Peter Brantley was at an Open Content Alliance conference last week, and in How to Build an Online Library, his comments on the conference, he says, “Ideally, one would create a works level database based on FRBR.”
As discussed earlier, the Perseus Project has this, but it would be very useful if similar sites did too, such as Project Gutenberg. There, for example, there’s no relationship shown between The Adventure of the Dying Detective by Sir Arthur Conan Doyle as a text file and as an audio file. One work, one expression, two manifestations.
(There is a relationship noted in the audio version, which notes the source and gives the ID number of the story being read. It’s read by a computer, so don’t rush to give it a listen if you can read it instead.)