A LibraryThing Blog posting informs me that recently their database contained 8,388,608 records – apparently this is half of 224, the largest number you can store in three bytes. It’s also the limit for MySQL’s “signed medium integer” (111111111111111111111111).  They have now rebuilt the database – and won’t hit another antibiotic barrier until they hold information on 8.4 billion books!

This is clearly a ‘cool’ service (it was also recently featured on Slashdot) . How did they succeed in being so popular?  What lessons can the JISC development community learn from the popularity of the service?  Any suggestions?