CiTO + SWAN: The Web Semantics of Bibliographic Records, Citations, Evidence and Discourse Relationships
This is a revised version, now accepted, after an "accept with minor revisions". The reviews below are from the original submission.
Solicited review by John Westbrook:
The authors report on the harmonization of ontologies representing citation typing and discourse in the field of neuroscience. In describing their work the authors present a very high-level description of the importance of ontologies and semantic alignment, and then jump into the details of the particular semantic challenges in this domain. While the latter is the bulk of the hard work that has been done here, there is perhaps some room for additional examples to that have motivated this body of work.
The introductory section describing ontologies and importance of semantic harmonization is rather generic and not well associated to the topic at hand. One could easily begin this manuscript with the section that begins with "The ontology harmonization activity described in this paper…".
Rather than present this generic discussion, the manuscript would be strengthened with the inclusion of a broader set of use cases. For example, some of the following could be described:
• The mechanism for inking conventional and open source publications with blog and forum entities where a single URL may embody a range of opinions.
• Integration with emerging schemes for identifying individual contributors and researchers.
• The mechanics of handling the retraction of citations not accompanied by a body of literature describing the reason for the retraction.
Either the wrong version of Figure 7 is provided in the Conclusion section or this figure and its description add little to the discussion.
Solicited review by Cameron Neylon:
The effective harmonisation and integragtion of complementary and competitive ontologies is a key area for development if the promise of semantic web technologies in the sciences is to be realised. This paper describes the harmonisation of two suites of ontologies, both of which are gaining significant use, and which had the potential to become competing standards.
The paper offers two important contributions. The first is the harmonised suites of ontologies themselves, the second is a clearly described example of the process of such harmonisations. The process, challenges, and the conditions that made the harmonisation feasible are clearly described and this will be useful for future efforts.
The manuscript reads very clearly throughout and key issues are well described. A minor suggestion which may help the naive reader would be to signpost specific issues via subheadings. There are a number of key points made, distinction between citation as an act and the citation as the thing being cited, directionality of citation, importance of distinctions within FRBR and manifestations. My feeling is that incorporating those statements as subheadings could help guide the reader to the key points.
From my perspective I think it would have been helpful for one of "before and after" examples to appear in the main text. This might mean bringing the supplementary figure into the main text (the supplementary figure isn't included in the document I have reviewed) or alternately showing a before and after version of Figure 1. A visual representation of how this has changed and new connections have been forged would have been helpful for me to understand the changes.
Finally, it would be valuable, although I appreciate not straightforward, to see some more detail on the social processes by which agreement on issues was reached and in particular how disagreements were resolved. Was resolution through a purely consensus approach, reached by argumentation? Were rules of engagement or resolution agreed in advance? Are there suggestions on how best to approach harmonisation activities, particularly as the size of groups involved increases?