General citations

Mar 2019
User request
Tagged examples, please include standard ones as well! Would like to have guidance and best practice on the standard too: Journal article, Book (whole), Chapter in book, Conference proceeding.”
The current sub-group is concentrating on Emerging Reference Types (Websites, Multimedia, Data, PrePrints & Software), does a new sub-group need to look at giving additional support on more traditional reference types above and beyond the existing recommendations?

JATS4R comment
To be revisited and to include from this draft recommendation:
Multimedia citations

User comment

All of the major citation styles (APA, MLA, Chicago, IEEE) are outmoded. They were designed in the mid-1900s to facilitate finding a book or article in your favorite library, and if not found, to enable a librarian to obtain it via interlibrary loan, and finally, if necessary, to facilitate ordering a copy from its publisher. These styles no longer meet the needs of a computerized society. They wrongly elevate form over function and do not take into account advances in technology.

Nearly all readers of academic and professional publications have easy access to the Internet via PCs, laptops, iPads, and/or smart phones. A written description of an item they want to obtain is not nearly as helpful as hot-links to websites that can be immediately displayed with a mouse click or finger tap. This is especially true if the link indicates the type of information that will be displayed.

Readers of a PDF or eBook need to be able to use URLs. They do not need to see them much less know when they were viewed successfully. These readers would much rather see an indication of what they are likely to retrieve if they click a link. These should be links to safe, stable websites that provide additional data that a typical reader may need about persons, places, things, events and concepts.

The displayed link should indicate its type. For example: review, preview, summary, synopsis, abstract, descript, analysis, critique, comment, preface, intro, foreword, TOC, refs, page1, excerpt, pdf, full text, download, borrow, CV, home page, bio, wiki, quotes, thesis, interview, audio, video, obit.

The number of link tabs and their names should not be restricted. All that is required is uniform usage within a particular book or article. Of course, the tags must be meaningful. Tags such as “click here” and “link” should not be used. Tags for multiple links of the same type should be unique. For example: review1, review2, review 3. If displaying the target of a link requires payment or a paid subscription, the link should include the cost or “sub”, e.g., download_US$24, or online_sub.

Here are some sample endnotes from An Existentialist Theory of the Human Spirit (Volume 1): To Love and Create … or Not by Shlomo G. Shoham:

58 Gershom Scholem (Stanford), Major Trends in Jewish Mysticism, NY: Schocken Books, 1961, p. 13, preview, First Lecture, review, full text. Some Kabbalist systems regard infinity as the equivalent of the Keter of the Sephirot. See Joseph ben Abraham Gikatila, Shaare Orah, Jerusalem: Bialik Institute, 1970, p. 138.

85 Hayim Vital, Sefer Ha’likutim, Jerusalem, 1913; Gershom Scholem (Stanford), Major Trends in Jewish Mysticism, NY: Schocken Books, 1961.

144 Talmud Yoma 77b (Sefaria in Hebrew, p. 233 in English translation online). Referring to the abduction of Dinah, Genesis 34:2.

176 Genesis Rabba 22b, full text.

The same approach can be used for references in alphanumeric order by author and date of publication. For example, these endnotes are from Biblical Humanism and Suicide Prevention: Where Did the Greeks Go Wrong?, a contribution by Kalman J. Kaplan to a Festschrift in honor of Shlomo G. Shoham.

Golden, William W. (p. 189) (1900). Maimonides’ Prayer for Physicians. Transactions of the Medical Society of West Virginia, 33, pp, 414-415; text by Fred Rosner, text by Dr. Harry Friedenwald.

Kaplan, Kalman J. & Cantz, Paul (2017) Biblical Psychotherapy: Reclaiming Scriptural Narratives for Positive Psychology and Suicide Prevention. Lanham, Boulder, New York, London: Lexington Books, pp. 260, web page, 74 slide summary.

Muntner, Suessman (1977). Medicine in ancient Israel in Fred Rosner (ed.), Medicine in the Bible and the Talmud: Selections from Classical Jewish Sources, Hoboken, NJ: KTAV, pp. 3-20.

