what's a DL for?
Jun. 13th, 2006 04:43 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
deborahCarl Lagoze, on metadata aggregation and the NSDL experience:
Carl discussed problems with the great multiracial hope which is OAI-PMH harvesting to form union catalogues, and I completely agree with everything he said until his final point.
Firstly I'm not convinced I agree, and secondly I'm not sure that this is the digital library's job to provide these resources. Maybe I'm perpetuating a false dichotomy, but it seems to me that this job should be provided by teaching and learning software (or similar tools outside of universities), not by libraries, just as a brick-and-mortar library provides resources to support its users and, if its users are educators, they provide a framework for building relationships and annotations. To the user, perhaps, there should be no visible seams between the library and the annotation/relationship tools, but I'm not convinced that the library, per se, should be the back-end providing those tools. Carl said "If we're just into developing search and discovery, other people can do it better than we can", but I think he has it back to front. Organizing information so people can discover information is one of the things we've been doing forever; supporting a particular form of resource use is not necessarily something we do. We can (and in many cases we do -- in brick and mortars book clubs and reading groups come to mind), but it's less fundamental, IMNSHO, than resource storage and discovery.
Carl discussed problems with the great multiracial hope which is OAI-PMH harvesting to form union catalogues, and I completely agree with everything he said until his final point.
- "knowledge gap":
three things that are usually distinct- subject expertise
- metadata expertise
- technical expertise
three things that have helped- documentation
- hand-holding
- CWIS
- Harvested metadata are often not valuable
- OAI-PMH is difficult to understand and imperfect to implement (60% of harvests fail), therefore high human cost to maintain
- "A Digital Library is not made by metadata. That is, people don't just want to discover, they want user-based annotations and relationships." This, to me, is the controversial crux of his argument, and I'm frustrated that this was his closing point.
Firstly I'm not convinced I agree, and secondly I'm not sure that this is the digital library's job to provide these resources. Maybe I'm perpetuating a false dichotomy, but it seems to me that this job should be provided by teaching and learning software (or similar tools outside of universities), not by libraries, just as a brick-and-mortar library provides resources to support its users and, if its users are educators, they provide a framework for building relationships and annotations. To the user, perhaps, there should be no visible seams between the library and the annotation/relationship tools, but I'm not convinced that the library, per se, should be the back-end providing those tools. Carl said "If we're just into developing search and discovery, other people can do it better than we can", but I think he has it back to front. Organizing information so people can discover information is one of the things we've been doing forever; supporting a particular form of resource use is not necessarily something we do. We can (and in many cases we do -- in brick and mortars book clubs and reading groups come to mind), but it's less fundamental, IMNSHO, than resource storage and discovery.
