The Local Index
Where could we go with CLOCK? That’s one of the questions we’re asking as we draw the project to a close. As mentioned in previous posts, our thoughts move towards a localized index. This would basically, rather than replicate all the information on the distributed databases, hold entries for each uniquely referenced bibilographic work. In a nutshell, it would contain rudimentary search (Title, ISBN, Author) content and specific location (URi) content. This would allow users to search for very specific records without having to span a search across multiple datasets, something our adventures in SPARQL proved to be quite tricky due to the weight of the data coming from these endpoints.
We’ve made a pretty diagram to show how this would work! (click on it for a bigger view)