rekoros

Bulk of the series

And a good day to you, sir!

yes, exactly!

After using NixOS for a bit and eventually learning to rely on it, life without it seems inconceivably complex

When I was in college in 2001, I went to the library and checked out Kaare Christian's book called "The UNIX Operating System". One of the early chapters covered vi - I'd telnet into the school's Sun server with a pretty early version of vi (one-level undo) and follow the examples. Never looked back!

This should go into dictionary as definition for "fuck around and find out"

Oban is great!

A true derailer

One way is to become an apprentice with an experienced plumber - ideally skilled in hydronic heating.

I've been using daisyUI for a couple of years and really love it - it's a quick, reliable way to rig up a nice looking/functioning UI with minimal work, while making things appear reasonably uniform across pages/workflows.

The combination of Tailwind and daisyUI made it possible for me - a backend developer - to pretentd to be somewhat competent in frontendland, which has been incredibly handy, work-wise.

I've been using Azure's "Document Intelligence" thingy (prebuilt "read" model) to extract text from PDFs with pretty good results [1]. Their terminology is so bad, it's easy to dismiss the whole thing for another Microsoft pile, but it actually, like, for real, works.

[1] https://learn.microsoft.com/en-us/azure/ai-services/document...

You're dealing with much larger datasets than I have, so far - mine is only a few million vectors. I have a hard constraint on resources, so had to get things working quickly in a relatively gutless environment.

pgvector - I wasn't able to get HNSW to build/rebuild quickly [enough] with a few million vectors. Very possibly I was doing something wrong, but fun research time ran out and I needed to get back to building features.

I ended up breaking up/sharding HNSW across multiple tables, but I'm dealing with many distinct datasets, each one just small enough to make HNSW effective in terms of index build/rebuild performance.

The article suggests IVF for larger datasets - this is the direction I'd certainly explore, but I've not personally had to deal with it. HNSW sharding/partitioning might actually work even for a very large - sharded/partitioned - dataset, where each query is a parallelized map/reduce operation.

Yep, I believe it

HNSW has been great for relatively small datasets (a website with 30K pages), but it’s a dangerous stretch for anything bigger

I've had browser and Linux VM logging to HyperDX with great success, but have been struggling to get OTel logging working with Azure Functions. Turns out, new (currently in Preview) "Flex Consumption" functions [0] natively support OTel and work with HyperDX.

[0] https://azure.microsoft.com/en-us/updates/public-preview-azu...

This may be the one area where some form of autocracy has merit :-)

We've been using Nix flakes and direnv (https://direnv.net/) for developer environments and NixOS with https://github.com/serokell/deploy-rs for prod/deploys - takes serious digging and time to set up, but excellent experience with it so far.

Great work and writing - thank you!

I’ve been working on (with a couple of friends) a one-click observability solution for Supabase: https://supafana.com

This our first experience in infrastructure SaaS (we opted for Azure) - way harder than we planned for, but seems to work!

I don’t fully understand the words you’ve used, but it feels that RU market is too small to justify building plants?

If you’re looking for an open source feature flag option, also check out https://www.growthbook.io/

Talking to my 24 y-o daughter:

- Have you ever heard of ICQ? - No

I don’t know, but I doubt that they’re so smart that they’re the only ones who “deserve”, somehow, to be blessed with functional CS101.

I heard that it’s a special form of torture, since those CS programs must be hard, but I don’t really buy it.