Ah, the metadata dilemma. Too much is too expensive while too little makes the whole project pointless since no one will find it. For digital libraries in general I think this is an area that is both art and science. For my digital collection in particular, the potential audience and known creators are who I have in mind when I am experimenting with cataloging. What I mean by that is I envision my collection of webcomics as being of popular culture interest and not for a specialized field or academia. For general users, who are comfortable online, traditional subject terms are not adequate as they can be old fashioned or non-intuitive. Because of that I have been playing mainly with key words and tags. Neither are perfect. Key words have potential as being natural language based and a well known search method. I think it is the system most users will be comfortable with. I personally like tagging as method of description but without high user involvement and/or collection density it doesn’t necessarily work well. In both cases, consistency is dependent on me (the administrator) paying attention and keeping track of what I had chosen in previous cases. The only way around this problem I can think of is to include decision making for terminology in the planning stages. And then hope one is prescient enough to cast the net wide enough to give full coverage.
No comments:
Post a Comment