home.social

Search

495 results for “zeab”

  1. Completely in on () usage. You can even do this with . 😄

    Leave updates only from SSE. Model all mutations, but one way to ingest updates as they are available. No poll, websockets, or error handling chaos. Just SSE. 😉

    yagni.club/3m475dwkjvc2o?auth_

  2. @mradcliffe I would say do look over in general.

    They have a web server that is pretty versatile. And if you know , the main parser, you can interop with your language of choice. Cooklang also have a parser that can make things simple. 😅

    I went over cookLang a while back, out of interest. Ended being absolutely fascinated by the dedication and work that has gone to the language for recipes. 😎

  3. Kind of wild has its own federation setup now. 😅

    Sure it uses . No fancy . Still, fascinating to see something like this.

    More of the web trying to index itself as primary search engines lose all their meaning to what search means. 🫠

    cooklang.org/blog/13-the-dishw

  4. And in a twist of design, here's doing durable execution directly with and an event bus framework in called .

    So just in case you all in go, this may be the way to go instead of . 😅

    threedots.tech/post/sqlite-dur

  5. 0.5 is even more reason just to pick . You can get pretty far with litestream and have it fully manageable for duplication.

    Lower the N tier of your app. 😉

    fly.io/blog/litestream-v050-is

  6. Profile control in is kind of outstanding. I have this very problem with monorepo dev work. Not all the time you need to load everything under the sun. 😅

    Going to see how to fit on these very repos. 😁

    devenv.sh/blog/2025/09/17/deve

  7. Here I go again. Diving deep into modern design.

    feature introduction for materialized views in their fork looks interesting, to say the least. 😅

    turso.tech/blog/introducing-re

  8. Here I go again. Diving deep into modern #rdbms design.

    #turso feature introduction for materialized views in their #sqlite fork looks interesting, to say the least. 😅

    turso.tech/blog/introducing-re

  9. Here I go again. Diving deep into modern #rdbms design.

    #turso feature introduction for materialized views in their #sqlite fork looks interesting, to say the least. 😅

    turso.tech/blog/introducing-re

  10. Here I go again. Diving deep into modern #rdbms design.

    #turso feature introduction for materialized views in their #sqlite fork looks interesting, to say the least. 😅

    turso.tech/blog/introducing-re

  11. It may seem simple enough to get on the browser. But with synchronization baked in, very different tale. work here is pretty amazing. Curious to see how far they can go. 😎

    turso.tech/blog/introducing-tu

  12. It may seem simple enough to get #sqlite on the browser. But with synchronization baked in, very different tale. #turso work here is pretty amazing. Curious to see how far they can go. 😎

    turso.tech/blog/introducing-tu

  13. It may seem simple enough to get #sqlite on the browser. But with synchronization baked in, very different tale. #turso work here is pretty amazing. Curious to see how far they can go. 😎

    turso.tech/blog/introducing-tu

  14. It may seem simple enough to get #sqlite on the browser. But with synchronization baked in, very different tale. #turso work here is pretty amazing. Curious to see how far they can go. 😎

    turso.tech/blog/introducing-tu

  15. Another project for in . Hold up! It's a fork of . Using instead of . 😅

    The difference here is how much datastar use (). It's EVERYWHERE. Very potent stuff.

    Hadn't yet tried datastar, but maybe soon with this. 😁

    starhtml.com/

  16. Found out that has utility, using notify-reload. And it works with . 😄 I now have the terminal opening as fast I once did with . I can't turn back anymore... 😅

    ghostty.org/docs/linux/systemd

  17. I'm totally with the idea.

    At dayjob, we have needed a emulator, but held back due to the exact point made: not core business. 🫠

    This I feel is more common than people realize. We get a common way, and tooling can improve all over. 😎

    mitchellh.com/writing/libghost

  18. Oh snap. v18 will support ! This is brilliant. Full filtering and sorting accessibility from the ID alone. 🤯 Truly can't wait. I been migrating all my ids to v7 already. I'm ready for the benefits. 😎

    thenile.dev/blog/uuidv7

  19. 3.0 is out. The live streaming ability looks kind of wild. Was not expecting to be able to stream my terminal usage. But now... I can totally see doing this. 😁

    blog.asciinema.org/post/three-

  20. Logical replication in is the secret sauce of data prowess. 😅 The WAL can be the difference between distress or just stress. 🤣

    usage for their search instance is prime example. Replicate and go. A whole lot saner than the full ETL path. 🫠

    paradedb.com/blog/etl_vs_logic

  21. Starting to learn how to use with for . It's an itch I had.

    Been meaning to build up everything with pyinfra. But I need tests. In world, I had tests for everything using Molecule. It worked well enough, but was quite complex.

    However, with pyinfra, the tests can be isolated integration test containers! 😎

    Now I can go crazy with the amount of tests for every operation done. 😆

  22. Found out about and kind of in love. It's simple enough. No fluff and can do operations with keybindings. Absolute joy. 😄

    Honestly, I don't know why I haven't been trying file managers. I'm like 99.9999% in the terminal. 😅

  23. Very good write up on and . These things have such good improvements in general.

    Not always is more ram feasible either. 😅 Looking at you gddr5 😒

    linuxblog.io/zswap-better-than

  24. On #k8s, the complexity is on knative rollout. But if you strip down to nothing, it becomes easier. knative serving CRDs with #kourier, and you get something that work for requests cross nodes easy. pro and con are, you now have a whole bloody k8s cluster. 🤣 Heavy at start, but light in maintaining after. Especially if this is a #talos or single node #microk8s/#k3s setup. 😅

  25. On , the complexity is on knative rollout. But if you strip down to nothing, it becomes easier. knative serving CRDs with , and you get something that work for requests cross nodes easy. pro and con are, you now have a whole bloody k8s cluster. 🤣 Heavy at start, but light in maintaining after. Especially if this is a or single node /#k3s setup. 😅