An intelligent authoring model for subsidiary legislation and regulatory instrument drafting within construction and engineering industry

• We provide a basis for improving the representation and authoring of legal documents.
• We propose an intelligent XML authoring workflow model for regulatory documents.
• We show that a user-oriented authoring model produces high quality, linked datasets.

Of primary importance within the domain of open data and more specifically open legislation, lies the essential central requirement for data to be available in a user-oriented manner; whereby the public and professionals alike can consume, share, reproduce it upon request and utilise it when time demands. Focusing specifically on subsidiary legislation (SL), current drafting workflows fall far short of addressing this vision. Whilst a significant amount of recent research has focused on less technical issues such as the actual definition of open data itself, the case for open data as to its management within a legal context, etc. due to the domain specific nature of producing robust and technically accurate open datasets, little work has been done on techniques for the drafting of legislative resources (in particular SL) as open data. To address this problem our work exercises a use-case driven from the domain of sustainable design and construction. As a validation vehicle, we select Scottish building regulations, which govern 32 local authorities across the country as a typical example of such legislation. Our work focusses on three areas of particular importance, (i) observing ongoing practice within the government organisations responsible for the drafting and publication of the aforementioned texts, (ii) understanding the means and methods utilised by local authorities which have been tasked with ensuring that standards of compliance are met within all cases of design and construction, and (iii) reviewing and understanding how construction stakeholders actually execute their activities with respect to the texts in question. The outcome of our study has resulted in a methodology and subsequent production of an intelligent XML authoring workflow model (DROID-SL) for such documents which displays how legal texts of this nature can be better consumed within our society. We demonstrate that by adopting a user-oriented drafting vision, it is possible to produce high quality, user-oriented, linked open datasets which wholly embrace the fast moving area of open legislation.