This talk was given May 29th, 2024 prior to version 1.0.0 in June 2024.

For anyone trying this on version v1.0.0: The function now isn't called 'boilerplate' anymore. It is 'quack'.

solid talk - so much value and great presentation

Thank you for sharing your experience - very informative.

The on-disk vector format can be revolutionary, IMO.

Do you have any DuckCons in the US?

Great video again. DuckCon videos have all been great. Duckdb makes working with Parquet files so easy. Over the next few years I believe that you are going to see many organisations moving away from data API's to instead publishing the data to Parquet files. It has so many benefits and .... is so, so, much easier. A great example is the likes of IMDB and TMDB. They, along with probably about 90% of companies and sites, have their full data sets hidden behind APIs which are a pain to use if you want say import large parts of or want an adhoc type query such as show me the movies both Tom Cruise and Bard Pitt have been in. If the data was instead a Parquet file you could simply use the likes of Duckdb to query it directly.

I love it! Such a fantastic solution! Does the SaaS platforms willing to accept or still work with a closed mindset? 🦆

Thank you. Very informative.

love it 👍

wow she’s amazing!

Exceptional work! Would love to have parquet query optimisations that compete with those of Clickhouse

Greatest database of all time.

Looking forward for duckdb 1.0.

Great talk! Thanks

very impressive software, my favorite by far!

Love the idea behind duckdbt. One use case that I find useful is to share one duckdb database with 2 or 3 services and use duckdbt as a means to get around the concurrency issues that arise otherwise (failed to get a lock on .db file) Also duckdbt is a terrible idea for the name. Even google corrects it to "showing results for duckdb instead"

Everything described is so bang on, excited for our Python duckdbt future!

I find this video entertaining!!

Congratulations, onward and upward!

Congrats on the inauguration! I love the rant, so on point with the idea of Flugscham for data. So much waste!

I used duckDB for a small analysis project lately. Loved it! Great job guys, big thanks for what you do!

bold move to showcase an in memory local database on a keynote which is sponsored by databricks who want to sell cloud cluster compute. :D

Great talk, did not know you were this good of a speaker Josh. Thanks for mentioning me, I was pleasantly suprised 😉

nice

Great talk, great presentation skills.

Using it for financial time series alongside Polars and like it a lot so far.

13:30 "Yo dog, I heard you like plugins" 😭😭😭

Enjoyed it very much

Awesome Presentation

Pandas wants to be Tidyverse when it grows up😮

What is this program called?

Where can I find dates and places did future Duck Con?

Where is the road map to 1.0?

Great talk! I am advocating DuckDB everywhere as much as possible - these talks make it easier to persuade others.

This is truly brilliant. I would struggle to name a more impressive piece of recent software. There are so many user cases for this but one of the primary ones is for ETL. Import the data (from file, url etc) into duckdb, quickly enrich it using its amazing sql and analytical prowess, and then export the data to a modern format file or data store. Quite frankly, its user-friendly version of SQL embarrasses the likes of SQL Server and Oracle.

Hello Jessica, very nice explanation. I want to know a little bit about byte cache concept that you mentioned in your talk to cache the parquet files. Can you please explain it?

So much cool stuff, I love it 😃

This talk is amazing! The MDS in a box is such a refreshing concept and so incredibly powerful with DuckDB. Also, the delivery of the jokes was impeccable. Top-notch stuff.

nice. hello from indonesia 👍

I am also interested in the slides.

Doing the things you love is really fantastic no matter what the result will be, enjoy!

Are the slides available somewhere?

Slides ?

Thank you!