Compound bilingual journal titles may be fine

[fgnievinski]

Hi. I wanted to share a comment about the guide on Better Practices in Journal Metadata, section Combining Multilingual Metadata in a Single Field, which says:

This happens frequently when multilingual journals want multilingual metadata to appear on one article page, so they combine multilingual metadata in a single field instead of using the separate fields in OJS

I wanted to suggest using the article title or abstract instead of journal title to exemplify the problems of combining multilingual metadata in a single field. First, because it's a much more frequent issue, as it affects every individual article.

The second and more important reason is because some journals seem to genuinely prefer a compound bilingual title. For example, the Canadian Psychology/Psychologie canadienne appears literally as such in the journal full-title field of the article metadata deposited in CrossRef, for example:

<journal_metadata language="en">
<full_title>Canadian Psychology/Psychologie canadienne</full_title>

The practice seems recognized, for example, in the APA Style blog:

Note for this example that Canadian Psychology/Psychologie canadienne is a bilingual journal that is published with a [compound] bilingual title; if the journal title were only in French it would not be necessary to translate it [the journal title] in the reference.

Bussières, E.-L., St-Germain, A., Dubé, M., & Richard, M.-C. (2017). Efficacité et efficience des programmes de transition à la vie adulte: Une revue systématique [Effectiveness and efficiency of adult transition programs: A systematic review]. Canadian Psychology/Psychologie canadienne, 58, 354–365. https://doi.org/10.1037/cap0000104

[fgnievinski]

I also note the journal used as an example doesn't seem to fully follow the Guide's recommendations. More specifically, the journal title entered in OJS for the French locale is in French-only, but the journal title entered in English is in compound bilingual form:

Savoie, D. J. (2010). New Brunswick: Let’s Not Waste a Crisis. Revue d’études Sur Le Nouveau-Brunswick, 1.

Savoie, D. J. (2010). New Brunswick: Let’s Not Waste a Crisis. Journal of New Brunswick Studies / Revue d’études Sur Le Nouveau-Brunswick, 1.

https://journals.lib.unb.ca/index.php/JNBS/article/view/18191

[AhemNason]

I'll answer these in reverse order. Number 2, I know that this journal doesn't follow these rules. I host it at my institution and I also wrote this portion of the guide. I needed screenshots, and made them aspirational.

I also note the journal used as an example doesn't seem to fully follow the Guide's recommendations. More specifically, the journal title entered in OJS for the French locale is in French-only, but the journal title entered in English is in compound bilingual form:

Savoie, D. J. (2010). New Brunswick: Let’s Not Waste a Crisis. Revue d’études Sur Le Nouveau-Brunswick, 1.

Savoie, D. J. (2010). New Brunswick: Let’s Not Waste a Crisis. Journal of New Brunswick Studies / Revue d’études Sur Le Nouveau-Brunswick, 1.

https://journals.lib.unb.ca/index.php/JNBS/article/view/18191

[AhemNason]

The second and more important reason is because some journals seem to genuinely prefer a compound bilingual title. For example, the Canadian Psychology/Psychologie canadienne appears literally as such in the journal full-title field of the article metadata deposited in CrossRef, for example:

It's true that this is common and that a lot of journals very much want to assert their identity in this way. The issue is in the consistency of the metadata they present broadly.

Canadian Psychology is a great example because it's indexed variably across a number of platforms with different names. For one, it's ISSN is only registered against the English name. You can see that here: https://portal.issn.org/api/search?search[]=MUST=allissnbis=%220708-5591%22&search_id=14724982

The ISSN is hugely important to the name of a journal. If the journal insisted on this bilingual title generally and its ISSN was registered that way, I'd say this was fine. There's a registered "parallel title", instead.

Secondly, if I look this journal up in a variety of indexes and my own ILS, I don't get a consistent title. Sometimes I get both... but here are the titles I pull:

• Canadian Psychology / Psychologie canadienne
• Canadian Psychology/Psychologie canadienne
• Canadian Psychology
• Canadian Psychology. Psychologie canadienne.

