully pulls the information from the website but naturally not all data is necessarily available to pull. Or the tool can't grab certain data.

3. I want to manually enter this data in zotero. As you see in the picture I posted in the original post, for the resource with title 'The Constitution', zotero's web plugin has NOT populated the creator column or the date column. I want to edit these fields. When I left click on the record in the table, the side nav shows for that record. But the 'keys' don't necessarily match the names in the table.

Essentially, I want to always ensure (at a minimum) that all records have a title, creator, and date populated. The menu on the sidenav sometimes has date, author, effective date, accessed date, creator, title, short title, and a bunch of other details. It is not always clear which fields I need to populate to 'ensure' that I fill the three fields that I care about.

My question therefore is, what is the best way to ensure I populate this data.

To ensure I don't have duplicates, I want to always populate those three fields. These fields serve as keys for RTF scan.

Is my clarification helpful?

On the ODT scan, I had trouble setting that up. But i will take a second look.

adamsmith

"Creator" is a generic label that includes author, director, programmer, and whatever other category of, well, creator, exists across various item types.
"Date" refers to the primary date (i.e. "Date" for most item types).
"Title" is always right under the item type and is (and is literally "Title" for most item types)

KrisMatrix

So you are saying that I can't just create my own key and enter the label information? I have to find out what field maps to creator for each entry and enter the information for that field (which is what I currently do)?

adamsmith

Yes, the whole idea of a "Key" doesn't apply to RTF Scan. If you want to use actual citation keys, look at the other tools I listed.

KrisMatrix

I will have to look into so of the tools.
- I know of overleaf. Great if I want to focus primarily on latex.
- RStudio I thought was just an IDE for R.
- Fidus writer seems interesting. Don't know anything about it.
- Markdown is a useful format, but I am not sure how it serves as a bib manager and citation tool.

But it sounds like you are saying be prepared to work on a paper, and stick with a specific citation style for that paper. I use RTF citation because rtf scan allows me to swap RTF citation with any other style such as IEEE, Chicago, MLA, etc.

Perhaps you are saying that if stuck with overleaf, I could still use zotero for managing by references, but use overleaf with it somehow?

adamsmith

Oveleaf has Zotero integration, as has RStudio (which you can also just use as a Markdown editor, no need to use with R) and you can use Markdown (with pandoc) more generally and keep your references in Zotero, yes: many people do.

KrisMatrix

OK. I think I'd rather use latex over markdown. Let me poke around with overleaf. I have only used latex offline before. The website seems to say that online integration with Zotero is a paid feature.

KrisMatrix

Hey @adamsmith, Since you seem to know pandoc. Perhaps you can comment on my hypothetical use?

I use a writing app called scrivener. It CAN export to latex, but it sucks. It does a better job export to word docx format. It has a bunch of other export format, but I haven't tried them. It possible can export to some sort of markdown format.

Anyway, my thought is, write in scrivener. Export to docx or markdown. Any citations, use the csl style that apparently pandoc supports. Now, I don't properly understand how the csl referencing system works, so that will take experimentation. But have references in zotero. Export references to biblatex format file. Then using pandoc run:

pandoc --bibliography=zotero.bib --citeproc -o [someoutputformat] test.md
pandoc --bibliography=zotero.bib --citeproc -o [someoutputformat] test.docx

I HAVE tested pandoc for converting docx to md or tex files and I am satisfied with them. However, I haven't tried out replacing the references with the csl style.

adamsmith

This should definitely work with Markdown, yes. I don't believe pandoc will recognize citations in .docx format (but I'm not sure)

KrisMatrix

Ok. So Scrivener to docx (assuming its conversion to markdown doesn't suck). Then docx to markdown. Then markdown to whatever format, but her I use the bib for conversion.

It occurred to me just now that I don't know how to tell pandoc to use a particular citation style when it creates citations and bibliography.

adamsmith

All covered in the manual: https://pandoc.org/MANUAL.html#specifying-a-citation-style

KrisMatrix

Thank you adamsmith. I was able to get things to work. Here is another question for which you might have a recommended workflow. You may not know Scrivener, but for the sake of discussion, assume that I am writing in MS Word.

It seems to me that in order to get the unique key for biblatex keys, I need to select my references then export to a .bib file. Until I do this, I don't know what the keys are. Am I correct? This means that if I am researching on the fly as I am writing, then I won't know the keys to add into the written document?

It seems like the workflow I have to follow is:
1. Collect references before and as I write.
2. Write but add some search and replace signifier...say $$cite within the text.
3. When I am done with paper, select my references in zotero. Then export to .bib.
4. Then do the manual work of replacing $$cite with the keys from the .bib.

Is there a better way? There seems to also be a better bibtex plugin, but I am not sure what that is for exactly.

adamsmith

You'll want to do 1 for every workflow, but you can use the BBT add-on for convenient access to cute keys. It comes with lots.of documentation and there are a bunch of blogspot on using it with pandoc and markdown

KrisMatrix

Ok. I 'think' I got a relatively good method with the better bibtex plugin. It seems to offer a way to set quick copy to give me \cite{keyname}. This seems to follow a latex system. Since I am writing in word, it might have been better if just got keyname without the \cite{} and perhaps even better one that is pandoc format (i.e. [@keyname], but this is close enough. I suppose I could write perl script that swaps the \cite for pandoc format. Unless you know that there is a better way?

adamsmith

BBT has support for pandoc's citekey format: https://retorque.re/zotero-better-bibtex/citing/index.html#drag-and-drophotkey-citations