By the end of the day (Friday), I hope to have some demos of how amazon content can be transformed into formats of interest to me:  RSS, OPML, MOA2, and IMS-CP.  I hope to be able to use nothing more than XSLT and a number of web services that others are making available, such as the w3c XSLT service and the W3C Tidy service (which converts HTML to XHTML).

Some progress:  here's pictures from a keyword search on "Einstein" in the books section of amazon converted to a MOA2 book, rendered in the viewer.  Details to follow on how I did it.

I can now dynamically generate MOA2 objects for an amazon search.  It's awkward but seems to work -- using just the tools available on the Web.

Consider a search for Einstein.  You can the "einstein" in the URL with your desired term.  If you want to have a space, you must use %2520 instead.  So, a search for "Johann Sebastian Bach" would require "Johann%2520Sebastian%2520Bach".  Such a replacement would yield an amazon search for Johann Sebastian Bach.

Obviously, this is very awkward -- but it works.

(What I'm doing:  I've written some XSLT to do a minimal conversion of amazon XML to the MOA2 format -- and fed this output to the MOA2 viewer.  More details to follow.)