Wheat Publishing
Instant Open Source

When you couple Wheat's "every object has a URL" and "every object can be published on a web site" with the environment being a collaborative web site itself, you get a qualitatively different way to publish code.

As Art Cabral said: "...the heuristics inherent in World Wide Wheat are built over time. Every routine that is written adds to the collective intelligence of what can be created next..."

Or as I say: "When you code on a Wheat Server, you code for the world"

The experience of using open source for many systems is tedious at best. With all due respect, using CPAN is a chore.

Wheat makes it almost trivial: Carol publishes an library to manage calendars on her Wheat server. Doug can use it by simply making instances of it on his server. (The object's parent pointer is just a URL to the classes on Carol's server.) Carol can publish version 1, version 2 and the bleeding edge version. Doug can choose whichever he wants. During development, he'll automatically get bug fixes as Carol publishes them. When he's ready to deploy he'll grab a copy to keep locally.