From PKC
Jump to navigation Jump to search


To help manage personalized and project-based data asset, PKC is designed to run in scalable configurations, starting from personalized computing resources, up to multi-server cloud environments. The overall approach of PKC is try to minimize the entry-barrier of automated workflow and create an inclusive field of growth driven by Moore's Law. The system design work of PKC will try to follow principles presented in the book: Specifying Systems[1].


Create a basic set of services, files, and page content to help individuals operate a MediaWiki website on any machine of their choosing. This will allow them to continuously work with their own data assets, independent of future changes.

Success Criteria

  1. Allow Individuals to install an instance of MediaWiki service by reading this PKC/ file.
  2. Make all textual content, executable software images, installation scripts in the public domain, so that everyone can share and use them at will.
  3. Provide instructions to learn about how to use PKC in the initial MediaWiki's database, so that people can start learning to use PKC through their own instance of MediaWiki.

Required Resources

  1. A computer[2] that you have access to its "root" or "administrator" previledge.
  2. Operatng Systems that support Docker: Windows 10, Mac OS X Big Sur 11.2.3 and Linux.
  3. Under Windows 10 Environment, some VPN software might interfere with Docker's Windows Subsystem for Linux, a.k.a. Windows Subsystem for Linux (WSL), you will need to remove VPN software before installing [Docker]. In case you don't want to remvoe your VPN software, or your Docker and Bash have compatibility issues, please try to [VirtualBox PKC] solution.
  4. Access to the Internet during intallation time. Please try to perform the installation on a network with 10Mbps+ to the Internet. After installation, this system can operated without access to the Internet.
  5. Relevant software implementation examples and architectural patterns in the field of Knowledge Management

Installation Procedure/Implementation Process

  1. Installation must be conducted by a person who possess the root password of the target machine. When it starts installing, root password needs to be provided in a command line interface (CLI).
  2. You will need Git, the software installed on your machine. You may follow the instructions on Git Software Installation and Download.
  3. Install Docker, the Installation instructions and downloadable files can be found here:
  4. After the success of Docker installation, launch a terminal application, or often called command line interface (CLI)interface, move your working directory using the command cd", meaning change your working directory, to a directory in your file system. For Mac OS X and Linux operating systems, Terminal applications are bundled during installation. For Windows 10 PKC Installation, please refer to this page.

Assume the directory you keep your working files is called: Workspace, your terminal application/command line interface (CLI) should present a prompt, similar to the following format: (your environment might also display machine name and user name along with the working directory name)


Download the entire package using git. You may copy the instruction as follows:

 ~/userName/Workspace$ git clone
  1. After the git clone instruction copied relevant data to your working directory, using the command cd to the PKC directory that contains the script, and type: ./ to execute the script.(If you are running Windows machine, just type:.\up.bat or just up.)
 ~/userName/Workspace$ cd PKC
 ~/userName/Workspace/PKC$ ./
  1. After you submitted the ./ in your command line interface, you will be asked for a password, type in the root password for the machine you are trying to install.
  2. Depends on your network connection, this process will download around 5GB of data, so that it will take a while.
  3. Open a browser: type the following URL to the browser's URL field: http://localhost:9352.
  4. In case you need to have administrator access, use Admin as user name, and metalanguage as password.
  5. If you plan to set up https for your PKC, please refer to the page:Nginx as Reverse Proxy.

Expected Outcome

  1. Every 30 minutes, all the changes you made to your local instance of MediaWiki will be automatically backed up to the directory's "backup/" sub-directory.
  2. The textual content stored in MediaWiki's database can will be stored in an XML file: XLPLATEST.xml
  3. All the uploaded files, assuming the file names are accepted by the host operating system, will be dumped to the "backup/MediaFile/" sub-directory.
  4. New applications and data processing patterns can be defined by PageFlow

Boundary Conditions

  1. We can not warrant any reliability, completeness, and accuracy of this installation procedure. Any action you take upon this information and execute this script is at your own risk, We will not be liable for any losses and damages in connection to the use of the actions and software prescribed here.
  2. We have only tested on a small number of machines and configurations, your mileage may vary.
  3. Do not remove any of the files in the directory with backup/, such as docker-compose.yml and the LocalSettings.php. These files are the configuration files for Docker and MediaWiki respectively. Missing them, this system will cease to work.
  4. Many pre-defined parameters can be replaced in the .env file stored in the PKC directory. It allows flexible configuration of your system. For more information, please read Configure your own PKC instance.
  5. For the sake of reducing typos, the project has moved from to For the current implementation of Github, these two git repository names point to the same source. When GitHub change this practice, it will change. We recommend you to use the shorter version.
  1. Lua error: Internal error: The interpreter exited with status 126.
  2. Lua error: Internal error: The interpreter exited with status 126.