Integrated Systems Database
The Integrated Systems Database, is the comprehensive distributed semi-structured compendium that forms the effective basis of all trans-planetary human culture. The ISD was developed on Earth as as a way of combating issues related to network latency and throughput with real-time packet-switched networks in the interplanetary context. After a generation or so, most people simply refer to the ISD as "the database."
ISD Origins and the Mega Network
The Mega Network
Network engineers were able extend and bridge the planetary data networks to allow access from orbital outpost and even off-planet colonies. This "mega-network" even functioned for a while, but by ?2300 after couple of particularly turbulent solar flare seasons, the leading architects and engineers decided that it was better to leave the network effectively segmented and develop asynchronous systems for transmitting and sharing data.
Logically the mega network never stopped existing. Except for a splinter group on Miranda in the late 27th century, planetary network engineers were very careful to make sure that the address space didn't overlap, even if the interlinks weren't established and messages had to be queued for transmission that often took months to complete given stellar distances.
Even if the network could have been reassembled, technically possible or not, there were no real demand for for reassembly. Off-world data from a few days or weeks ago was more than anyone could process or use, so the impact was functionally insignificant. While most of the network administrators tended toward the nostalgic, ?Morgan Winters, the administrator responsible for formalizing the segmentation, said "we didn't colonize the solar system so we could read the paper back home."
Asynchronicity and Distributed Systems
Apart from the political overtones of the solar system colonization project and the convenience of a definable historical moment for network segmentation, the downfall of the mega network reflected an ongoing movement away from synchronously operating networks, and a movement to embrace asynchronicity in network and application design. Asynchronicity was, in effect, trending long before the network segmentation.
The later success of the ISD caused many to reevaluate the importance of this transition phase. The slow shift in network interaction was as much about hard performance limits and changing needs as it was about ideology. The shift happened in stages. Planetary subnets implemented request queues and large caching infrastructure to regulate and control network use, and application developers began writing applications and networked software that ran very locally, and that didn't require persistent connections. Directories were routinely cached locally, caches could deliver notifications to the relevant users when content refreshed. "Peering" outposts throughout the ?interplanetary transport network provided an additional layer of caching.
ISD's system is distributed in design, and is based around the idea of "records." Every object in the ISD is a record, and while there were standard selection of formats, the content of records is largely freeform. With some exceptions records are immutable, but versioned, so as a system ISD can incorporate updates and additions in the distributed context.
A field of library science even emerged to find different ways of more efficiently of addressing collisions and update conflicts in "the database, and these "?database gnomes," were employed by every organization on every settlement. Eventually, however, the synchronization procedures improved such most people, even the ISD Administrators never saw database conflicts or synchronization problems. It was possible to force error conditions, but even in strenuous activity the system was robust enough to prevent them from occurring in production.
The Early History of the ISD
ISD grew out of the attempts to replace books and book systems in the post-analog age. The early networks on Earth were difficult to index and structure, and therefore use. Additionally, maintaining the network was very expensive: the infrastructural requirements including the demand for real time network routing and constant availability never reached equilibrium with the ability of the network to generate revenue particularly in context of the fracturing economic structure on Earth.
The inability of the old networks to truly support asynchronous interactions and trans-planetary operation, however, were the decisive factor in their abandonment.
While the "new networks," did not immediately achieve interplanetary scale, and did carry their own infrastructure costs, the engineers, librarians, editors, and publishers, all relished the opportunity to be able to redesign information transmission to correct the mistakes of the previous generation. There were errors: the ISD and its immediate Mega-network predecessors were not effective application transport layers like the large scale public-IP networks that preceded them, but there were applications that built upon and facilitated ISD development.