using MAJAX and Google Book Classes in mashups

continuing the conference.... This presentation, by Annette Bailey of Virginia Tech, focuses on creating mashups using MAJAX (Millennium AJAX), software that enables mixing catalog data into web pages. This package is freely available from Virginia Tech and its use does not require support from III. She depicted the skill set required to employ MAJAX as relatively low, with knowledge of HTML sufficient.

Bailey is now showing a new books list from the VT Libs site, which includes status information extracted from Millennium; the creation of a list of books on a specific topic on a faculty member's page, including book covers extracted from Syndetics and clickable item status information from Millennium. Bailey noted that MAJAX uses caching and sends 1 request per unique item per page to the server, regardless of the number of span tags referencing an item. MAJAX does impose an additional load on the Millennium server; there is a configurable limit of 100 records per second per page. She noted that MAJAX works only on web pages from within the same domain as the Millennium system (would be *.wsu.edu in our case).

Browser testing: Works in Internet Explorer 6.x and Mozilla Firefox; Opera support is qualified.


-Installing: upload majax.js, majax.html to your catalog's /screens directories

-Employing: include tags, with the src attribute pointing to the majax.js file;
on web pages, use tags and MAJAX syntax

Bailey went on to describe some specific examples, such as the use of majax-harvard-reference, which extracts data and formats it as a bibliographic reference.

Going back to this morning's presentation on III development: It's hard to tell where the III business model ends and customer development begins. In this case, Virginia Tech is distributing a package that strongly supplements Millennium, possibly crowding out III from building a product that provides this support. At this moment, Bailey is going over the nuts and bolts of the MAJAX API (although she hasn't used that term), describing additional methods to produce a customized product from the software. As Bailey noted towards the conclusion, MAJAX "opens up the black box" of Millennium data for other uses.

---

Annette Bailey continued with a discussion on a different development effort involving the Google Book Search Book Viewability API. This API supports client-side access to a small set of Google Book Search information (examples, thumbnails of book covers, links to [previews of] full-text information that is available). She accomplished this using the Google Book Classes.

She provided an example: Searching (currently on the VT Millennium Web OPAC staging port) providing an embedded icon within bibliographic records, enabling users to click out from the catalog to Google Books. Here's some additional info.