DuckDB Labs released #DuckLake 1.0 - a data lake format that stores table metadata in a SQL database, rather than spreading it across object storage files.

Key features:
• catalog-stored small updates
• improved sorting and partitioning
• compatibility with Iceberg-style data features

Learn more ⇨ https://bit.ly/48PsPIS

#InfoQ #DuckDB #ApacheIceberg #AI #DataLake #DataStorage

Off the back of yesterday and looking at #DuckDB and the UI extension, I did a quite write-up.

Other than the issue I had around doing a lot of schema changes, it's a really nice UI which doesn't make you feel like you've ended up with the "entry level" interface.

https://dazfuller.uk/posts/2026-05-04-exploring-duckdb-ui/

Exploring DuckDB UI

I had a little time for a change and so thought I’d spend a bit of time exploring something different. One of the things that I find weird is the number of devs who say that they “can’t do SQL”. SQL is one of those languages that was created to allow business people to query a database using a language which makes sense to them. It has definitely grown since that lofty ideal, and of course, each database system has its own variation (called a dialect).

reality.foldLeft

Also been playing with #DuckDB UI this afternoon. It’s a great place to start for exploration, and if you don’t have any UI tools available to you.

Did find out that if you’re making a lot of schema changes then it gets itself into a mess and won’t open again. So had to drop and recreate the db file a few times.

Probably do a write up on it tomorrow. Along with a Microsoft Fabric RTI demo for later in the week

Over the last couple of months I had to do a lot of number crunching in order to write some #designdocs. While writing such docs is always useful, I recently added a lot to the fun-dimension with #Jupyter and #DuckDB 😅 This might be useful to others, so here's a quick summary of my setup with a simple example: https://zerokspot.com/weblog/2026/05/03/simple-data-analysis-setup/ #blogged
Simple data analysis setup

Oh look, another "groundbreaking" #blog post about #DuckDB from a self-proclaimed data wizard. 🙄 Apparently, the limitations of basic text queries are just too much for our hero, who bravely delves into the wild world of Full-Text Search. 🌟 Spoiler alert: it's as thrilling as watching paint dry on a data frame. 🥱
https://peterdohertys.website/blog-posts/full-text-search-w-duckdb.html #DataWizard #FullTextSearch #DataFrame #HackerNews #ngated
Full-Text Search with DuckDB - peterdohertys.website

Pete Doherty is a NYC based software developer

Full-Text Search with DuckDB - peterdohertys.website

Pete Doherty is a NYC based software developer

Full-Text Search with DuckDB - peterdohertys.website

Pete Doherty is a NYC based software developer

Here's a blog post about my initial experiments with full-text search in #DuckDB.

TL;DR It's a good, simple FTS solution for unconventional data sets. If you do need a more fully-fledged FTS solution (e.g. #postgres then there's a clear migration path.

https://peterdohertys.website/blog-posts/full-text-search-w-duckdb.html

Full-Text Search with DuckDB - peterdohertys.website

Pete Doherty is a NYC based software developer

I wrote a blog post about my initial rounds of experimentation with #DuckDB.

TL;DR I came away very impressed and am excited about using it in the future.

https://peterdohertys.website/blog-posts/dab-of-duck.html

A Dab of DuckDB - peterdohertys.website

Pete Doherty is a NYC based software developer

#duckdnng, a shameless #DuckDB extension which is LLM enabled copy pasta of the excellent handcrafted #Rstats #nanonext C code and API design