Relational Database
topic on 5 shows · 15 statements across 11 episodes
In Depth 
Founders 
No Priors 
the MAD Podcast 
the a16z Podcast
15 statements about Relational Database, every show
Pursuing High Apparent Risk Deters Competition and Improves Winning Odds
“The bigger the apparent risk, the fewer people will try to go there. We would surely lose if we had to face serious competition. But if we were all alone in pursuit of our goal of building the first commercial relational database system, we had a chance to win…”
Tech Press Originally Claimed Relational Databases Would Never Be Useful
“The technical press at the time was saying that the relational database would never be useful.”
Relational databases perform terribly at full-text search workloads
“Understandable, but staggering to me how bad relational databases are at full text search. Just, they're just awful.”
Goyal: Relational databases are fully capable of adding HNSW vector indices
“Relational databases are perfectly capable of adding HNSW indices to them.”
Huber: Vector databases primarily handle un-databased unstructured data
“Vector databases are primarily about unstructured data. It's actually taking data that had no database that knew about it and loading it in for the very first time. Most applications of this stuff is not about taking data that's already in your relational data…”
Multi-level graph queries are extremely hard to run in MongoDB or SQL
“So this type of query in a relational database or something like MongoDB is going to be very hard to run. You've got a lot of joins and you've got very complicated queries.”
Eifrem: Neo4j is frequently 1,000 times faster for connected data queries
“We've optimized every layer in the stack of the database architecture completely around connected data. We're not built on top of a different database or anything. It's a native architecture. And that means that if you want to query along how things are connec…”
Stewart: Modern relational databases must be shared-nothing distributed SQL
“We think that again, if you were to build a relational database today with the current state of art, you would end up with a shared nothing, a SQL database that supports transactions.”
Stewart: Migrating to NoSQL sacrificed ACID consistency, creating major development drawbacks
“You traded scale for acid, but that's something you lost when you're moving away from the relational database. And it turned out that losing acid Was a big deal.”
Frank Chen: AI development tools are predominantly open source compared to databases
“In fact, one of the big differences between the relational database and AI tools is that a lot of the AI tools are actually open source.”
Horowitz: Relational databases function like Microsoft Excel on steroids
“A relational database is basically like Microsoft Excel on steroids. If you take Microsoft Excel, and you've got tables and rows, and you've got, you know, columns, and it's kind of like a relational database, but what a relational database does is give you an…”
Horowitz: MongoDB's key innovation was enabling indexing and searching inside documents
“The key thing that we did is, you know, you can always just store everything about a person in a single cell in a relational database. The thing that we did that was interesting was let you actually index on it and search on it and modify it in sort of very sm…”
Liu: Most software is just a CRUD interface over relational databases
“The vast majority of software out there in the world is not going towards some exciting new technology or, you know, not some, you know, innovative new consumer product, but really it's a reinvention of the same wheel again and again. And that is a create, rea…”
Liu: Collaborative undo is harder in relational databases than document editors
“In a relational data context, however, it's a much harder problem to solve, and you have to deal with things like, what if, you know, one person changes a column type from a linked field to a text field, and then somebody else was making edits to that field at…”
Hadoop was designed for new data problems, not relational database issues
“What we didn't understand at the time, and it's been a pretty common feeling, is Hadoop wasn't built to solve the problem we'd been solving with relational databases. It was designed to solve a new problem, and it turned out that new problem was going to be ve…”