When OUP was first developing its data model, the digital content was created by converting scholarly monographs that had already been published in print. The intention was to sell the online content as a way of making these books more widely available.
The online version had to be as faithful as possible to the original print version. This
meant that bibliographic references had to be captured as plain text, as submitted by the
author. Each reference was wrapped in a <bibItem> element.
Having the references link to library catalogues (via OpenURL[8]) was
desirable, so we chose to identify the components of a reference by copying them into
attribute values on the <bibItem> element.
OpenURLs and other search links are generated on the platform by parsing these attributes.
This model also lets us add value to abbreviated references (e.g. “Ibid.”)
Example 12. OUP data model for a book reference
(Using Vancouver style for the text of the reference.)
<bibItem
class="book"
author="Lalloo F|Kerr B|Friedman J|Evans G"
title="Risk assessment and management in cancer genetics"
date="2005"
place="Oxford"
publisher="Oxford University Press"
>
Lalloo F, Kerr B, Friedman J, Evans G.
Risk assessment and management in cancer genetics.
Oxford: Oxford University Press; 2005
</bibItem>Example 13. OUP data model for a journal article reference
(Using Vancouver style for the text of the reference.)
<bibItem
class="journalArticle"
author="Salomone A|Bozzo A|Di Corcia D|Gerace E|Vincenti M"
title="Occupational exposure to alcohol-based hand sanitizers:
the diagnostic role of alcohol biomarkers in hair"
date="2018"
vol="42"
journalIssue="3"
page="157"
pageLast="162"
doiTarget="10.1093/jat/bkx094"
>
Salomone A, Bozzo A, Di Corcia D, Gerace E, Vincenti M.
Occupational exposure to alcohol-based hand sanitizers:
the diagnostic role of alcohol biomarkers in hair.
J Anal Toxicol. 2018 Apr 1;42(3):157–162,
doi:10.1093/jat/bkx094
</bibItem>