Help:Archive Machine

From autonomous archive
Jump to navigation Jump to search


The Archiving Machine is a combination of hardware and software designed to facilitate, and encourage interaction with the Autonomous Archive. Its design principles are developed from the motivation of re-activation of static documents, and collaborative community engagement with the archive of the Poortgebouw. The parts that make up the Archive Machine are found hardware consisting of; a PC, screen, keyboard and scanner, and of software written in bash and python that utilises Optical Character Recognition, online translation, and automatic Media Wiki uploading and page creation.

Information pertaining to the usage of the Archive Machine can be found at the Getting Started page. For a more detailed overview of the systems and methods of archiving with this system, including the use of properties and advanced searching, go to the Archival Methods page.


The house's history can be seen from its very beginnings in not only in contents of the archive documents, but also in the autonomous existence of the building and its community. This too is evident in the technological and network infrastructure in the buildling. This Archive Machine by nature, continues that tradition in its form, and function.

All the hardware was either sourced from the inhabitants, or found abandoned in various corners of the building. Where possible hardware was fixed, or Frankenstined together so as to meet its functional requirements. The software was coded using open source, libre software libraries and is intricately documented, so as to enable further development, deployment and adoption.

Furthermore, all the systems are not reliant on internet access and thus are functional regardless of API support, or any other proprietary service providers. All hardware is non-specialised, consumer level devices that can be fixed, replaced, and upgraded. The software and hardware does not dictate a use of the system, and it is open to adaptation where desired.


The technologies used, and their setups are recored below. All information contained in this page is intended as both documentation of the process for archival purposes, but also as a maintenance aid for users of this archive. Both technical and ideological information has been recorded here.


All hardware has been provided by donations in the house. The setup is not reliant on any specialised equipment, and the software choices can be installed on most cheap consumer computers. It is in keeping with the community of the Poortgebouw that this setup has no extraordinary components unless totally necessary. The ability for this setup to grow, adapt, and be repaired by anyone with an intermediate to advanced interest is extremely important. Reliance on one person would not be sustainable in the adoption of this system and thus steps have been taken to keep all the hardware as mainstream as possible.

As of writing this, the archive machine is a donated Dell Optiplex 280 with no upgrades or modifications. The scanner is a half working multi-function printer scanner, in which the print functions are inoperable. The software on the archive machine should be modular enough to have any replacement as far as hardware input is concerned.


The software is a combination of custom programmed scripts and opensource applications. You will find here an overview of the various systems setup on the Archive Machine, some of the motivations for choosing this software, and basic functionality. For further understanding of these systems relevant references have been provided in the way of external links, or local wiki pages.


The base of this system is Media Wiki. The design choice to use this system was primarily driven from its ease of use and its familiarity amongst most internet users. As it is the same system upon which Wikipedia is built, for users who wish to browse the Archive there should be no usability issues. For the archive system (both technological and systematic) it is transparent in the editing of complex document relationships, that can be explained and edited in human language. Very little specialist knowledge or exclusive access is required to interact with, or modify the system.

Semantic Media-Wiki[edit]

Semantic Media-Wiki is a plugin for Media-Wiki that allows for more complex relational information between documents. More granular properties can be added to a page so that it may be collated or searched easily. As the documents in the archive have complex relationships to each other, it is important that their datas be recorded in such away that they can be acted upon. This activation allows for interesting possibilities now and in the future in regards to new document generation, or more precise access to important information.

The archival interface currently asks for properties like Author, Year and Document type. Once a page is created in the wiki however, it is easy to expand these properties further. Information on how to add these properties can be found on the Semantic Media-Wiki website, or on the Archival Methods page in this wiki.


The AutoArchiver is the software interface for the Archiving Machine. It allows for the digitisation of the physical archive, and facilitates in the uploading of images, and the creation of wiki pages per document. The code is written in Bash and Python and is freely available and editable from the AutoArchiver page. All libraries, tools and the code itself are open-source and freely available to anyone.

The software's function is not to make the operation seamless but to break the initial barrier of uploading and page creation on the wiki. It provides a way to easily generate relevant files and pages on the wiki that can then be further edited by any user in the community.


The archive machine is running Debian 8 stable version (which was the latest stable build at time of writing). The install is minimal with no desktop environment. The machine only has one user, no special hardware nor software installed. There is only one user on the machine aside from the root user called AutoArchiver. The machine has also been setup for SSH access so that it may be administered remotely (albeit within the Poortgebouw).


It has always been the intention to keep the archive machine local within the Poortgebouw. There is no external access to the wiki, nor any other services. It is important that the community has sole access to the wiki as it contains sometimes sensitive data pertaining to the house and its past or present inhabitants. In the future, opening up some of the data to be accessible externally may be beneficial, but at this initial stage no a priority.

The machine is connected to the building network through a relay of various routers connected via wifi. Due to the complexity of the network in the building inter building connection is a challenge that will have to be overcome. There is no central DHCP server, and thus inter network connections to the archive machine may not be possible at this time. Furthermore, future desires to have the archive accessible via the internet, will pose difficult insofar as port forwarding and other firewall restrictions are concerned.

Git Hell[edit]

Debug data: