Meet PLACE – An open source personal knowledge server, powered by Atomic Data. Learn more from this interview with Joep Meindertsma.
Can you briefly introduce yourself and your team?
We are Ontola and Applied Knowledge systems, two small software companies who specialize in building open source software that helps people to have more control over their data.
Ontola is a Dutch tech company. They have developed Argu.co, an e-democracy platform that helps governments to engage their citizens in decision making processes. Their team consists of four people, three of which develop software. Ontola has decided to fully focus on developing Atomic Data and its implementation, Atomic Server.
Applied Knowledeg Systems (AKS) is a UK-based software and consulting firm that specialises in knowledge management software. AKS is developing the Terraphim search engine, which uses AI to provide users with better ways to search through their own data. Terraphim is partially powered by Atomic Data.
What is your motivation to work in the data portability field?
We believe that data portability is essential to protecting the best interests of individuals. Tech giants have the power to build extremely well-integrated software that is very convenient to use, but built on vendor lock-in. Switching from the Office 365 to Google Docs, for example, is very difficult. We need open standards that help to make switching costs as low as possible. The big tech companies do not have any incentive to fund this. Even smaller software companies tend to try finding a lock-in strategy, as that helps get their company funded by VCs. So we need projects like DAPSI that fill this gap!
In simple words, what challenges does your project address?
We focus on making data models easily reusable. This means: developers should easily be able to verify any data, make sure that it conforms to some shape, without having to manually map or configure anything.
Currently, developers often have to reinvent the wheel. When an app developer has to add a `birthdate` field to a user model, this operation is often not a trivial one. They need to migrate the database, update the schema, add a class, update serializers, update the API docs, update the client methods, update the views… It’s a lot of work, and even if they are done, the data is hardly re-usable, as another dev will probably have their own incompatible definition of a User/birthdate. Atomic Data offers a way to re-use this `birthdate` concept, making the problem irrelevant.
What solution are you developing?
We came up with Atomic Data after having worked intensely with Linked Data (RDF) for the past few years. We went all-in on this technology, as we believed in the vision of Tim Berners-Lee’s semantic web. However, as we got more involved, we started to see the real technological challenges in this field. This led us to write up a bunch of best practices, which ended up becoming a full-on modular specification (the Atomic Data documentation). Our CEO Joep started with this project in his spare time, and built various server-side and frontend applications to prove that it would work. This implementation is quickly becoming a powerful and versatile database that provides a clean, simple and familiar UI while giving developers the performance and simplicity they want.
What are the next steps?
Up until now, we focused on building the fundamental technology and specifications required for data interoperability. The next steps focus on improving our main product for end-user, which means getting it to a level where it allows individuals to easily manage data and knowledge through a web application. We want Atomic-Server to become a serious alternative to tools like Notion, and we know we’re heading in that direction.