The Pangaia project's purpose is take the
WikiWay to its ultimate fulfillment:
WikiVersionTwo. It extends wiki by adding a reputation system for users (forming a meritocracy like in academia; see
UserRanking) and
PerItemVoting for each contribution.
It will move the Internet away from the geo- and server-centric legacy into a
PeerToPeer architecture, adding a
ThreeDimensionalVisualizationModel to the Internet (the
PresentationLayer). It will also attempt to do other neat things, like make a
UltimateArchitecture by revamping the
ObjectOrientedProgramming paradigm and make a
DataEcosystem. See
ObjectOrientedRefactored. The Pangaia project offers the first
UnifiedDataModel,
In other terms, it will advance
InformationScience, to hold and categorize all knowledge of civilization. More than the Library of Congress classification system (or the Dewey Decimal system) did for printed media, it will create a system of organizing all knowledge, implementing a
TheoryOfKnowledge and realizing that
AllDataRelatesToOtherData. As such, it will fulfill
CopyRight law and provide true
FairUse communities down to the every word.
Ultimately, this project will make use of a new
DataStructure called a
FractalGraph - "the data structure to end all data structures!". As a companion to the
FractalGraph is a
MessagePassingFramework to allow nodes to communicate to one another. (See
ObjectOrientedRefactored).
Once this is done, the
DeskTop will go away. Or you could call it something like an
ExoKernel. Pangaia = (
LanguageIsAnOs +
DataEcosystem) +
ThreeDimensionalVisualizationModel. In this formulation, once the paradigm shifts, Python gets re-optimized for the
UnifiedObjectArchitecture, becomes the
ExoKernel and the tree-like
FileSystem, since
AllDataRelatesToOtherData, becomes a fully-ordered
NameSpace. The VPython Visualization model becomes the User's interface into
CyberSpace.
Think of it like this, you're used to viewing the O.S. as something for the computer, but a computer doesn't need an O.S. It was an accommodation to handle applications, but back in the old days, applications didn't know about each other, and created little data silos for themselves. Now, with the Internet especially, all data can relate to everything else. And with the rise of personal computing, each machine became a single-user system managing their personal data and associations.
So make this explicit. Use the 3-d graph to manage data directly and visually.
I don't understand this paragraph. For me the operating system is just that, something which enables the computer hardware to be accessed by the operations specified by the user or users through the software layers. This is something which is getting to be more and more complicated as the hardware evolves. Over time the access to the hardware has been made more remote by generalizations which have been added to the operating system. How can you manage without these layers? --
JohnFletcher
- You are correct, that is what an operating system is (has been). With a UnifiedDataModel, all data relates to everything else, so if you have a good data model, you can simplify everything, because you can go right to the hardware from the application itself. You no longer need a hundred apps for managing data in different ways (word processor, spreadsheet, databases) all with their own DataSilos, because now everything is related to everything, especially with the internet.
I mean something like an
ExoKernel architecture: instead of a "middle-man" organizing CPU scheduling of different applications, the application
IS the OS; i.e. the application and the
OperatingSystem become <ONE>. The application becomes the whole access point into the
CyberSpace, while being microoptimized for CPU utilization and a broad view for data I/O. Each node in the object model is a process, and since it's designed to be a unified model, it is the main focus at any given time...
- {The author of this page has probably been directly or indirectly influenced by NakedObjects or another DirectManipulation OO environment like the TomatoIde. The idea of being able to interact directly, in a consistent fashion, with everything from the hardware on up is conceptually appealing, but has not yet been demonstrated to be practical to general users.}
- Actually, it comes from the Python interpreter environment which has a high degree of inspection capability built in.
But possibly it will require a
TrustedSystem utopia that doesn't exist. Yet since a big part of the concept is about
FairCredit assignment, the possibility of such may be attainable.
But if you want the codez to create the
SelfOrganizing PerfectSystem, you'll have to find
MarkJanssen.
Otherwise, just help popularize
VisualPython and
ProcessingLanguage.
PeerToPeer is overrated. Central servers are cheap for text-centric apps, making it easier to keep everything in synch. And if everybody reinvents the wiki in their own little cloud, then hyperlinks or their equiv become useless or confusing.
- No, you are thinking small scale. The purpose of building a UnifiedDataModel is to allow PeerToPeer to scale towards billions of nodes.
See also
GlassBeadGame,
BlueAbyss,
TreeOfLife,
EverythingTwo,
PythonThreeThousand
CategoryOpenSource
AprilThirteen