 |
|
|
A lot of effort has been expended on developing metadata standards for particular communities of users. These are generally suitable for their purpose. Some effort has been put into defining metadata standards that can be used across a wide range of communities. However, these are generally less successful in that something that is supposed to serve many rarely has a well defined area of application, leading to items being missed out that are important to some users. For example, one metadata system I have some knowledge of is meant for geospatial data. I deal with atmospheric data a lot, where information on the height that the data applies to is of paramount importance. Guess what is missing from the geospatial metadata system I mentioned earlier? This is a simple example. There are more esoteric ones dealing with space plasma data and astronomical data. So, I ask the question, "just how practical is it to attempt to define a single metadata standard?" My own feeling is that it is too difficult, but I would love to be proven wrong. If we accept that it is too difficult to define a single metadata standard, how many different ones do we need? Can we develop hierarchical systems instead where items at the top level are common across all areas with only specific branches having more detailed items defined? |