Zenodo items vs. Zotero items

bjohas

Zenodo has these items types (https://zenodo.org/deposit/new):

Publication:
Book*
Book section*
Conference paper*
Journal article*
Patent*
Preprint+
Project deliverable
Project milestone
Proposal+
Report*
Software documentation+
Thesis*
Technical note
Working paper+
Other+

Poster+
Presentation*
Dataset+
Image+
Video/Audio*
Software
Lesson+

I've marked the ones that are available in Zotero with "*".

I understand that the categories in Zotero are partially about "what the item actually is" vs. "how the item should be formatted in a bibliography" (i.e. how the CSL will format the item). However, I wonder it may be worth considering some of these (e.g. marked with "+")?

I think "preprint" would be really useful for our uses. Proposal, working paper, dataset, lessons, and other could be helpful too.

Clearly the typology of assets you anybody would want to record is infinite. But perhaps it worth considering them? (Are there standards for these types?)

Alternatively, I would love to have sub-type field in Zotero, where I can just say what the item is (in our context).

(Also cf. https://forums.zotero.org/discussion/15636/changes-to-fields-and-item-types-for-zotero-5-1.)

What do others think?

adamsmith

Subtypes exist for some Zotero item types (e.g. "type" for presentation, thesis, and report). Freetext fields aren't ideal though and can mess quite a bit with citation styles.

Item types aren't standardized and everyone kind of does what they want, which leads to the messy situation we're in. Obviously we can't just add everything or Zotero becomes unusable.

We'll definitely add dataset to Zotero & it exists in CSL already.

We've always treated working paper and preprint as reports in Zotero and that works fairly well, but I can see the case for adding that as a category -- that'd clearly be a single category, though. I don't think there's actually a significant difference between the two.

I'm not sold on "proposal" or "lesson" -- there's never been serious demand for those.

Image is artwork in Zotero.

Poster is a type of presentation.

bwiernik

We have in the past discussed creating a separate Image item type to distinguish non-art images (e.g., technical photographs, historical pictures, figures), in the same manner as the distinction between Film and Video Recording. I’m still in favor of that.

bjohas

+1 for preprint and
+1 for a separate image item type.

Is there a guide in the Zotero help pages somewhere for "I have an item that type X, how do I put it into Zotero?" That would say: "Posters? Use Presentation, add type: 'poster'"?

mark

+1 for adding preprint in the UI (possibly by renaming report to preprint)
+1 for the image item type

Motivation
When the original list was devised, the present surge in preprint servers couldn't be foreseen; now many (new) users will expect this to exist as an item type; indeed in most academics' daily practice it's going to be much more salient than many other existing categories like Film, Statute, Instant Message, Podcast

Likewise for Image, the original list is strongly humanities-bent, with lots of granularity for history and art-related types, reflecting the origins of Zotero. As Z has significantly expanded its userbase, I think having Image as a separate type in the UI would be helpful to many users.

bjohas

I agree with Mark about the additions, but not the removal/renaming of report :)

theadri

Report item type is central those, like who work mainly with grey literature and should not be renamed. Other useful items type to add to the list would be :

+Technical note
+Working paper
+ image / photography
+ dataset
+ proposal

bjohas

I agree, report should stay! I've I've counted the votes right, we have:

Preprint+++
Working paper++
Proposal++
Image+++
Dataset++ (will come)
Technical note+

I do think that preprint is really important, partially because it's so common in some disciplines (with preprint servers) and needs to be distinguished from the ultimate journal paper that results. Happy with preprint/working paper as a single category (maybe with a 'type', as the type often appears in the citation). @adamsmith, do you think that might be possible?

galicarnax

"We've always treated working paper and preprint as reports in Zotero"

Actually, the trouble with preprints is not only that there is no separate Zotero item type for them, but also that there is no consistency across translators. E.g., preprints from SSRN are imported as Reports. Yet, preprints from arXiv are imported as Journal Articles. Of course, that's good for preprints which have been accepted in a journal and have associated DOIs, but it's not good for unpublished preprints. For me, it sounds much more reasonable to map preprints to existing Manuscript type, rather than to Report or Journal Article.

adamsmith

we had arXiv preprints as reports and got a lot of feedback that that doesn't work well, in particular because most bibtex styles are set up to treat them like journal articles and in the disciplines covered by arXiv, that's probably the dominant way to format bibliographies.

galicarnax

Thanks, I've searched the forum on this topic, and from what I can tell, the problem was that all arXiv preprints were mapped to Report, even those which were published in a journal and so were surely citeable as @article in bibtex. But I rarely see that unpublished preprints are cited as @article - actually, @article type requires 'journal' field which is irrelevant for unpublished pieces. Usually they are cited as @misc (which allows to include 'eprint' field -- see, e.g. https://arxiv2bibtex.org/). Ideally, arXiv translators should import papers from arXiv as Journal Articles if they have DOIs, and as Manuscripts (or at least Reports) otherwise.

adamsmith

Here's the thread in question: https://forums.zotero.org/discussion/comment/163677/#Comment_163677

and it's specifically also referring to the preprints.

bwiernik

Just to also comment from the psychology perspective—Journal Article also creates a more reasonably formatted reference in APA style, with PsyArXiV or whatever the (p)reprint server is being listed as the “Publication”, particularly if the server provides a DOI or other ID, compared to what is possible with Report.

