the Book Kismet project v 0.1

I’ve started working on a project called Book Kismet. The initial seed for the idea was this message I posted to a Google Group devoted to an innovative “social books” project housed at the University of Toronto, and it’s my hope that BK will wind up being a useful tool for them.

The functionality I want to see in version 0.1 is as follows:

user supplies author name, book title or ISBN

system provides list of ISBNs & titles from one of:

  • Amazon API
  • Google Book Search API
  • LibraryThing API
  • OpenBook API

user selects either author or book (by ISBN)

system provides (some of):

  • cover image via one of {Amazon, LT, GBS, Open}
  • link to Amazon detail page; link to Amazon reviews for boook
  • link to GBS title page; link to GBS reviews for book
  • link to LibraryThing detail page for this book (all ISBNs); link to LibraryThing book reviews
  • link to Google API answer set on “author lastname” “author firstname” “discussion forum” “booktitle”
  • embedded Google Trends widget with book title and author name as keywords
  • Google Groups (old Usenet) search results

etc.

Hello world! from Book Kismet

I have been thinking about the challenge of integrating physical and digital worlds via [social books] and I have a suggestion.  I think you need to be designing at a higher level of abstraction: really, designing a standard instead of a device.

Think about it this way.

Class of bibliographic entities:
  • printed book (i think “codex” is too jargony)
  • journal article
  • report
  • e-book

Class of physical Enablers

  • CueCat
  • Kindle 1 & 2, Kindle DX, etc.
  • Sony E-Reader

Class of entity-level citation schemes, e.g.

  • APA
  • Chicago Manual
  • BibTex
  • EndNote
  • Digital Object Identifiers

Class of “pinpoint” citation services, e.g.

  • Legal standards (West, F.2d 1033)
  • Scientific standards: Nature 355:321 12 October
  • Paragraph and line numbering schemes

Class of web resources

  • Single purpose websites (1 per book)
  • Google Book repository
  • Amazon catalog
  • LibraryThing

Class of web services

  • Annotation
  • Discussion
  • Recommending

Right now, we have a variety of entities pursuing efforts to connect all these classes with single threads e.g. Amazon connects e-book documents with Kindle with the Amazon catalog with recommending. Kindle is a closed system so that thread is the only you one can follow if you own the Kindle class of Enabler.  the proposed sBook would connect codex books using a custom-built Enabler with some undetermined citation format with purpose built websites and offer Discussion and Recommending services.

What is really needed, IMHO, is an open, platform-agnostic architecture that allows mix and match of all these classes. I believe Kindle is eventually going to be a limited success (not a failure, just a 10% of the market type thing) because it locks the reader into a single thread of classes. I’m more optimistic about Google Book Search because I think their physical enabler will be any device that can read a PDF and I think they will eventually ave a good citation standard and robust discussion services at GBS.