Closed Maelplaine closed 5 years ago
Looking at JATS4R, maybe should use the first matching:
<article>
<front>
<article-meta>
<pub-date date-type="original-publication" iso-8601-date="[something]">
[anything]
</pub-date>
</article-meta>
</front>
</article>
The above is now uncertain, as it sounds like JATS4R's recommendation is going to change. /cc @Melissa37
Yes, I don't think we have time to wait for the JATS4R recommendation on versioning to be finalised - could take months and months. I would stick with something PubMedCentral recommends, for now.
Having read the PMC guidance and JATS docs, going to suggest <pub-date date-type="pub" iso-8601-date="[something]">
. Though I'm not sure how common @iso-8601-date
is here, so might have to pick out <year>
, <month>
and <day>
too.
The date is now to be all uppercase. Decision made to do this mid March 2019. @chugginselifesciences
Going to implement <pub-date date-type="pub" iso-8601-date="[YYYY-MM-DD]">
, otherwise use <day>[DD]</day><month>[MM]</month><year>[YYYY]</year>
(and skip if it still doesn't match, so partial dates etc aren't supported).
Pattern library: https://github.com/libero/pattern-library/pull/75 Browser: https://github.com/libero/browser/pull/66 JATS support: https://github.com/libero/jats-support/pull/10
Problem / Motivation
- WHO reader, publisher - WHEN landing on an article page - WHERE article header, below the authors - WHY so people know it’s still relevant or new ## Assumptions - Current format is suitable for all the journals: Jul 24, 2018 - Localization format is Jul 24, 2018 but we should allow for other locales - The default time and timezone is 00:00:00UTC ## Clarification needed - Does the source of truth for the date comes from the XML content, not from the moment in which the - article is really published in the article store? Can both work? > Yes but it would need a validation to make sure the dates match Assuming that there is no versioning of article and assuming that the timezone of the publisher and system match - Can these include time? And timezone? > No it doesn't (eLife, Hindawi, RSC) - What dates do we need? eLife uses 3 (publication date, status date, version date) > For IJM use the date when published in the forthcoming article section > RSC displays the 'article received' date, 'accepted' date and 'first published' date > Hindawi: Submitted, Revised, Accepted, Published - Are these Lax dates? publication date, status date, version date > TECH TEAM? ## Technical notes - Dates should be stored as timestamps (relevant database types) - Display should take (one of the) journal’s defined locale(s), not user’s. - Should be stored in UTC, and timezone (if required) stored separately ## User interface / Wireframes