stephenpiment
stephenpiment on Hacker News
Author of the essay here.
I've found that Snow Crash is one of those books that pretty strongly divides readers. Some people are very fond of it; others bounce off of it, often commenting that something about it rubs them the wrong way. Not many readers are in the middle.
I'm definitely in the former camp, even though I can see the book's weaknesses from a writing perspective. In the essay, I wanted to express some thoughts on why the book's fans respond to it, to the point that it has been rather influential in the world beyond fiction.
Honestly, I don't know that CL was ever exactly beloved among Lispers. (Scheme may fit that description better). CL was a pragmatic construct from the beginning, responding to many conflicting demands.
I have to agree with the other comments that you seem to have prematurely narrowed your options. The requirements that you mention are pretty broad and don't really pick out Riak or Cassandra in particular.
You may want to ask, What are your requirements for data consistency? Will you need to build reliable abstractions on top of your data store? If so, you may want to look at other options, such as FoundationDB.
Stephen Pimentel, foundationdb.com
I think it's great that someone is taking a detailed look at distinct isolation levels for different systems. It's also worth looking at distinct levels for atomicity, consistency, and durability. One could then define an overall "pH" level for systems.
I would take exception to Bailis' statement that "it is impossible to simultaneously achieve high availability and serializability." At FoundationDB, we do exactly that.
Stephen Pimentel
foundationdb.com
Clearly, there are different usage regimes where different solutions will make sense. Nonetheless, there's a really strong case to be made that SSDs have entered a sweet spot in terms of price/performance for databases, and this trend is only accelerating. Here's one discussion of the rationale:
http://www.foundationdb.com/#SSDs.