Review Comment:
The authors have revised Section 4 to add content in a part of the manuscript that is now named Sec. 4.1.1 showing that users are querying content in the OpenCitation's REST API in a way that has a high similarity score to user queries of the SPARQL API endpoint, i.e. users are querying similar things in both the REST APIs and the SPARQL APIs. This gives readers a sense that they are functionally equivalent and combined with the number of queries to the REST APIs and the SPARQL endpoint, they show there is potentially a shift towards using the REST APIs. This suggests a possible change in user behaviour in querying using RAMOSE REST APIs from SPARQL queries showing potential adoption of the REST APIs from users in the context of OpenCitations data reuse.
In closely revisiting the SWJ's definition of demonstrable impact (http://www.semantic-web-journal.net/faq#q20), an evaluation of RAMOSE's impact follows:
"(I) impact beyond your own range of influence (e..g., uptake of your work by other research groups)"
- There isn't sufficient evidence of this given as it is too early to expect widespread community uptake of RAMOSE at this moment in time. It appears there has not been additions in the manuscript on this issue
"(II) impact within your own range of influence (e.g., use of your results in a collaborative research project)"
- There is evidence in the manuscript to demonstrate some indirect impact, i.e. RAMOSE's key role in OpenCitations. Section 4.1.2 lists other endpoints using RAMOSE but are related to OpenCitations events. Sec. 4.1.2. demonstrates adoption of OpenCitations REST APIs in other tools. The outcome of third party tools accessing OpenCitations' REST APIs demonstrates indirect impact - as the RAMOSE tool was used to develop those REST APIs and access OpenCitation's content successfully. The examples provided for third party applications are are potentially impactful ones (e.g. a Zotero client for OpenCitations). However, there is a question about whether this could be achieved without RAMOSE, that is, developing a REST API without RAMOSE.
"(III) potential impact"
- Anecdotal evidence is provided that RDF data providers would require such a tool as RAMOSE for delivering RDF data via REST APIs
As noted in a previous review, tools like RAMOSE have an important contribution to the Semantic Web community as a tools that may allow RDF data to be more easily accessed. On the above basis, there is demonstrable impact on the Criterion (II) - using RAMOSE helped develop the OpenCitations REST API, which led to third party adoption of OpenCitations and benefits to the Open Science community.
Overall, this revision addresses previous comments (such as the comparison table in page 13). Some minor revisions suggested below.
Minor revision
- p.12 "SQARQL" -> SPARQL?
Section 4. Section headers 4.1.1 and 4.1.2 are added, i.e. sub-subsections. However, would it make more sense to have them as subsections, e.g. 4.1 and 4.2 respectively?
|