Hello, I don’t know if this idea requires its own talk session or if it could be folded into some of the others, but here goes. I’m currently a cataloging librarian, so naturally my focus is on bibliographic metadata. Since I work in an art library, I catalog a lot of art books, but I sometimes wonder if the cataloging I do is sufficient. In particular, I’ve been wondering if art researchers would appreciate knowing what works of art are reproduced or described in a given publication. This would probably require more time and effort than most cataloging outfits can afford, but I’m hoping that current technology might reduce some of that to manageable levels. A couple ideas:

• Scanning lists of illustrations/plates into PDFs that can be linked from the catalog record, the way some libraries currently do with tables of contents
• Using trusted sources of (linked) open data, such as CONA or museum websites, to identify works of art and possibly even verify/record their publication histories

Of course, there are potential issues with this. Simply digitizing published information may not be sufficient; for example, how would one account for artworks that are untitled, retitled, or even mis-titled? Some form of authority control might help, but that would require even more time and research on the part of the cataloger; imagine having to do that for 10, 20, or 100+ works in a single monograph. Finally, where would this information—scans, metadata, etc.—be stored/kept? For now, though, the foremost question in my mind is the one I started with: does anyone, aside from me, want to know what artworks are in what books?