Simplify affiliations

[JGilbert-eLife]

I wonder if the affiliation divisions are over complex - what is the use case for this information? Why are we still giving space for fax? Why would we want to record complements?

I think there probably should be JATS4R input on this section with respect to the ideal set of information recorded for an affiliation.

[michael]

I also thought of just specifying address and leave that free form (multi-line). We can have additional (redundant) fields for city, and country, so this information can be used for searching and for rendering the affiliation as a string.

Thoughts?

[fabiobatalha]

@michael

For us the Institution name, Orgdiv1-3, city, state, and country are relevant elements and should be able to be identified separately. The other fields are not important, they can be collapsed into address.

Maybe website can have also a exclusive element, but nowadays it is not used in Érudit.

[lpanebr]

I know Scielo Brasil also requires orgdivs and country[1] for every aff WITH the ISO country code, and if city and state exist they should be captured as named-content.[2]

...
<aff id="aff01">
    <label>1</label>
    <institution content-type="orgname">Fundação Oswaldo Cruz</institution>
    <institution content-type="orgdiv1">Escola Nacional de Saúde Pública Sérgio Arouca</institution>
    <institution content-type="orgdiv2">Centro de Estudos da Saúde do Trabalhador e Ecologia Humana</institution>
    <addr-line>
        <named-content content-type="city">Manguinhos</named-content>
        <named-content content-type="state">RJ</named-content>
    </addr-line>
    <country country="BR">Brasil</country>
    <email>maurosilva@foo.com</email>
    <institution content-type="original">Prof. da Fundação Oswaldo Cruz; da Escola Nacional de Saúde Pública Sérgio Arouca, do Centro de Estudos da Saúde do Trabalhador e Ecologia Humana. RJ - Manguinhos / Brasil. maurosilva@foo.com </institution>
</aff>
...

[1] http://docs.scielo.org/projects/scielo-publishing-schema/pt_BR/1.8-branch/tagset/elemento-country.html [2] http://docs.scielo.org/projects/scielo-publishing-schema/pt_BR/1.8-branch/tagset/elemento-aff.html#elemento-aff

[Melissa37]

Aside: I notice you use this: <named-content content-type="city">Manguinhos</named-content> So does eLife. However, JATS has a <city> tag and the recent JATS4R author and affiliations group leader was surprised when I requested greater detail in their proposal around granular tagging. She was really confused as to why we did that and we should just use <city>. I cannot remember now. why we made that decision. Can I ask why you do that?

[fabiobatalha]

As far as I Know, this was a previous requirement from SciELO once they manage this information in this way.

The SciELO Publishing Schema was created a few years ago. I’m not sure but probably it should be the way to do that in JATS on that time. To assure that I need to check JATS 1.0 or previous...

I fill comfortable to change it to city and state now. On Wed, 7 Nov 2018 at 09:01 Melissa Harrison notifications@github.com wrote:

Aside: I notice you use this:

Manguinhos

So does eLife. However, JATS has a tag and the recent JATS4R author and affiliations group leader was surprised when I requested greater detail in their proposal around granular tagging. She was really confused as to why we did that and we should just use . I cannot remember now. why we made that decision. Can I ask why you do that?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/substance/texture/issues/993#issuecomment-436631986, or mute the thread https://github.com/notifications/unsubscribe-auth/AAexTuZedywvuB_XzGueLCm6Mv3J_WMkks5usueqgaJpZM4X1y8o .

[lpanebr]

@fabiobatalha is right. Scielo guidelines are based on JATS 1.0 were tags for city and state did not exist.

[Melissa37]

@fabiobatalha and @lpanebr thanks for the insights! I'll work on eLife updating our xml specs to use <city> instead