Ideas and code for a better Internet

The Premise

More and more our lives are dependent on web services run by large corporations like Google and Facebook. What we read and watch, what we buy, and how we do business is determined in large part by self-interested entities.

Can we apply the principles of openness, collaboration, and decentralization upon which Linux, Wikipedia, and Bitcoin were built to create a alternative platform — one that guarantees security, privacy, and user control?

Current Projects

A cloud for the people, by the people, & of the people.

In which we own and control the web services that our lives depend on.

A cloud platform for hosting open source web applications — like Amazon's or Google's but cooperatively run — with built-in data sharing and a participatory funding model.

Unfurl uses Git, Ansible, and Terraform to record and deploy changes to your DevOps infrastructure.

Unfurl organizes this history into "ensembles", which represent instances of running services.

Ensembles can be started, stopped, backed up, restored, migrated, cloned and forked.

They are designed to be the core building blocks of an open, reproducible, and decentralized cloud infrastructure.

Older Projects

Vesper is open source framework for web application development that enables content from disparate sources to be replicated and recombined in a consistent and composable manner.


Vesper provides a data access layer that lets applications update and query a variety of datastores (including SQL, NOSQL, RDF, plain text) as JSON. It includes a JSON query language that is designed to be translated to the native datastore’s query language. Its’ JSON model is compatible with the Semantic Web/Linked Data model without requiring developers to have familiarity with that technology.


Vesper can maintain change history and metadata contexts and optionally store it separate from the primary store. This enables data to be replicated and merged intelligently, similar to distributed revision control system like Git.


Putting aside all the above, we just want develop web applications rapidly and agily and Vesper’s JSON-centric model helps us do that by greatly simplifying data management.

Learn more about Vesper here or get the code here.

Rhizome is the ancestor of the Vesper technology. First released in 2003, it was arguably the first semantic wiki. This project has been dormant for a long time but many of the ideas expressed in the papers below live on in the Vesper design.

Rhizome aimed to help bring about a new kind of commons — one of ideas. This commons wouldn't comprise just a web of interlinked pages of content, as is the current World Wide Web, but a web of relationships between the underlying ideas and distinctions that the content implies: a permanent, universally accessible interlinking of content based on imputed semantics such as concepts, definitions, or structured argumentation.


"Building a Semantic Wiki," IEEE Intelligent Systems, pp. 87-91, September/October, 2005 pdf html

"Bringing the “Wiki-Way” to the Semantic Web with Rhizome" SemWiki2006 - From Wiki to Semantics (2006), pp. 222-229 pdf

"RxPath: a mapping of RDF to the XPath Data Model" Proceedings of Extreme Markup Languages, 2006 html

Source Code

Github mirror

Sourceforge Project (ancient)