Re: [jats-list] Re: @xml:lang on <related-article>

["Alf Eaton eaton.alf@xxxxxxxxx" <jats-list-service@xxxxxxxxxxxxxxxxxxxxxx>]

Given that HTML has an `hreflang` attribute which accompanies the `href`
attribute for exactly this use (usually used to link to alternate languages
of the current page, e.g. `<link rel="alternate" hreflang="fr" href="
https://example.com/";>`), it would be nice if there was an equivalent
`xlink:hreflang`, but unfortunately as far as I can tell there isn't. I
suppose JATS' `related-article` element wouldn't allow the use of
`xhtml:hreflang` imported from the XHTML namespace?

Alf

On Mon, 14 Jan 2019 at 18:03, Lizzi, Vincent
vincent.lizzi@xxxxxxxxxxxxxxxxxxxx <jats-list-service@xxxxxxxxxxxxxxxxxxxxxx>
wrote:

> Hi Laura,
>
>
>
> There might be a reasonable addition to the JATS tag set here.
>
>
>
> I have encountered a similar situation to the one that you described, in
> which there is a need to tag a link to a translation of the current article
> and the translation is not necessarily going to be located on the same
> platform. The solution at the time was to use xml:lang, although this is
> not a proper usage of xml:lang. For example:
>
>
>
> <related-article related-article-type="translated-article"
> xlink:href="10.0000/example" xml:lang="de"/>
>
>
>
> The XLink specification (https://www.w3.org/TR/xlink11/) provides an
> xlink:title attribute which bindicates a human-readable description of
the
> resourceb. The xlink:title attribute is intended to hold a description of
> the resource that is a target of the link, so it could be used to describe
> the language of the translation. For example:
>
>
>
> <related-article related-article-type="translated-article"
> xlink:href="10.0000/example" xlink:title="German translation"/>
>
>
>
> However, this would be more easily ready by humans than by machines. It
> would be possible to tag xlink:title="de" but this seems to be against the
> intended purpose of xlink:title to hold a human-readable description.
>
>
>
> There is no xlink:lang attribute, but such an attribute would be useful in
> this situation.
>
>
>
> The proposal would then be to add an attribute on <related-article>,
> possibly named something like ext-link-lang, to indicate the primary
> language of the related article using a machine readable value that
> conforms to the same rules as xml:lang.
>
>
>
> Thanks,
>
> Vincent
>
>
>
>
>
> *From:* Randall, Laura (NIH/NLM/NCBI) [E] laura.randall@xxxxxxx <
> jats-list-service@xxxxxxxxxxxxxxxxxxxxxx>
> *Sent:* Monday, January 14, 2019 9:12 AM
> *To:* jats-list@xxxxxxxxxxxxxxxxxxxxxx
> *Subject:* [jats-list] Re: @xml:lang on <related-article>
>
>
>
> Thanks, everyone, for the thoughtful comments.
>
> Mark, I appreciate you sending that resource and it very clearly states
> that what I was going to do is absolutely the wrong thing.
>
> To the folks who suggested I add the information about the target's
> language to an element within the related-article, I agree that that is a
> good solution, but only when there actually is content in the element.
> Related-article can be empty and, in my case, is.
>
> Gareth, your take that this should be a function of the delivery platform
> is actually why I'm running into this problem. In PMC, it is. When we build
> links between articles published in different languages, we'll pull the
> language information of the target from our database and create a link that
> includes the specific language information. So my trouble comes when we
> don't have the target article and thus don't know what the target language
> is. I had suggested to my group that we default to a generic phrase to
> replace the specific language ("another language"), but a colleague pointed
> out that that wasn't particularly helpful to the user.
>
> Aaannnddd...down the rabbit hole I went.
>
> I think Bruce brought up some good points and there are implications of
> JATS having some attributes everywhere that the Standing committee didn't
> necessarily anticipate. I think that in this specific case of having
> @xml:lang everywhere, we (the SC) might consider adding a reference to the
> W3C document to the non-normative documentation to clarify the use. But
> that's a suggestion for the committee, so I'll be submitting a comment
> requesting this.
>
> Again, thank you all for your comments.
>
> Laura
>
> ________________________
> Laura Randall
> laura.randall@xxxxxxx
> NCBI/NLM/NIH
>
> -----Original Message-----
> From: Randall, Laura (NIH/NLM/NCBI) [E] laura.randall@xxxxxxx <
> jats-list-service@xxxxxxxxxxxxxxxxxxxxxx>
> Sent: Thursday, January 10, 2019 8:40 AM
> To: jats-list@xxxxxxxxxxxxxxxxxxxxxx
> Subject: [jats-list] @xml:lang on <related-article>
>
> Hi, all.
> I have a situation where I have the English language version of an article
> that was also published in German. I want to include a related-article in
> the one I have that points to the German-language version of the article,
> so I'm using <related-article>. Along with the citation of the German
> article, I would *like* to be able to record its language so I can build a
> link that identifies the language as part of the citation (for user
> convenience).
>
> I'm wondering if I can use the xml:lang attribute on the related-article
> element to do this.
>
> Now before ya'll grab your pitchforks and yell 'TAG ABUSER', hear me out.
>
> Definition of related-article:
> Description of a journal article related to the content but published
> separately. May include a link to the related article.
>
> Definition of xml:lang:
> The language of the intellectual content of the element for which this is
> an attribute.
>
> So the official definition of related-article says it is a description of
> another journal article and xml:lang says it specifies the language of the
> intellectual content of the element. "The element" in this case describes
> the target article, so my possibly twisted (out of desperation this time)
> take on this is that in the same way the vol and page attributes describe
> the volume and page of the target, so, too, should the xml:lang describe
> the target.
>
> I'm like 99.99% positive that the Standing Committee didn't consider this
> use when they(we) added xml:lang everywhere, but I think this follows the
> rules and definitions.
>
> Does anyone have any thoughts on this?
>
> Thanks (and sorry),
> Laura
> ________________________
> Laura Randall
> laura.randall@xxxxxxx
> NCBI/NLM/NIH
> JATS-List info and archive <http://www.mulberrytech.com/JATS/JATS-List/>
> EasyUnsubscribe <http://lists.mulberrytech.com/unsub/jats-list/216958> (by
> email <>)