Crossref themselves wouldn't vet against two languages in a title field. There's no review of submitted metadata. Any journal that puts both languages in the journal name field in OJS and deposits with Crossref would show the same. And Crossref does not currently have full multilingual metadata support (although they're working on it, based on JATs implementation of repeatable fields with language attributes).

The real recommendation here is to be as consistent as possible. Especially against ISSN.

I wanted to suggest using the article title or abstract instead of journal-title to exemplify the problems of combining multilingual metadata in a single field. First, because it's a much more frequent issue, as it affects every individual article.

I don't disagree with this! Abstract, in particular, we see a lot of. A note to @kaitlinnewson that adding these other examples to the doc would be welcome.

The practice seems recognized, for example, in the APA Style blog:

We recognize a lot of non-ideal behaviours in metadata because they're so frequent. And it's not surprising. But some citations for this journal will say Canadian Psychology, and some will have both languages, depending on where the user read that content. And the issue is that it may seem to some like they are talking about two different journals, not one journal that sometimes has both names in a single title field.

But! The beauty of recommendations is that they are ignorable. The committee did our best to promote metadata that would be consistently presented and legible downstream to better identify materials clearly across platforms. But, we also know that it's a tall order and one that isn't always supported (like how Crossref doesn't currently have a secondary language title field).

[fgnievinski]

Thanks for your comments, @AhemNason. I'm wrangling with my own bilingual journal titles, reason why I'm seeking guidance.

If you search for the online ISSN instead of the printed ISSN, then both titles are recognized for that journal:
https://portal.issn.org/resource/ISSN/1878-7304

Title proper: Canadian psychology
Parallel title: Psychologie canadienne

I expected the third variant, Canadian Psychology/Psychologie canadienne, to be registered explicitly as well, but it isn't. Maybe the ISSN register limits parallel titles to only one?

The long compound bilingual form does appear as the preferred form in the CrossRef Title List, along with a minor variation using hyphen instead of slash:

I totally agree with the recommendation to be as consistent as possible. My impression this is super important for CrossRef and all downstream indexing. Here's a related comment in the CrossRef forum:

We require that all future metadata deposits for that journal must contain journal-level metadata that matches the record established in our system by the first deposit. So, if you submit a journal title one way in the first deposit and then later switch to using a “parallel” or alternate title, the metadata deposit will fail with an error message that goes like <msg>ISSN "[ISSN]" has already been assigned, issn ([ISSN]) is assigned to another title ([title we have in our records])</msg>

So, I'm starting to suspect the recommendation for bilingual journals should actually be to keep the journal title fixed, to either language or to the compound bilingual form. If we're going to shoehorn mulitlingual content into a monolingual indexing infrastructure, then journal titles should not change when the content language changes, does it make sense? I.e., except for the journal title, every piece of multilingual metadata (abstracts, article titles, etc.) would still be stored in their own locales (internationalized and localized fields). I'm seeking confirmation from CrossRef about this interpretation, see more details in their forum.

PS: proudly ex-UNB here: https://unbscholar.lib.unb.ca/islandora/object/unbscholar%3A8512

[fgnievinski]

So, I'm starting to suspect the recommendation for bilingual journals should actually be to keep the journal title fixed, to either language or to the compound bilingual form. If we're going to shoehorn mulitlingual content into a monolingual indexing infrastructure, then journal titles should not change when the content language changes, does it make sense? I.e., except for the journal title, every piece of multilingual metadata (abstracts, article titles, etc.) would still be stored in their own locales (internationalized and localized fields). I'm seeking confirmation from CrossRef about this interpretation, see more details in their forum.

CrossRef folks seem to have confirmed the above interpretation:

For practical purposes, based on the way our title record system is setup (the purpose of which is mainly to prevent duplicate records by duplicate publishers for the same journal) you need to pick one journal title and stick with it. And, the language of the journal title does not need to match the language of the article title.

Maybe this recommendation could be included in the OJS documentation and guides?

[fgnievinski]

I think this issue could be moved into discussions, similar to #7569
pkp/pkp-lib#7569