On 04-May-2012, at 6:10 AM, Karen Coyle wrote: > > > On 5/3/12 4:35 PM, Ben Companjen wrote: > >>> OL would need to have a hierarchical layer in between editions and the >>> files that represent volumes, but at the moment that layer doesn't >>> exist, and there isn't a way to associate files with volumes. >> >> Wait a minute... My statistics script caught a couple of OL IDs of >> volumes, e.g. /books/OL13439942M/v1. It redirects to >> <http://openlibrary.org/books/OL13439942M/The_ingenious_gentleman_Don_Quixote_of_La_Mancha>, >> which in the right column shows links to all separate volumes. So it >> is possible! It seems AnandBot may have linked the files to the record >> in 2009. >> The JSON view<http://openlibrary.org/books/OL13439942M.json> shows >> how these links are stored: under the key "volumes", as a list of >> /type/volume objects. >> It's likely then, that there are 26 editions with links to multiple volumes. > > That's a great find! Perhaps Anand can explain whether this could be > done over the rest of OL. It might fix a lot of errors.
Open Library has a field in edition records for storing ocaid, the IA identifier, that contains a single value. This created us two troubles. 1. If the same edition is scanned more than once, there is no way to represent it. We often end up creating duplicate editions with different ocaids, which is not really the right way. 2. The case of multi-volume scans, where an edition is split into multiple volumes and each is scanned separately. We've tried to handle that by using a "volumes" filed in edition records. This filed takes a list of values, each of type /type/volume. Each entry in volumes contains the volume number, label and the IA id. It didn't really work well and we stopped doing it. You can still see traces of it in some editions. Anand _______________________________________________ Ol-discuss mailing list Ol-discuss@archive.org http://mail.archive.org/cgi-bin/mailman/listinfo/ol-discuss To unsubscribe from this mailing list, send email to ol-discuss-unsubscr...@archive.org
