The Road to a Standardized Graph Query Language: GQL, Part 2

The Road to a Standardized Graph Query Language: GQL, Part 2

Last month I wrote about the ISO meeting in Brisbane, working towards a standard query language for property graphs.  I recently returned from Berlin, where the World Wide Web Consortium (W3C) held a special workshop to build bridges between the semantic graph and property graph communities. The 3-day conference covered a range of topics, grouped into three broad categories:

  • Problems & Opportunities
  • Interoperation
  • Standards Evolution

I discovered this W3C workshop to be filled with a dedicated and diverse group of experts from industry and academia, who are passionate about extracting value from linked data. In the spirit of openness and collaboration, each session’s attendees took notes on a shared Google doc. You can see the W3C graph conference agenda and notes here. Noted industry journalist George Anadiotis was in attendance and wrote this excellent piece on how this standardization effort signals a turning point for the graph database market.

Applying my background in both hardware design and graph algorithms, I moderated a session on Queries and Computation. I also made invited comments at the SQL & GQL and Graph Query and Interoperation sessions. Many of my Neo4j and Oracle colleagues from the ISO Property Graph working group were there, with similar roles, alongside our peers from the Semantic Graph community. There was a lot of open discussion about the state of the technology, visions for the future, and building bridges. Bridges among the data models and the human communities.

Following the official W3C workshop, an ad hoc group called the Property Graph Schema Working Group met face-to-face for the first time. This group formed following a previous community effort, G-CORE, to express some core needs for a property graph language. While this group has no official standing, there are participants here, like me, who are also involved with the ISO working group. It’s a great way for academics and others who don’t have the time to invest in the formal ISO and ANSI efforts to have their voices heard.

So where are? We’re moving forward, coming together, with growing momentum. The key outcomes from the week in Berlin, as I see it:

  • A Bridge for Data: We will define a standard conceptual model for property graphs. Once we have this, then we can talk about serializations (standard data formats, so different systems can transfer data more easily).
  • A Bridge for People: W3C is setting up the liaisons to collaborate with ISO’s GQL effort, to provide comments with an eye towards interoperation with semantic graphs.
  • A Bridge to the Future: Being in the thick of things, I’m able to see the suggestions for GQL that will benefit users. TigerGraph isn’t waiting. We’re already evolving GSQL to deliver enhanced capabilities and more standard formats.

I’ll keep you informed on new developments. Thanks for listening.

Victor Lee
Director of Product Management