by Jack Krupansky - Base Technology
March 3, 2008 Current Version, Prior version: March 3, 2008
This document is a proposal for Factpad (the name is a tribute to the simplicity of Notepad), a tool that is designed to make it very easy for consumers to author, store, manipulate, publish, and share "facts" that interest them, and to do it in a way that is completely in alignment with the philosophy of the Semantic Web.
The basic idea is to represent a list of Semantic Web "facts" as a standard text "file" in appearance to the user, allowing them to create, edit, delete, search, sort, etc. in standard text format, while maintaining the Semantic Web "database" inside the editor and storing the database as either (or both) a standard XML file or into a "Factbase" as described above. Besides the database engine, the editor would be built around a collection of micro-formatters to format discrete facts in clean text form (user-selectable views) and micro-parsers to parse clean text facts into Semantic Web triple form. Overall, what the user sees in Factpad is simply a database "view", or actually a composite of a sequence of views. The editor needs to be really, really simple ("Notepad-simple"), but also have the ability to show the underlying fact details and select alternative views for various types of facts. Kind of like the old "show format codes" in WordStar! A fancier editor mode might include outline-style show and hide features, but the basic editor needs to be absolutely "blank white page", dirt-simple. It really is just a text editor, but with an engine behind the scene to match text and "facts".
Yes, a GUI would also be appropriate an an alternative interface for Factpad, but that would only supplement to capabilities of Factpad and would add any new semantic-level capability. In fact, the best GUI would be to support multiple Factpad text windows, possibly with lines or simply arranging them in visual structures to facilitate references between facts and to simplify text entry.
The next step is to have a "Factcloud" tool to allow the user to relate their local facts to the factbases of other users and "authorities" (schemas and ontologies.) The main function of Factcloud is to add context to local facts. A secondary purpose of Factcloud is to provide nested libraries of views, micro-formatters, and micro-parsers that the user and Factpad editor can build upon. Factcloud would support the concept of a dictionary stack for context so that user dictionaries, community dictionaries, and dictionaries for user-recognized experts/authorities can be arranged and prioritized to suit the needs and interests of the user while allowing the user to exploit and build on the work of others, and for others to utilize the user's work in the same way. The Factcloud too is the "connector" between local Factpad editing and cloud-level fact coordination.
Factbase can be used at four levels:
In short, there are three core components here:
The ultimate goal is two-fold:
By achieving this two-part goal, Factpad will be enabling a far-richer consumer-centric Semantic Web.
That is the idea in a nutshell. More refinement to come.
I will continue to revise and extend this core vision and concept, but at this stage I do not have any intention of turning the vision/concept into a commercial product or service.
Although this document is copyright-protected, it may be freely copied or excerpted and even implemented as a commercial product or service in whole or in part, provided that there be an acknowledgement of this base document as being the provenance of the core vision/concept.
No my knowledge, there are no comparable products or services, although there have been efforts to produce databases for Semantic Web RDF triples, so-called triple stores.
-- Jack Krupansky
Please contact us with any questions or comments.
[ Home ]
Updated: March 04, 2008 12:49:19 AM -0500
Copyright © 2008 John W. Krupansky d/b/a Base Technology