James Phillips presents the origins of NoSQL, followed by a comparison of various NoSQL solutions and ending with an architect’s view of Couchbase.
James Phillips is co-founder of Couchbase. In 1984, at age 17, he co-founded his first software company, Fifth Generation Systems – acquired by Symantec in 1993 forming the foundation of Symantec's PC backup software business. Immediately prior to Couchbase, James was co-founder and CEO of Akimbi Systems, a virtualization software company acquired by VMware in 2006.
Software is changing the world; QCon aims to empower software development by facilitating the spread of knowledge and innovation in the enterprise software development community; to achieve this, QCon is organized as a practitioner-driven conference designed for people influencing innovation in their teams: team leads, architects, project managers, engineering directors.
NoSQL and elastic scalability
As a user of Redis, I think it's a sad truth that indeed all the other NoSQL solutions you mention, Redis included, are not dynamically scalable. Redis does not yet have a working clustering feature and so is effectively limited to the memory of a single node; Cassandra (from what I've heard) has a fidgety non-transparent ring-based clustering model; and Neo4j requires complex sharding.
I personally use Redis for its simplicity and for me Couch/Mongo is not an optimal solution. However I think the "dynamic elasticity" you mention is a very compelling feature, for me it's the basic promise of NoSQL, that we should be able to scale easily to support any load. It may be that this is more difficult to a achieve in a non-document model, but IMO it's high time that vendors in the NoSQL world rise to the challenge of providing transparent scalability, however their data is structured.