[Gsoc] map - reduce on zope data

[Tim Nash]

The proposal is to find a way to run distributed map/reduce on data
that is primarily published by zope. This could be by running
map/reduce on xml files stored using localfs or on objects stored in
the zodb.  There is a java map/reduce project called hadoop that is
getting very usable.  A student may want to come up with an efficient
way for zope to utilize the hadoop library. If the zope 3 on jython
project goes ahead without the ZODB, maybe they could use zope to
publish files stored in the hadoop file system. That could make a base
for a very scalable zope.

Note: I cannot mentor this proposal, and it may not actually be useful
to the average zope site.

Thanks!
Tim Nash