Heard a lot about news, but there’s stuff around making TV programmes richer, more interesting. Quite often they want a list of books that are interesting or relevant. Sometimes they have a list, or they want people to contribute to a list. So where should people link to? Amazon is the de facto place to link to? Should the BBC link to Amazon? Probably not. Wikipedia would be nice, but if a book’s not there you could add it, if it was “notable”. There’s also OpenLibrary, which wants one page per book for every book ever made, but it seems early days. WorldCat is interesting, it tells you which library has the book near you, but mainly focused on US. There’s the Library of Congress, which has a lot of data. Then there’s LibraryThing and BookKeeper, which is about “my” collection of books, so people logged in could see which of their friends have also read that book.

On the music site, it’s quite clear, they link to official website or MySpace page. But for books there’s no clear source.

This is not about the text, unless it’s public domain and available, but about the metadata.

Link to the publisher. Author’s websites. Penguin tried to build book groups around their books, so there’s a page for each book and can do a book group about these things. Harper Collins have a network of interest that they’re building around, Authonomy, BookArmy.

Amazon are canonical, whether we like it or not. They have the metadata, synopsis, reviews etc.

Libraries have huge collection, but they don’t necessarily have the synopsis. They just have classification data.

Amazon have an API so could someone not just use that data?

Would it make sense for the BBC to make a page for every book it talks about?

How to you let users share their preferred link. Not sure if the BBC should be making their own library, given how many others are in that space. Should be partnering with another service.

Open Library website talks about how they might work with World Cat, so landscape may change in front of us. Shelfari, owned by Amazon.

Link to a variety of sites?

Every book as an ISBN, so perhaps use that as a gateway to other places that list books by ISBN.

Wikipedia has a page for every ISBN which links to pages that have ISBN built URLs.

Two ISBNs, the US and the Worldwide ones. Two standards. Library of Congress has its own identifiers which are broader than just books.

Do we need a way of marking up book titles, semantic web. Could you RDF8 tag to ISBN number in with the book, then later on you can scrape all those pages.

To make it more complex, ISBN refers to one edition to one book, but sometimes you want the whole work, rather than the edition. Need a way to pull that together.

This is why Amazon number, the ASPN, which shows you an individual book or collection of editions.

BBC policy about linking – what does the audience expect? Is it inline or in the sidebar? what are you saying? Here’s a place to buy it? Here’s a history of the book? Depends on what the audience expects. Technically, it’s easy but editorial problematic. Can’t be seen to be promoting Amazon. Would feel like that was unfairly promoting Amazon, but Amazon is the de facto place to link to for books.

Could link to something like Froogle, a list of places you can buy the book, but you’re pushing people to buy the book and promoting a site.

Amazon and Wikipedia are the main ones from an information point of view. BBC should work with open sources of data.

Think of the archive as a live thing where we add new things, like the list of books, that would be valuable. If the annotated book used by the researcher who put the programme together was made available that would be fascinating.

In Our Time, Melvyn Bragg, lots of notes from researcher but would need sanitising before it could be published because the notes are very detailed. After Our Time, wiki, set up but is a bit dead. If you had the research notes to go with the transcript, that would be great. Very thorough research.

What about other BBC sites, aren’t they doing already? Once you have an identifier per book in the system, you can aggregate content within the BBC site about “these are all the programmes or sites that are talking about this book” .

Something like Dewey Decimal system. BBC already has topics which functions a bit like that, if you have a page about the Cold War it aggregates all the BBC’s info about the Cold War, but should also link to thinks offline and books etc.

Namespace – Emma, the book, film, TV series, which one? Wikipedia deals with that ok.

Tom Coates did a lot of work around radio programme pages, analogous problem. When looking at a page on /programmes, the object is “episode” and it knows the broadcast of that episode. And with books sometimes you’re interested in a particular edition because of, say, the cover, and sometimes you want to talk about the text.

But that seems like an edge case, generally talk about a book, not an edition.

Some of the sites listed earlier are doing good work in aggregating them into works, not editions.

Books in translation with popular titles, e.g. Latin classics in translation, if you’re grouping the book in that how would you deal with versions? Publishing industry is struggling with that anyway – how do you deal with Ladybird version of the Three Musketeers, and the original version?

Should semantically mark things anyway, whatever is done.

What does the user expect? Do they want to go and buy it? Do they want to find out more? Do they already know the book and want supporting information?

Further reading is different to a bibliography. A bibliography would be relatively easy to do, but further reading would be something to let others do. Wiki, audience participation, which then leads on to gaming.

Technorati Tags: BeeBCamp2