Relic is a web mash-up that retrieves and presents data from a range of different information services. It is designed as a utility to assist selection and acquisitions processes for Resource List materials.

It works as follows:

An ISBN-13 or ISBN-10 is submitted as a search to Relic.

Any hyphens or spaces are removed from the search string and it is validated (ensuring that the 'ISBN' contains only numbers, and is either 10 or 13 characters in length).

The ISBN is submitted to the Open Library catalogue, and (if a match is found) metadata is returned and and displayed in the search output (along with a medium sized cover, if one is available). Where the Open Library record contains these values, the metadata will include deep-links to the catalogue record for this ISBN in Open Library, the Library of Congress and LibraryThing.

Preconfigured search links are then built dynamically for Coutts Oasis, Amazon.co.uk and AbeBooks (offering both ISBN and Title searches) and displayed in the search output.

The ISBN is then checked against the existing index of ISBNs in the Resource List Management System (Talis Aspire). For each list that the ISBN appears on, the list name will be returned and displayed in the search output (which forms a clickable deep-link to that list in the RLMS).

Next, the ISBN is checked against the index of ISBNs in the Library Catalogue. If there is a match for that ISBN, metadata from the catalogue record is returned and live holdings data. In addition, deep-links to the Full View of Record and Holdings screen are configured and displayed. An option to 'search for other editions' (by Title phrase) is also composed and displayed.

Finally, the ISBN is submitted to the LibraryThing ISBN service, which returns a list of related ISBN-10 or ISBN-13 values which are then displayed in the search output. Clicking on one of the ISBNs submits that related ISBN as a new Relic search string, re-running each of the searches afresh against that alternate ISBN.

Fixes

The following bug fixes are planned:

• Error handling: Error handling for malformed responses or search time-outs needs to be improved
• ISBN-10 X: Validation rules for ISBN search values need to accommodate 'X' and 'x' values in positon 10 in ISBN-10
• Title clean-up: title strings need to be more extensively cleaned-up prior to the formatting of the search URLs

Enhancements

The following potential enhancements are under review:

• Alternate metadata source: where a search against Open Library returns no response, metadata could be requested from an alternative source before progressing
• ISBN only searches: A 'no match' response from Open Library removes the vendor search option: ISBN only searches could be available even when no matching metadata is returned
• Google Books preview: An additional Dynamic Search look-up on Google Books could be configured to enable pre-checked preview options through Google Books
• xISBN: The option to integrate OCLC's xISBN as an alternative to LibraryThing could be explored. The xISBN service returns additional data (such as edition information) which might improve the efficiency of related ISBN review
• ISBN or: The option to string boolean 'or' ISBN requests could be reviewed (particularly 'ISBN-10 or ISBN-13' searches on the Library Catalogue)
• Vendors: Deep-links to additional vendor and suppliers could be added; vendor web services for metadata retrieval could be explored
• DOI and LCN: additional search options for DOI and LCN values could be integrated - LCN initiated searches would rejoin a similar search pattern to ISBN initiated ones; DOI searches would not require Library Catalogue or vendor look-ups, but could trigger CrossRef API metadata retrieval