So what is new? Did they manage to figure out why it cannot work under a heavy load? It that Cassandra's or general Java issues?
In fact, many of these database implementations produced by NoSQL movement aren't anti-SQL, they're anti-MySQL; and the authors don't realize it as they've used PostgreSQL or Oracle in depth.
I seriously doubt this is a Java issue, and it might not be a Cassandra issue either... it could very well just be a an issue with how Digg is coded.
However, the key thing to me was that they were fixing break the world bugs a few days before the release, which tells us their engineering and release management was horribly broken. I don't see how we can blame Java, Cassandra or their modifications to the latter when they ignored clear signals their new total rewrite wasn't ready for prime time.
This is a Digg Fail and not a Cassandra Fail.