Crazy Wisdom

Crazy Wisdom

Episode #415: Rethinking Databases: EdgeDB’s Blueprint for a Developer-Friendly Future

Episode #415 · December 6, 2024 · 55 min

MP3 · Apple Podcasts · Spotify

About this episode

On this episode of the Crazy Wisdom Podcast, host Stewart Alsop is joined by Yury Selivanov, the CEO and co-founder of EdgeDB, for a fascinating discussion about the reinvention of relational databases. Yury explains how EdgeDB addresses modern application development challenges by improving developer experience and rethinking decades-old database paradigms. They explore how foundational technologies evolve, the parallels between software and real-world systems like the electrical grid, and the emerging role of AI in coding and system design. You can connect with Yury through his personal Twitter account @1st1 (https://twitter.com/1st1) and EdgeDB’s official Twitter @EdgeDatabase (https://twitter.com/edgedatabase).

Check out this GPT we trained on the conversation!

Timestamps

00:00 Introduction to the Crazy Wisdom Podcast
00:27 What is EdgeDB?
00:58 The Evolution of Databases
04:36 Understanding SQL and Relational Databases
07:48 The Importance of Database Relationships
09:27 Schema vs. No-Schema Databases
14:14 EdgeDB: SQL 2.0 and Developer Experience
23:09 The Future of Databases and AI Integration
26:43 AI's Role in Software Development
27:20 Challenges with AI-Generated Code
29:56 Human-AI Collaboration in Coding
34:00 Future of Programming Languages
44:28 Junior Developers and AI Tools
50:02 EdgeDB's Vision and Future Plans

Key Insights
  1. Reimagining Relational Databases: Yury Selivanov explains how EdgeDB represents a modern rethinking of relational databases. Unlike traditional databases designed with 1970s paradigms, EdgeDB focuses on improving developer experience by introducing object-oriented schemas and hierarchical query capabilities, bridging the gap between modern programming needs and legacy systems.
  2. Bridging Data Models and Code: A key challenge in software development is the object-relational impedance mismatch, where relational database tables do not naturally map to object-based data models in programming languages. EdgeDB addresses this by providing a high-level data model and query language that aligns with how developers think and work, eliminating the need for complex ORMs.
  3. Advancing Query Language Design: Traditional SQL, while powerful, can be cumbersome for application development. EdgeDB introduces EdgeQL, a modern query language designed for readability, hierarchical data handling, and developer productivity. This new language reduces the friction of working with relational data in real-world software projects.
  4. AI as a Tool, Not a Replacement: While AI has transformed coding productivity, Yury emphasizes that it is a tool to assist, not replace, developers. LLMs like GPT can generate code, but the resulting systems still require human oversight for debugging, optimization, and long-term maintenance, highlighting the enduring importance of experienced engineers.
  5. The Role of Schema in Data Integrity: Schema-defined databases like EdgeDB allow developers to codify business logic and enforce data integrity directly within the database. This reduces the need for application-level checks, simplifying the codebase while ensuring robust data consistency—a feature that remains critical even in the era of AI.
  6. Integrating AI into Databases: EdgeDB is exploring innovative integrations of AI, such as automatic embedding generation and retrieval-augmented generation (RAG) endpoints, to enhance data usability and simplify complex workflows. These capabilities position EdgeDB as a forward-thinking tool in the rapidly evolving landscape of AI-enhanced software.
  7. Balancing Adoption and Usability: To encourage adoption, EdgeDB is incorporating familiar tools like SQL alongside its advanced features, lowering the learning curve for new users. This approach combines innovation with accessibility, ensuring that developers can transition seamlessly to the platform while benefiting from its modern capabilities.
Episode transcript
Stewart Alsop III00:00

Welcome to the Crazy Wisdom Podcast. This podcast is for you. If you have an insane drive to find the truth of things, it's not the good answers that we seek, but the good questions. I interview a range of different guests from many different fields, all with the intention to uncover the simple truths that are hidden in plain sight. Most people don't want to go there. I go there, my guests go there, and you benefit. Please let me know if you enjoy these episodes and as always, subscribe on itunes, Spotify or wherever you listen to the podcasts.

Stewart Alsop III00:37

Welcome to the Crazy Wisdom Podcast. Today my guest is Yuri Selivanov and He is the CEO and co founder of EdgedB. And we're going to get into what exactly EdgedB is because I have a AI whisperer down here in Buenos Aires who's a senior developer and he told me that EdgedB might, might actually fit into what I want to build. And so welcome to the show, Yuri.

Yuri Selivanov01:00

Thank you so much for having me.

Stewart Alsop III01:02

So what is the one to three liner for edgedB? What does edgedB do?

Yuri Selivanov01:11

It's essentially a reinvention of relational database from scratch. Given modern requirements to application development, we actually want to make databases easier to work with. And it's not a big secret in the industry that developers don't really like databases and dealing with them. And one of the reason for reasons for that is that relational databases were invented in the 70s and most of the decisions for the syntax of the query language, for the data model, how you define your databases, it was defined in that period and it was never properly challenged. So while the rest of the ecosystem evolved drastically, we have so many different programming languages that didn't exist in the 70s or even like 20 years since 70s. We haven't seen anything like that happen to the database world and with HDB will want to change that.

Your first full transcript is free. After that, an email opens every transcript in the index — a list of readers we can write to, not a guest book.

Subscribe