Storing #RDF or #LinkedData is the easy part (I am working on #activitypub client, so everything is JSON-LD anyway), but as soon as I need to formulate queries like "Give me list of the activities from all the actors who are on the following collection of my user, sorted by published/updated date", the amount of data mangling that I need to do is already making me think "ok, why don't I just push all of this to sqlite?"
Oh _no_
in #CouchDB, Mango querying does not mix with using "" as document key.
"One or more conditions is missing a field name."
Stressful start to the year?
You have one less worry with #CouchDB 
Our latest article covers @couchdb’s #checksums feature: a built-in safeguard that helps protect your data from the impact of ever-dreaded disk corruption.
It’s just one of the features that makes CouchDB your data’s safe place.
Full post on our blog: https://neighbourhood.ie/blog/2025/01/22/how-couchdb-prevents-data-corruption-checksums
We published a four part series about how to build a real-time multi-user Kanban board with @couchdb and @sveltejs
Its is complete with code for all steps, so it’s not just a draw-the-rest-of-the-owl tutorial:
1. Setup, intro, first steps: https://neighbourhood.ie/blog/2024/12/05/realtime-multiuser-kanban-board-with-couchdb
2. Automatic Conflict Resolution: https://neighbourhood.ie/blog/2024/12/11/automatic-conflict-resolution
3. Manual Conflict Resolution: https://neighbourhood.ie/blog/2024/12/18/manual-conflict-resolution-with-couchdb-and-svelte
4. Avoiding Conflicts with Resource Locking: https://neighbourhood.ie/blog/2025/01/15/resource-locking-with-couchdb-and-svelte
For 12 years already, I've been on/off experimenting (and prototyping) to create a compendium & knowledge graph about computational & generative art, the different movements/genres/mediums (e.g. architecture, design, visual art, audio/music, sculpture, kinetic/motion, film/animation, text etc.), influences, definitions/references, people (artists, curators, writers, thinkers), collectives, organizations (foundations, galleries, museums), exhibitions/festivals, tools & tool makers, common techniques used, art platforms — everything tagged and also supporting to browse by time (centuries & decades, currently starting ~1600) and region...
Attached are some screenshots of my first prototypes from 2011/12 (using #CouchDB, then #neo4j for storage & my own tools for force-directed graph layout) and of the current prototype using #Logseq (https://logseq.com)... The latter is working great for now and feeling, I'm getting somewhere this time, also because I have to make it work (for work!). This is all still just a beginning, hundreds of more people, orgs, projects & references to import and re-check from older versions. The current contents are _very_ biased to my own network/trajectory in/through this space...
Ps. Following up with all the folks & materials I've included already, I'm realizing again and again just how I've been the most naïve and _worst_ person to monetize (mostly not even trying!) my art/contributions... For 20 years I've filed 90% of hundreds of my projects under "experiments", "sketches", "demos", "tutorials" (often also to help illustrate techniques of my open source tools), only to realize (not for the first time) almost everyone else of my old peers has been way less selective and been attaching way more importance to all of their outputs... Maybe one day I will learn, before it's too late...
#QuickJS and #Nouveau are a couple #CouchDB 3.4.0 + 3.4.1 updates you can see at our Oct 9 Meetup.
@janl will shed light on 3.4.1 + where 3.4.0 went. Alex Feyerke and @ninette will detail new features:
Talk #1: CouchDB 3.4.0 / 3.4.1 Release Retrospective (Jan Lehnardt)
Talk #2: Introducing QuickJS: CouchDB’s New JavaScript Engine (Ninette Adhikari)
Talk #3: First Steps With Nouveau, CouchDB’s New Fulltext Search (Alex Feyerke)
We’re excited to host you!
Sign up: https://vi.to/hubs/couchdb-berlin
Maybe something along the lines of #couchdb I imagine? But also for files (complex objects) and a standardised protocol.
ICMY: I’ll be dropping some serious #CouchDB knowledge this Wednesday.
All of the things I’ve learned working on and with CouchDB since 2007 when it comes to database and document design.
Many bits also easily translate to other Document DBs and Key-Value Stores, so this is going to be useful even beyond CouchDB.
Sign up below 
https://narrativ.es/@janl/112406412346562343
From the Linux Update newsletter: Open source database management systems offer flexibility and lower costs while avoiding vendor lock-in, but how do you choose the one that's right for your organization? https://www.linux-magazine.com/Issues/2024/282/Data-Management #DBMS #Firebird #MySQL #PostgreSQL #MariaDB #NoSQL #CUBRID #ApacheCassandra #MongoDB #JanusGraph #Redis #CouchDB
@javahippie bis Du dann bei #CouchDB landest @janl will approve
Anyone tested PouchDB in the browser and set it up so it syncs with CouchDB? I'm trying to make a minimal thing and was wondering if it could sync a blob/attachment. That way I could get away with some html/css/javascript on github.io and only CouchDB as a webserver.
I got nerdsniped by @tef & @sushee to demonstrate how relatively easy it is to build a CRAQ (https://timilearning.com/posts/mit-6.824/lecture-9-craq/) system atop @couchdb — Showing that as much as CouchDB is a distributed database in itself, it is also a very nice toolkit to build other kinds of distributed databases with it :)
The demo is in Node.js with some private JS/HTTP API glue that is not relevant to the demo: https://gist.github.com/janl/d0ba3544631e054e59fbd5f4d3ee96f8
Let me know if you ahave any questions.
Sneak peek for a new thing we made, formal announcements will commence tomorrow: https://neighbourhood.ie/products-and-services/structured-query-server #CouchDB #SQL #HellFrozeOver
