Welcome to ArcadeDB: Play with Data

And the Quest for a New Business Model for Open Source Projects

·

5 min read

Welcome to ArcadeDB: Play with Data

Sept 1st, 2021 - by Luca Garulli Founder of ArcadeDB (Founder/CEO of OrientDB)

The First Multi-Model Database

OrientDB was the first Multi-Model DBMS on the market. In 2009, I started working on OrientDB as an Open Source project (license Apache 2). In 2012, I was invited to Cologne, Germany to present this novel idea in a keynote session at the NoSQL matter conference. These are the first 2 slides presenting the concept of Multi-Model (full presentation available).

image.png

The feedback received from my presentation was pretty binary:

  1. This is ridiculous. You’re acting like an RDBMS with one product that fits all sizes! NoSQL is about having different datastores, one per use case. If your application is complex, just use polyglot persistence by using many of them. Easy peasy.
  2. I love this idea, basically, it's a general-purpose DBMS based on NoSQL instead of the Relational Model!

After Germany, I decided to rebrand OrientDB as a Multi-Model DBMS to push this "graph document database" concept forward. OrientDB was a community-driven project with zero marketing budget. At first, I created a company just to respond to the increased demand for training and commercial support. We grew by hiring the most active committers on Github. Accepting outside investment was very tempting, but bootstrapping the company allowed us to make our own rules and helped us grow lean and organically into a debt-free, profitable company. After only a few years in the business, having fun working on something innovative and exciting, our user base started to grow exponentially. Clients started coming to us from all around the world. Big companies bet on us, the underdog, like Apple, Lockheed Martin, British Telecom, Barclays, just to mention a few, and of course an endless number of startups. But that was just the peak of the iceberg in regards to users. OrientDB uses the Apache 2 Open Source license, which means everyone can use our product for free. We never really knew how big the real user base was.

After only 6 years, OrientDB was acquired by Callidus Cloud, a public company on Nasdaq. After 4 months, it was acquired by SAP.

What’s next?

What should I do now? What do I really love? That was simple: coding and working with talented people with bright minds and innovative ideas for resolving complicated problems. Discovering new technology, crunching data, open-source stuff: this is what I love. I don’t love the idea of going back to the full startup business with the goal of making money to pay for things like the sales team, marketing, advertising, lawyers and market analysts.

And suddenly a new idea: what if I run the company as an Open Source foundation? A foundation that has development and research as the main goal? Not like the Apache Software Foundation. I was thinking more about a new model where companies that need support can still find a home, but the income is distributed among contributors that are actually working to support the project.

Almost all of the companies I know in the technology business started with a handful of core tech guys, but they ended up having marketing/sales as the number one cost. What if I cut everything out except R&D? No sales team. Zero budget for advertising. Only word of mouth. Like Tesla.

I like it. I couldn’t find any reference of existing companies that are using this business model (but if you know other companies that are embracing this model, please let me know in the comments), so for now, I’m going to call it “The Arcade Business Model”. “Arcade” because it’s (1) the fun period of videogames in the '90s (today mostly associated with the retro-gaming), and (2) “arcade” means “ a series of arches with their columns or piers ”, namely connections between columns and piers.

I love thinking of this model as a way to encourage the connections of people who have fun developing cutting-edge technology. We’re going to experiment and adjust this model to make a lean and sustainable business for both sides: contributors and clients. Contributors can be finally get paid for their work with the Open Source and clients can use this advanced technology for FREE (no strings attached). If some clients want a new feature, custom work, direct access to the support team and/or getting an issue fixed ASAP, then they can sponsor the project through one of the available tiers in GitHub Sponsor as a Recurrent or One-Time payment.

A lot of thoughts, but now it's time for action. I’m thrilled and excited to announce ArcadeDB!

arcadedb-logo.png

It shares the same philosophy and principles of the Multi-Model, but with a brand new transactional engine, built with mechanical sympathy concepts in mind to squeeze the most out of modern multi-core hardware. ArcadeDB is a powerful, full-fledged DBMS stack that you can also embed. It's not a low-level library, like RocksDB or LevelDB.

ArcadeDB not only supports OrientDB SQL and Apache Gremlin but also Neo4j Cypher (Open Cypher) and MongoDB Query Language and basic Redis API for Key/Value operations. That’s right, you can use all these languages on the same database, in the same transaction!

Creating and supporting drivers for every single programming language was extremely difficult with OrientDB. With ArcadeDB, you can use a Postgres driver or directly with HTTP/JSON API. If you’re coming from MongoDB, you could use it with any MongoDB driver executing MongoDB queries.

ArcadeDB is licensed Open Source with Apache 2 license. Check out the project on GitHub: github.com/ArcadeData/arcadedb.

The documentation is online at: docs.arcadedb.com and the main website is arcadedb.com.

I’m going to update this blog with use cases from users that switched to ArcadeDB and new interesting ideas about the multi-model concept and data in general. OrientDB's success story was written by a large team of passionate and talented people. And this is the magic I’d like to find again with ArcadeDB.

If you are interested in participating in the project, please follow us on our new channels:

GitHub: github.com/ArcadeData/arcadedb

Twitter: twitter.com/arcade_db

LinkedIn: linkedin.com/company/arcadedb

Facebook: facebook.com/arcadedb


Have Fun with Data,

Luca Garulli, Founder of ArcadeDB (Founder/CEO of OrientDB)


All product names, logos, and brands are the property of their respective owners. All company, product, and service names used in this website are for identification purposes only. Use of these names, logos, and brands does not imply endorsement. OrientDB is a registered trademark of SAP, Inc. Cypher is a registered trademark of Neo4j, Inc. MongoDB is registered trademarks of MongoDB, Inc. Redis is registered trademarks of Redis Ltd.