Hacker Newsnew | past | comments | ask | show | jobs | submit | andersmurphy's commentslogin


Fun read.

I've been tripped up by the where in partial indexes before. Same goes for expression indexes.


Thank you for saying that too!

Hope this explanation helped explain why, at least a little bit.


Don't forget Jruby and Groovy.

Touche!

And Clojure and Scala. So really Clojure is number 2. :-)

I mean the definition of death is when the light fades right? Everything else is just an approximation. Would be wild if true.

It would be wild because it would be wrong. We acknowledge someone who has suffered brain death to be truly dead, even if their body can be kept alive (and presumably shining!) for weeks and weeks through modern medicine. If this light is a side effect of biochemical processes then presumably someone who is decapitated will also continue to shine for minutes.

The fading of the light would be sufficient but not necessary for someone to be considered dead, so it would make a poor definition.


Sleeping probably also fades light.

It does require a mind shift as generally you have one connection morphing in all your updates and do CQRS.

The connection per component model that mesh uses is fine until you have concerns that cross across components (this was an issue hotwire also ran into before they introduced morph/refresh).

Instead you have one endpoint per page re-render the whole page via morph on every change. You still have backend components they just send requests up and get their updates via the sse connection for that page. Kinda like view = f (state) just over the network.


+1 for datastar. It lets you do really dumb push based streaming html stuff where HTMX isnt quite fast enough (great for multiplayer apps).[1]

Super excited about triptych too! Thanks for pushing that.

- [1] https://checkboxes.andersmurphy.com


I was quite excited for Datastart until I saw that they keep a good chunk of the functionality behind a proprietary license [1]. It's not even about the price (it is a bit steep though, but I guess if you use it in a commercial project it will pay for itself), it's the proprietary part.

[1] https://data-star.dev/reference/datastar_pro


A "good chunk of functionality" is a mischaracterisation. I don't use anything from pro and I use datastar at work. I do believe in making open source maintainable though so bought the license.

The pro stuff is mostly a collection of foot guns you shouldn't use and are a support burden for the core team. In some niche corporate context they are useful. You can also implement your own plugins with the same functionality if you want it's just going to cost you time in instead of money.

I find devs complaining about paying for things never gets old. A one off life time license? How scandalous! Sustainable open source? Disgusting. Oh a proprietary AI model that is built on others work without their concent and steals my data? Only 100$ a month? Take my money!


It's not about the money, it's about the fact that I would depend on vendor-locked functionality. I cannot maintain my own fork of it if for whatever reason I had to. I don't want to fork datastar if I don't have to, but I want to be able to do so if I have to.

Alpine does have a "pro" version as well, but it's just copy&paste code examples. I am not complaining about Alpine, and I even got my employer to pay the maintainer more than what the datastart guys are charging for their pro version. So again, this is not about money.


You get the entire source with pro. Nothing is stopping you maintaining your own private fork for use in your commercial product. If datastar the non profit disappear tomorrow you would still have the source. So where is the problem?

> your own private fork

That right there is the problem, it's maintainer lock-in, I cannot share my improvements with anyone else or benefit from other people's improvements.


So you'd prefer if it was free and GPL?

Your recent post on using datastar with SSE to power game of life [1] was excellent. Thanks for writing it!

[1] https://andersmurphy.com/2025/04/07/clojure-realtime-collabo...


Thank you! I've been meaning to write one on the nitty gritty of virtual scroll, hopefully I'll find some time next month.

That would be great to see. Likewise more details about your custom compression settings.

Datastar folks have some cool videos - https://data-star.dev/videos

HTMX does have web socket support! I've been experimenting with it...

Datastar uses SSE. In practice you almost always want SSE rather than websockets or your own UDP protocol. Websockets are an operational nightmare once you get off the happy path.

HTMX also has a SSE extension. I’m working on a multiplayer game as a hobby project and am not too far along.

> Either Ruby Central puts controls in place to ensure the safety and stability of the infrastructure we are responsible for, or lose the funding that we use to keep those things online and going.

Seems pretty clear after reading this. If 1-2 companies pulling funding is enough for them to force you to to what they want, its hard to stay independent.


But for scale you don't want multiple writers. You want a single writer so that you can batch.


Yeah sqlite runs circles around postgresql when it comes to writes if you know what you are doing. Single writer lets you batch and batching is king for writes.


Consider applying for YC's Winter 2026 batch! Applications are open till Nov 10

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: