News for hackers. Stuff that matters.
Re:Nvidia – open-source auto-ontology by gavinray on Tuesday November 03, 2026 @11:29AM (#50354084)
I want to attempt an explanation as it seems a lot of folks here didn't grok the very terribly written README

"Auto Ontology" is sort of a poor name.

Here's the problem:

Your company has data scattered across multiple types of databases. People who make decisions want to ask questions about that data.

The old way is they would go to a Data Analyst, and say, "Get me a report on so and so", and it was the Analyst's job to iron out the right series of queries and one-off scripts to generate the needed info.

With a system like this, the idea is that all your database models have been introspected and stored in a central metadata catalogue.

When a question is asked like "What were the top selling items in the month of May?", the metadata catalogue tells you "Products are stored in FOO_DB" and "Product_Sales are stored in BAR DB" and so you go and fetch the right data from the right place and do whatever in-memory calculations you need to do.

What the system provides is the implementation code for:

- Introspecting each of your DB's

- Aggregating model + metadata information in a central, virtual DB (VDB)

- Generating dialect-specific SQL from natural language queries and routing them to the right DB for the underlying table

- Combining results in-memory when a natural language query requires multiple datasources

- Handling proper permissions, etc

A GOOD system for doing this also integrates with your external API sources (OpenAPI, GraphQL, gRPC, whatever) and also documentation sources (cough, I work on this, cough).

Re:Nvidia – open-source auto-ontology by throwaway5752 on Tuesday November 03, 2026 @11:32AM (#50354140)
"seems a lot of folks here didn't grok the very terribly written README"

I tried to make it through the README and it was just unmotivated slop. Perhaps it reads better to agents. It is like the Van Halen rider story - if the authors didn't care to write a good README, what are the odds that they paid attention to the rest of the large codebase?

Re:Nvidia – open-source auto-ontology by gavinray on Tuesday November 03, 2026 @12:00PM (#50354571)
I had to go digging into the source to realize what it was, e.g.

https://github.com/NVIDIA/auto-ontology/blob/main/auto_ontol...

Re:Nvidia – open-source auto-ontology by Twirrim on Tuesday November 03, 2026 @12:02PM (#50354614)
I ended up throwing the README at an LLM and it gave me a much simpler and clearer explanation. Copying the bits of an explanation that helped me:

> The problem it’s trying to solve is that generating valid SQL doesn’t necessarily produce the right answer. An AI might find a column called revenue, but still choose the wrong dataset or misunderstand whether that figure includes things like refunds. Auto Ontology adds information about what the data means to help it make those decisions.

and then the one paragraph summary at the end:

> My reading: it’s a database chat application built around a reusable catalogue of business meaning. The question worth testing is whether that catalogue makes answers measurably more correct—and how much human work it takes to get the definitions right.

Re:Nvidia – open-source auto-ontology by throwaway5752 on Tuesday November 03, 2026 @01:40PM (#50356231)
I could easily solve it myself, but the point is that I don't need to spend my valuable time on a project that the author's done care enough about to give a proper readme. And semantic web/graph indexers and queriers are not new, even if there are new tools at our disposal for classifying and querying ontological structures.

If they can't do that, it's a signal they didn't care enough about the details to rely on this project, and there is more cool software than ever I can spend my finite time on.

That may not be particular fair, and it may be a great project, but the bar is just much higher now.

Re:Nvidia – open-source auto-ontology by Twirrim on Tuesday November 03, 2026 @05:48PM (#50359332)
Sorry, yeah, I completely failed to say in my message what I was originally aiming to say. Lost sight of the message part way through writing, after I got pinged at work for something.

The fact that I had to throw an LLM at it to cut down the slop and turn it into something more succinct and comprehensible is nuts. For myself, I'd be embarrassed if I released something with that kind of README on it.

Re:Nvidia – open-source auto-ontology by ewild on Tuesday November 03, 2026 @01:06PM (#50355682)
what happens when your column names dont accurately represent what data is being stored (i know i wish it did but it doesnt at my place right now) and its all esoteric knowledge held by our data scientists? my worry is now the guys above will think its answering their question without knowing that its either A. wrong or B. data they may not be able to use for legal reasons on the project they are trying to make.
Re:Nvidia – open-source auto-ontology by gavinray on Tuesday November 03, 2026 @02:01PM (#50356565)
In the product I work on, there's a Metadata layer that maps user-supplied definitions, notes and commentary onto the raw table descriptions

These are optional but it allows manual override when schema design is poor

Re:Nvidia – open-source auto-ontology by adolph on Tuesday November 03, 2026 @02:04PM (#50356602)
This is the key issue. I call it a semantic gap.

Example is a transaction application with a backing database: The users conduct information interactions with the application. The application stores relatively sparse data into the transactional database. The database has just enough metadata to route data from transactions but doesn't have the full meaning.

Example database column "updated_date" with type timestamp: Without looking at the data, how do you know if this column is populated when the application first stores data in a row? How do you know the timezone of the timestamp as a user would see it?

Real world example: Patient height in app and api is SI and both note the units. Patient height in the data warehouse is US Customary with no unit notation. The app and api make elaborate conversions for the underlying transaction database's US customary storage.

I think the promise of an LLM ontology is to read application code to understand the context for any particular database column. The linked software is not that.

Re:Nvidia – open-source auto-ontology by shostack on Tuesday November 03, 2026 @10:29PM (#50361209)
I deal with this in a marketing analytics org.

Something like autoontology would help with human enforced governance for "blessed" data sets. You still need the operational rigor and data maturity to build said data sets and drive the transformation to adopt them.

This is effectively a tooling and labeling layer. But it will not magically solve the messy people issues that have always been at the root of this.

Re:Nvidia – open-source auto-ontology by zerop on Wednesday November 04, 2026 @01:27AM (#50362083)
Discovering ontology in complex real world large systems (involving multiple microservices) is quite hard. Best way IMO is to learn the database schema details, cross table joins etc from source code itself as it also has business context. This has worked for me quite well. But there are many other challenges like KPIs have fixed queries, cross product queries, pulling large data and getting right architecture for joining in-memory.