Pirke Aboth – The Ethics of the Talmud: Sayings of the Fathers, edited with introduction, translation and commentary by R. Travers Herford, preface by John J. Tepfer, (1962), New York: Schocken Books, pp. ix + 176.

Preuss, Julius (German wiki) (1978). Biblical and Talmudic Medicine, Fred Rosner (translator & editor). Northvale, NJ: Jason Aronson, pp. 652.

Readers of a printed book do need to see URLs in order to use them, but these should be tinyURLs composed of a re-usable prefix + 7 or 8 alphanumeric characters that retrieve the same data as the actual URLs. Although readers of a printed book cannot click the link tags, these tags are not completely valueless. They do indicate the type of data that can be found on the web via a manual search.

You can obtain tinyURLs by using the online process at
Bitly, Rebrandly, Google Analytics, and SEMrush are alternatives to tinyURL. For information about them, go to

JATS4R response:
This is an interesting comment, although it is askew to our mission here (and to the mission of JATS in general).

You are looking to change the way citaitons are written, processed, and displayed by applications. For JATS, we need to have sufficient tags available for all citations that exist in journal articles (and all citations that might ever exist in journal articles, although this involves a bit of soothsaying) to be represented in JATS XML. The goal of JATS4R is to make the use of JATS tagging as consistently applied as possible to simplify reuse of the content.

The changes you are looking for are going to have to be applied by the publishers of content (what needs to be in a citation) and the applications that display and distribute articles (how do citations need to be shown or “activated” so that they can be useful). I would caution you, however, against throwing away the basic metadata elements in citations in lieu of a simple link or identifier. The simple citation for an article (Journal, volume, first page (or other sequence number)) has been a recognizable name or identifier for an article for a long time, and having that information might be helpful in a future world that has moved beyond Handle. Also it can be helpful for readers to know things like journal title or article repository, authors, publication date, and article title so that they can make a decision to activate a citation.

User response:
The draft recommendation looks really good! Having clearer instructions about how to structure citation data would be very helpful – especially for new or less common citation types.

I do have a question / suggestion regarding the publication-type attribute. I have seen unfortunately endless variation of values for this attribute. (In fact, our code includes a giant mapping which we have been building up for years.) So my question… Has JATS4R considered using CSL types as the definitive / recommend list of values? This would be extremely helpful for my work – and for anyone who’s using CSL / Citeproc tooling.

As far as I know, CSL is by far the best spec / toolset for working with citations (used, fwiw, by Zotero). Were I JATS king, I would recommend everyone use CSL types when setting the publication-type attribute.

July 2022
How to tag references issued / published across a timespan within element-citation?
Use two <date> elements, differentiated using “date-type” attribute
    <!– Other elements omitted for brevity –>
    <date date-type=”publication-start”>
    <date date-type=”publication-end”>

Use <date> and <date-in-citation>:
    <!– Other elements omitted for brevity –>
    <date-in-citation content-type=”publication-end”>

Dec 2022
I’m trying to figure out how to map CSL/biblatex bib types to JATS and came across this page.

On one page [1] you seem to suggest that types should be things like “journal”, “book”, “letter”, “review”, “patent”, “report”, “standard”, “data”, or “working-paper”. On other pages you mention that the publication-type should be “software” [2] or “preprint” [3] for specific kinds of citations.

Would it be possible to get a full list of recommended values? I am also looking for suggested values for publication types like thesis, collections, web pages, book chapters and blog posts (most urgently).





  1. Recent new requests and “Citations” have been merged into this one “General citations” issue

  2. @coryschires @Izzcohen @johanneswilm please refer to this Topic re your comments or additional roadmap issues – they’ve all been conflated into this one.

Provide feedback on this recommendation

Please note you are commenting on this specific recommendation. To suggest a new recommendation, please follow the link on the homepage. By proceeding with your comment here, you understand that your comment will be publicly visible and you may be contacted by JATS4R in case of further clarification.

You may use markdown to format your comment. For example, to allow <> tags to display, please start and end that portion of your comment with three backtick characters, ```.