In the fast-moving world of AI-generated software, it's hard to tell signal from noise. Nick writes about what it means to have substance and how to avoid appearing like slop.
https://t.co/D7asosdR0H
Introducing Lazy Flush. Flush later for a 25% performance gain in TPC-C and improved Sysbench write benchmarks.
Read how we did it: https://t.co/r1VBGUcOXm
Hotdog. Not hotdog. ISP. Not ISP. π
We tested TypeSafe's Jev, their new System 1 model, in our internal leads pipeline to make fast, structured decisions that help filter out hosting providers and ISPs.
Dustin shares the setup and shows how it works π
https://t.co/x71MiZ9KVk
In this blog post, Neil spotlights Argon, a version-controlled database built on MongoDB, and compares it against DumboDB.
He looks at performance, architectural differences, and how each fits your workflow and use case. https://t.co/sy9dXwtXH3
Doltgres now supports server-side cursors!
This allows you to fetch rows in batches, move through results, and keep a consistent snapshot even when you switch branches within your session.
Closer to full Postgres compatibility. Zach gives a demo https://t.co/Bm3xkdoVj9
DumboDB's Mongo compatibility took a big leap forward.
Version 0.8.3 adds 29 operators, a new aggregation stage, and many other improvements.
Neil shares the progress, how it works, and what is coming next for Dumbo: https://t.co/6vI3LkWjBp
DoltLite is now on DoltHub! π
Work locally, commit offline, and push to DoltHub so other devices can pull your changes. Local work, shared progress. Built for agent workflows.
See the setup: https://t.co/hE97sQ3YVO
DoltLite is being used by @imbue_ai! π datalib mirrors your personal data into one place you own. Each store is a DoltLite DB, so every sync becomes a commit, keeping pipelines incremental and tracking every change!
Tim tried it π https://t.co/plumMSGcNm
Last week, we released an explainer video where Dolty walks the audience through how Dolt works. Here is a walkthrough of the framework and four questions to keep in mind throughout your creative process π
https://t.co/DOmVbEgELm
What happens when a commit graph has multiple merge bases? Dolt 2.6.0 adds dolt_merge_bases() and virtual merge bases to prevent silent merge errors. Neil explains with interactive diagrams: https://t.co/nQHKoB9r8f
Happy (belated) first birthday, Beads!
One year (or roughly 2,700 AI years) after launch, the original agentic memory tool continues to help coding agents work longer and remember more.
Congratulations to @Steve_Yegge and the Gas City team! π https://t.co/BnCSvOvEmN
Take a tour of Dolt's internals as our mascot, Dolty, shows you how traditional databases differ from the powerful Prolly Tree architecture that makes Dolt's version-controlled databases possible.
https://t.co/d4A2oH7j66
Dolt 2.5.0 now supports atomic transactions across multiple branches!
Zach demos how to update multiple branches in one transaction, commit them together, and use dolt_commit_all() to create a Dolt commit on every edited branch!
π https://t.co/u3jFkrv4qb
DumboDB, our version-controlled MongoDB-compatible database, now supports backups!
Back up and restore your databaseβs data state, including branches, commit history, and uncommitted changes. Protecting your DumboDB data has never been easier.
π https://t.co/B7CtbGtrKN
Doltgres is now natively supported in DoltHub!
Bring Git-like collaboration, pull requests, and row-level diffs to your Postgres-flavored databases.
This blog shows you howπ https://t.co/cQtOM7Yvq2