This section makes some fairly high level assumptions about user understanding of HTTP.
Such interactions with the data would depend largely upon the nature of the applications built around it.
"The openness of data is more an opportunity than a threat. One benefit may be a clarification of the licensing of descriptive metadata towards openness, thus facilitating the reusing and sharing of data and improving institutional visibility."
Open licenses aren't a benefit but an aspect of open data.
The term open is used here for the first time in the report. Its meaning should be made clear, e.g. by means of linking to a section where it is defined.
The distinction between “information produced or curated by libraries that describes resources or aids their discovery” and “data used primarily for library-management purposes” isn’t clear.
It’s clear that e.g. holdings information and user data are omitted. But what about circulation data for a resource, the number and frequency of lendings and co-occurence with lendings of other books? This can aid the discovery of resources through suggestions like “People who borrowed resource A also borrowed resource B, C and D” and thus might be covered by the report (but I think it isn’t).
The distinction between "information produced or curated by libraries that describes resources or aids their discovery" and "data used primarily for library-management purposes" isn't clear.
It's clear that e.g. holdings information and user data are omitted. But what about circulation data for a resource, the number and frequency of lendings and co-occurence with lendings of other books? This can aid the discovery of resources through suggestions like "People who borrowed resource A also borrowed resource B, C and D" and thus might be covered by the report (but I think it isn't).
The top down model has been driven by economics: we can't afford to provide the granularity that would serve the customers best. The potential to invert this is one of the exciting possibilities offered by Linked Data.
“Links can be used to expand indexes much easier than required for todays federated searching, and can offer users a nearly unlimited number of pathways for browsing”
Not sure what this sentence means.
It would be more accurate to characterize these as potential benefits, as they are still to be demonstrated on a meaningful scale.
The analogy of “stone soup” is inappropriate. It implies that libraries have nothing of value to contribute. Whereas library metadata is valuable information that is rich in relationships. Expressing this data as linked data is not trivial as it was not designed for this purpose.
Not exactly sure who the audience for this report is likely to be. If it is general library administrators, they are not likely to understand the term "global graph" unless it is defined here.
This whole section needs to make an even stronger case for why sharing library data as linked data is vitally important for the future of libraries, as well as how library data can contribute valuable data to the semantic web. I would suggest expanding this section and making it as persuasive as possible.
This seems to be lacking something here - add an explanation about how libraries have valuable skills and experience in creating and maintaining authority data and how those activities are important for linked data.
This is a thorough, well-researched, report on a topic that is very important to the future of libraries. I would like to see the report make the strongest possible case for moving forward on its various recommendations for library linked data, while at the same time not minimizing the obstacles that will need to be overcome to make such initiatives successful. To that end, I will contribute suggestions throughout the report that I believe will help to make it as persuasive as possible for supporting library linked data.
Jennifer Bowen
University of Rochester
eXtensible Catalog Organization
"since it uses HTTP, the Web's standard retrieval protocol." - Just noting a typo, in the lack of the possessive apostrophe. Please excuse me if this is out of scope.
Comments by Section