Resource wrangling : An implementation of Primo Resource Recommender service at State Library Victoria
Marcus Ferguson, SLV
Principles: find best way to present recommendations; control number of resources recommended; clearly identify subscription vs free. Include: databases, websites, research guides, custom types (collection pages, exhibition pages), ‘more to explore’ (originally things like library hours, now repurposed libguides for subjects) 488 resources – with 10,500+ tags. Maintaining this either through Back Office or spreadsheet upload was going to be difficult.
Built a spreadsheet with columns: list of keywords; database1..database5; website1..website3 etc; with dropdown menus to populate these. But then need to convert this. VLOOKUP wouldn’t work so needed custom function. Found a VBA function via Google. This operates on a new sheet to create a list of databases and all the tags used by it, plus a list of ‘other tags’ added manually for each one. Final sheet pulls it all together into the format Primo expects.
Finally also assigned icons to improve visual effect – found from vendor branding pages; website; or social media. Looks bad if a resource has none, so assign a default logo in that case.
Subscription database use ‘Login to access’ as URL text; free ones have title as URL text.
Added rr_database_icon_check as a keyword so can search for Primo for all of these and check that they’re still valid – mostly they’re pretty stable. If that changes, will grab them and store locally.
Final step is VBA macro to save export version and backup.
Looking forward: – need to assess impact of the May release “tag enrichment”; extend spreadsheet to include research guides; apply additional error checking; investigate ways to allow other librarians to work with the tags while managing change control.