First things first, a hearty thanks go to Tom Armitage for his superb engineering and data wrangling skills. The spelunker would not exist without you, Tom.
Second, in addition to the randomizer-by-type display we made for the home page, the spelunker has three main views:
The Date Graph is a visualization of just three fields in the collection database:
museum_number. The yellow bars that show representative
year_end dates for when each object was created. The yellow bars are given a "completeness" score too, which simple shows a simple scale of how many fields in that record are complete, or not. The brighter the row, the more complete the record. The blue map on the right shows the acquisition year for each object. If you see a red notch in that column, that means we weren't able to surface a date from the acquisition column that looked like a year. You can also use the form up the top to constrain the graph to show objects between specific years. 2000-2014 is a fun range, for example.
The Browse by Facets is also just what it says on the tin. The collections database has a few fields that are approachable to normal humans, like material, location in the museum, or place in the world. We've just made these things filters for the Big Dumb List so you can see stuff like about 400,000 objects are currently in storage, or 157 things from California.
The Big Dumb List is just that. The very first step in understanding any big database, just showing everything, and making simple ways to navigate around in it so you can quickly get a feel for its scale, and, especially, interconnection.
Also, please note: This site has not particularly been optimised. It's rough. Some of the pages take a while to load, especially the Date Graph. Your patience is welcome.
Thanks must also go to the V&A for making its catalogue metadata available. The V&A API has been online for some years now, and it's robust and well-documented.