Personally, I would like a Preprint item type in Zotero to visually distinguish it from journal articles, but I am generally of the opinion that that should be mapped to CSl article-journal, not report or a new item type.

The distinction between a preprint and a journal article is increasingly arbitrary anyway.

adamsmith

My view on the Zotero side is still this:

We've always treated working paper and preprint as reports in Zotero and that works fairly well, but I can see the case for adding that as a category -- that'd clearly be a single category, though. I don't think there's actually a significant difference between the two.

If we do make this a new item type -- suggestions for label very welcome (I fairly strongly dislike preprint) -- we could just map it to CSL's article, which is effectively unused and also seems like a good description: it's an article, but not published in a journal or magazine etc.

galicarnax

Sounds reasonable.

The only thing still mysterious for me is why this idea to map unpublished articles to Reports, when there is a dedicated item type in Zotero for this: Manuscript?

"An unpublished manuscript. Use this type for both historical documents and modern unpublished work (e.g., unpublished manuscripts, manuscripts submitted for publication, working papers that are not widely available)."

(https://www.zotero.org/support/kb/item_types_and_fields)

adamsmith

Working papers and preprints aren't really unpublished, though. They're published on a publishing platform which, typically, ensures permanence, stable linking etc. They're just not in a journal.

Unpublished work is both systematically different and commonly requires a different citation.

galicarnax

Well, for me, 'published' sounds closer to 'peer-reviewed' rather than to 'stable-linked' :) I would say preprints are deposited in a repository platform. Not a native English-speaker, though :)

bwiernik

In terms of how most citation styles talk about "published", it more broadly typically means "publicly available" and often "available with a stable link".

@adamsmith I think there are enough styles that request different formats for Government/Technical Reports (and similar) and Working Papers/Preprints to justify a different item type for them. CSL article seems like a really good fit, particularly because CSL doesn't do "fallback" types any longer.

Regarding a UI label, there is not clear winner in terms of existing usage between "working paper" and "preprint". The OSF preprint servers (SocArXiv, PsyArXiv, AgriRXiv, etc.), BioRXiv, and PeerJ PrePrints all use "Preprint" even when referring to e-prints, reprints, author-revised accepted manuscripts, etc. RePEc, ResearchGate, and SSRN mostly use "Working Paper" (as well as "Research Paper" https://www.ssrn.com/en/index.cfm/rps/). ArXiV uses "e-print". Institutional repositories seem to mostly use either "Working Paper" (e.g., http://www.baruch.cuny.edu/mspia/centers-and-institutes/center-for-nonprofit-strategy-and-management/working-papers.html) or no label at all (e.g., https://conservancy.umn.edu/pages/about/).

Given the mix of Preprint and Working Paper, I'm not sure either is particularly great (e.g., in Psychology, folks who talk about these types of works almost exclusively use "preprint" rather than "working paper"). I think a label like "Repository Paper" is sufficiently general without being that onerous or awkward.

adamsmith

Repository paper wouldn't work for established working paper series like NBER, which aren't on repositories, though... I also wonder if using a label that almost no one recognizes makes a lot of sense (that's e.g. the problem with Research paper, which I think is nice otherwise: no one is going to know what to put there -- what _isn't_ a research paper...)

Working paper/preprint is too long.

bwiernik

I mean, I would call NBER a repository, but that’s probably a distinction I’m just not familiar with.

I am not particularly fond of “E-Print” or “e-Print”, but those seem the most encompassing of the range of Working Paper/Preprint concepts. If the choice is between Working Paper and Preprint, I personally strongly prefer Preprint because that is the standard in my field and I think it much more accurately captures what, for example, arXiv is functionally, but Working Paper might be a more “general” feeling term, even if many members of this category aren’t being “worked” on.

bwiernik

CrossRef also uses the term “Posted Content”.
https://support.crossref.org/hc/en-us/articles/213126346-Posted-content-includes-preprints-?mobile_site=true

“Posted Paper” might not be a bad idea, as it captures the major distinctive feature of these items (they are posted rather than formally published) without the additional semantic baggage of Working Paper or Preprint.

bjohas

Preprint would work for me, and I'd use it specifically for works that are going to "appear in print" (potentially in a closed publication). So preprint means "last public version, before publisher formatting".

"Posted paper" or "repository paper" I've not heard before.

See follow-up question here: https://forums.zotero.org/discussion/72825/zotero-item-types-list-and-clarification

bothide

A preprint is not a manuscript, neither a publication. It has its own life, halfway between these two, and should be its own Zotero item.