Advanced Document Architecture (ADA), a software paper. An interoperable framework for microservices in scholarly publishing to introduce a P2P economy in research publishing for institutional takeback of infrastructures. The motivation behind the ADA software is to enable technological approaches to support research publications and related artifacts in the research cycle to become interoperable and free from restrictions placed on ‘use’ by systems provided by large scholarly publishers. ADA is a publishing microservices framework intended to lower the costs of publishing by automating publishing processes and drastically speed up software development. The technology employed for the ADA microservices are automatic deployment using Kubernetes clusters and content distribution using Apache Kafka data streaming. Microservices are a cornerstone of DevOps in commercial computing, where the FOSS suppliers have come to dominate business IT infrastructure provision, driving out closed source providers on price and innovation. ADA has a mission to support a similar change over in academe, with a vision that institutions use such cost savings to take back service provision and infrastructure in-house. ADA also aims to accelerate P2P networking of the wide array of existing FOSS publishing platforms to disintermediate the publishing market away from existing scholarly publishers who are not in the business of innovation. To support this P2P approach ADA proposes a metering service to allow for the payment for service provision by platforms. The ADA partners are building a working example of the microservice system as a proof-of-concept for community handover as a Kubenetes cluster and cloud deployment on Google Cloud Product (GCP). Example core components would be: services connector; multi-format document transformer; document validator; pipelines; metering and billing of services; and cryptographic IDs. ADA is takes inspiration from and sits alongside other ‘liberation technologies’ for research which are working on future publishing systems: Open Research Knowledge Graph; Solid and Dokieli; Project Jupyter; Open Knowledge Maps; or as in Herbert van de Sompel’s presentation ‘Scholarly Communication: Deconstruct & Decentralize?’; and as well as from support initiatives, such as: Invest in Open Infrastructure, or Research Software Engineering organisations such as De-RSE.

{"references": ["Auer, S\u00f6ren. \"Towards an Open Research Knowledge Graph.\" Zenodo, January 22, 2018. https://doi.org/10/gfv3rv.", "\"Solid: Social Linked Data | MIT CSAIL.\" Accessed October 21, 2018. https://www.csail.mit.edu/research/solid-social-linked-data.", "\"Dokieli.\" Accessed March 27, 2018. https://dokie.li/.", "\"Project Jupyter.\" Accessed December 6, 2017. http://www.jupyter.org.", "Maps, Open Knowledge. \"Open Knowledge Maps - A Visual Interface to the World's Scientific Knowledge.\" Open Knowledge Maps. Accessed November 27, 2018. https://openknowledgemaps.org/.", "Van de Sompel, Herbert. Scholarly Communication: Deconstruct & Decentralize? Washington, DC: CNI: Coalition for Networked Information, 2017. https://www.youtube.com/watch?v=o4nUe-6Ln-8.", "\"Home.\" Invest in Open Infrastructure. Accessed August 8, 2019. https://investinopen.org/.", "\"De-RSE.Org - Research Software Engineers (RSEs) - Responsible for Scientific Software.\" Accessed August 8, 2019. https://www.de-rse.org/de/.", "Svingen, Boerge. 2017. 'Publishing with Apache Kafka at The New York Times'. Confluent. 6 September 2017. https://www.confluent.io/blog/publishing-apache-kafka-new-york-times/.", "Tennant, Jon. 2019. 'These Figures Always Blow My Mind. We Spend $2.2 Trillion on Global Research & Development Each Year. Yet, We Can't Find a Better Way to Communicate the Outputs than by Outsourcing the Sharing of Our Knowledge to Commercial Third Parties. We Can, and Must, Do Better. #seed2019pic.Twitter.Com/5Kwl5tYDdT'. Tweet. @Protohedgehog (blog). 25 February 2019. https://twitter.com/Protohedgehog/status/1099973052710965248.", "Dar: Reproducible Document Archive. 2017. Reprint, Substance, 2018. https://github.com/substance/dar."]}