Required fields and additional documents types

autumnrustle

Hi. Is there way to:

1) mark requared fealds for each document tipe?

2) add new document types?

adamsmith

No to both, sorry. 2) might happen in the medium to long run. I don't think we'll do 1)

dstillman

(2) is still planned, but no ETA.

For (1), do you mean to show currently missing fields that would be required to cite an item in a given citation style?

I've long thought it would be useful to have something like that — say, to be able to set a style in Zotero and automatically show required field labels in red and (maybe) optional ones in orange when viewing items. @adamsmith, if we could do that technically, do you not think it would be useful?

adamsmith

I don't think this is possible technically from CSL's point of view. It's impossible to tell from CSL styles which fields are required and which are not -- e.g. you can obviously cite a book without an edition number (if there's only one edition), but the edition number would be required for a citation of a 3rd edition; or you can cite a journal article without page range if it's the online first version or an online only article, but would need it for a print version of an article etc.

So I'm quite skeptical about doing this with citaiton styles in mind. I could see, though, that a group would want to enforce certain data entry norms. That, I think, could be very useful.

autumnrustle

@dstillman Yes. Becose all styles must have minimal not empty filds.

dstillman

@adamsmith: Those examples make sense, but those would both count as optional fields, right? Are there not fields in some styles that are simply required, which are just inserted as empty strings in the output?

adamsmith

But the fields _are_ required when the information exists. Beyond the title (and URL + access date for webpages), I don't think there is any information that is required if you can't find it, so I really don't think the optional/required distinction makes sense.

bwiernik

In CSL, empty fields are simply not printed or styles might have some sort of filler text (e.g., anonymous, no date).

DWL-SDCA

"I don't think there is any information that is required if you can't find it...". While I agree that there will be incomplete information in some Zotero records until the complete metadata becomes available; I believe that one should _never_ cite anything that hasn't been fully read. If you have read it, you should provide a full description of the version you read. For an epub ahead of print, you cite it as an epub, etc. I believe that citations serve at least 2 key purposes: 1) give credit to those who produced the item; and 2) allow a reader to find and read the cited item. It isn't necessarily easy to flag all needed fields of a record to meet both of those purposes. For example, if you have a doi you may not need a page range but if you have neither of these you shouldn't be citing a journal article but, perhaps, a manuscript.

dstillman

Beyond the title (and URL + access date for webpages), I don't think there is any information that is required if you can't find it, so I really don't think the optional/required distinction makes sense.

Fair enough. This idea dates to the early days of Zotero, when I think various styles produced what was pretty clearly incorrect output (e.g., a comma followed by a period) when fields were missing. So if that's not the case anymore, I see how this wouldn't work.

Rintze

(I think that the field order in the right-hand pane could be improved a bit to put more stress on what generally are the important fields for citations; the inconsistent naming of "Publication" and "Journal Abbr" is also confusing. But regarding the discussion above I agree with the others that determining required fields based on CSL styles is not straightforward.)