[Chicago] Google App engine is python powered

[Pete]

On Apr 8, 2008, at 4:35 PM, Garrett Smith wrote:
>
> Google's restricting data persistence because they need to replicate  
> across servers in multiple locations. All of the distributed  
> databases I've seen (Couch, Mnesia, SimpleDB) compromise consistency  
> at some level to support distributed, non-blocking writes. If you  
> want to avoid the availability and scale problems of a single master/ 
> slave cluster, you'll need to venture out into the wilderness of  
> these strange data stores.

Another one to add to your list is my project GrassyKnoll: http://grassyknoll.googlecode.com

It's got a very similar model to the above.  It's distinguished by:

  * Python, Python, Python!
  * open source
  * choice of persistence layer - lucene, sqlite, gdbm, memcached, S3,  
metakit
  * JSON
  * RESTful HTTP throughout - the same code is used for client/server  
& internal replication
  * younger and needing assistance
  * did I mention Python?

> Where's *your* CouchDB t-shirt, loser.

New policy: All GrassyKnoll contributors get a free t-shirt!*

--Pete

* T-shirts may or may not: mention GrassyKnoll, be the correct size,  
or be clean.

-- 
Pete
pfein at pobox.com