Home

Issues

Use xISBN or thingISBN to group together Manifestations at the Work level. Because we're identifying things by ISBN, we're not connecting different editions of the same book.
Textbook editions are especially crazy. University libraries lend a lot of textbooks. Publishers put out a new edition every couple of years. It makes for a whole especially confusing space. Maybe it would be solved by the previous thing, or maybe it would require special attention to work well for our students.
ISBNs don't cover everything. Move to use OCLC numbers or Libary of Congress Control Numbers would mean everything was covered.

Robustify it.
Unit tests!
Add good logging.
Return title/author in the results. (In the vast majority of cases all the contributing data would agree on a title for a book, so we could just let majority rule. We all use copy cataloguing and buy from the same sources, I think. We'd want to get the titles when the libaries are giving us the data because then we don't have to go to an outside source, like OCLC or the Open Library, 50,000 times a month. Same with authors.)

We want one line of Javascript (or two or three) that anyone with any OPAC can use to embed SPLURGE results. We need to build up a library of such simple plug-in solutions.
Not all books have covers, or they can't be found. Solution: Look in more places, or make fake little generic ones that show the title and author.
Make sure widget is WCAG 2.0 compliant and meets accessibility standards

Sort out how we want the data formatted when we collect it, and then get it and get it regularly.
Putting code under GPL.
Releasing data under an open data license.
How to make the data anonmymous?