Been finishing of the repository report this week and reflecting on the processes that they have highlighted. The predominant thing that was missing for me was the lack of personalisation; I think I have a thing about this. Even when the repository holds author details or produces personal publications lists, they don’t utilise that information in metadata or search. Granted entering information from login details, name ad affiliations was fairly common, but none presented subject lists in the order of an individuals interests. This could be done as a simple count of times selected or derived more sophisticatedly form the content of the documents submitted. Repositories are in a unique position when compared to the rest of the Internets search and discovery functions, in that a large percentage of their end users are probably their authors. Hence they hold details on them that are all but impossible to get out of a guest user. Would be interesting to know if there are any stats on author repository use.
The other thing that bothered me was that they don’t appear to be looking closely at what current content management systems are offering. These are repositories but with a more personal and group focus. Handling any formats, and leaving content control mostly with the author. Often they link into other applications and acting as a storage and tracking system for all a users digital output. Anyway that’s all covered in my report, which is less opinionated, I think!
