Hey thanks everybody, I've been too busy to dig into any of your
suggestions but hugely appreciated. This group is awesome.

@Amanda, I actually remember signing up for Small Demons in beta and it
died before I got a chance to really explore it.
@Thomas, LibraryThing's charactername field looks very promising if the
list consistently gives main characters first billing.
@Shaun Trajectory is definitely interesting, though I've not thought of a
use case yet.
@Karen true about the authority problem - unless publishers wrap this sort
of info in ebook metadata?
@Joshua Like LibraryThing, its unclear if the character lists are actually
prioritized by significance.
@Joel Shame those resources look rather dusty. As for an IMDB for books, I
think LibraryThing or Amazon are better positioned than anyone.
@Brooke I'm absolutely certain its doable, but as @Amy points out its a
pain in the ass. Even if I simply take @Alexander's suggestion of the Le
Monde list, I have to scrape and scan and scrub for something that, in a
world where we can have nice things, this already exists in a
rough-and-ready incomplete but off-the-shelf dataset. It kinda blows my
mind it doesn't.

Not to mention there's the other step I mentioned, which is matching them
up with Gutenberg.org pages.

I'll keep you guys updated as I dig into all your ideas. Cheers!

Dave



On Wed, Apr 15, 2015 at 4:17 AM, Thomas Guignard <[log in to unmask]>
wrote:

> The LibraryThing API could also be used to retrieve what they call "Common
> Knowledge" tags, including character names but also place names etc.
>
> Example:
>
> https://www.librarything.com/services/rest/1.1/?method=librarything.ck.getwork&id=2773690&apikey=d231aa37c9b4f5d304a60a3d0ad1dad4
> (using the example API key)
> Look for the "characternames" field.
>
> As far as I can tell, however, there is no way to determine which of the
> characters are the "lead male" and "lead female" character short of
> assuming that the top listed characters are in effect the lead ones. Also,
> the API calls are limited to 1000 a day. But maybe an avenue to consider.
>
> t.
>
> On Tue, Apr 14, 2015 at 2:15 PM, Shaun Ellis <[log in to unmask]> wrote:
>
> > Another interesting startup in this area is Trajectory.
> >
> > Here's a list of Classics/Fiction via their JSON API (doc=isbn):
> > http://api.trajectory.com/api/v1/search/?q=&c=Fiction%20%2F%
> > 20Classics&limit=568
> >
> > Here's a "human readable" view:
> > http://www.trajectory.com/search/?q=&facets&c=Fiction%
> > 20%2F%20Classics&limit=568
> >
> > -Shaun
> >
> >
> > On 4/14/15 11:07 AM, Amanda French wrote:
> >
> >> What you *did* need for this interesting project was Small Demons, which
> >> was a for-profit company that was creating linked data from books --
> here's
> >> an article about it: http://www.theverge.com/2013/
> >> 3/1/4043298/building-an-atlas-for-books-with-small-demons
> >>
> >> But it shut down in 2013, and I have no idea what happened to the data.
> >> It might all have been commercial and proprietary, anyway. Article on
> its
> >> closure: http://www.latimes.com/books/jacketcopy/la-et-jc-small-
> >> demons-to-close-unless-buyer-appears-20131106-story.html
> >>
> >> Amanda
> >>
> >>
> >> On 4/13/15 10:12 PM, davesgonechina wrote:
> >>
> >>> So I have this idea I'd like to do for a hobby project, but it requires
> >>> finding a table that lists a classic novel, a Gutenberg.org link to an
> >>>
> >>>  <snip>
> >>
> >>
>