Tom and Liddy discussed the problems with the adaptability term on the phone. This is what seemed (to Liddy) to be the core of what Tom was worrying about.
It should first be said that although Tom seemed to indicate that this was the problem Andy and Pete were having with the work, I find no evidence of Andy refering to it anywhere. I suspect it is what was of concern to Pete when we discussed the issues in Madrid, but at that time I was not really getting the message!
So, the problem!
Describing resources for adaptability is tricky because it involves an iterative process in which resources may have a sort of gross description but then, if people do a really good job, individual components will also be described and the process of matching the resource components to the needs and preferences of users can be undertaken. In fact, it is very important to those who are providing accessible resources to their clientele that resources are described appropriately as they will depend on the metadata descriptions of available resources, so it is important that DC describers of resources provide suitable metadata.
So, the problem is how to do this.
First, there is a proposal for a term that would be used to describe the adaptability properties of resources that are not suitably described using other established DC terms. The characteristics (properties) fit into four categories, we think: access mode, role, flexibility and support tools. They are all properties of a resource that are relevant to its adaptability.
Then, there is recognition that some content developers or metadata creators know a lot about accessibility or use tools to assess accessibility. Such people are concerned with the adaptability of the resource for what are sometimes called 'accessibility' purposes - they are just one class of people who are expected to use the adaptability term. They are not concerned with 'access' as understood by librarians.
This means that sometimes the adaptability of a resource will be described using the adaptability term, and sometimes there will be further, more detailed, or just additional descriptions somewhere else in a separate record (or records).
So the problem is: how do these two sets of descriptions get to be integrated so they can be used as one? It is assumed that the content of both descriptions will look the same and be in the same syntax but just that it will be in different places.
The problem is not that there will be a description of the adaptability of a resource and then a description of the description of the adaptability of the term, as in the case where there is a creator property with the value John Smith and then more metadata about John Smith. In that case, the second lot of metadata is not about the resource but is about John Smith. If it were the same for adaptability, it would have to be that there was a description of the resource, and then, as already said, a description of that description. This is not contemplated!!
The problem is then, it seems, what about if the property adaptability is some string of characters that describes some adaptability properties and there is also a DOI on the Web that contains a lot more information of the same kind? Currently, the proposal is that the string will be in the metadata of the resource for the adaptability property and the other DOI will be associated with the RESOURCE by being identified as the value of a property 'hasAdaptabilityStatement'. This does not seem right. The DOI needs to be identified as metadata and associated with the resource in a more appropriate way.
This does not strike me as a problem that has anything to do with the metadata involved being about adaptability and I cannot imagine why the solution to this problem is not the same as the solution to the same situation in many other cases?
I can see that we should not use the hasAdaptabilityStatement in this way, and will work to fix that. I cannot see that it is an issue that makes any difference to what we are trying to do with the adaptability term???
Please clarify!!!