I'm mostly using GitLab right now, both self-hosted and their hosted platform. But am curious about Cursor Origin and certainly plan to try that out when it's available.

Go + Datastar is simpler, I much prefer it.

Yeah this combo fits as easily as a custom sized/fitted gloves would.

I would add a-h/templ for templates and it has Datastar support as well.

It is staggering how Datastar is not more popular.

I think people see the "pricing" and are turned off. The base is completely free! Some of the ecosystem is not. The big selling point for me, is the size of the library. HTMX and Alpine are a bit larger. The downside for me is needing to have a template type that supports streaming. (Thymeleaf in Java vs JTE) My controllers also have imperative code to support the streaming. None of this is enough of a turnoff however. For my simple apps, I'm extremely happy with how simple both HTMX and Datastar make my logic. (I replaced a React front-end and got rid of the whole build step).

I also prefer Datastar but the new features coming in HTMX4 closes the gap more. Happy to see the ecosystem converging on good patterns.

But the point is that Datastar only exists because htmx rejected all of those ideas. Now it's becoming a cheap, more complex, less powerful, heavier copy of Datastar. Just use the genuine article.

Last I checked Datastar doesn't support progressive enhancement. I like using <a href hx-boost=true> and <form method action hx-boost=true> so almost all my functionality works even without JavaScript. See https://dev.to/yawaramin/why-hx-boost-is-actually-the-most-i...

Its not evident to me that that is really a concern when datastar is so small, and most of the js is written inline.

Moreover, htmx can only do a fraction of what datastar does - you'd need to add on something like alpinejs to be able to try to fill the gaps. Both combined are much larger than datastar, and not nearly as integrated.

It’s a concern because I use htmx for loading and updating resources and I want those operations to work as well as possible even without JavaScript. I’ve never felt the need for Alpine.js, I have very little I need to do with custom JS, and it’s trivial to maintain that by hand.

I would just go all the way to using https://unpoly.com/ if I wanted the full-bang whiz.

Datastar has evolved into a suite of tools [1]. If anything the gap has widened.

Off-course these additional tools are not free [2].

[1] Rocket for WebComponents, Stellar for CSS

[2] Nothing wrong in charging but $349 is a tad too expensive for me. I will happily pay if it was in double digits.

While I do need E2E and SSO making Chatto not suitable, it looks pretty cool. NATS is cool as well. Hope to see it doing well.

Curious. Most places needing SSO also have legal retention requirements, making E2E a hard nonstarter. And most E2E users wouldn't want to rely on a SSO that could likely revoke their keys. What's your environment?

Chatto does have full SSO support (OIDC and some selected providers like Google, GitHub, and, ironically, Discord, with more coming in the next releases.)

It works well if you connect it the Brave Search API, but using it a scraper is fairly unreliable. Google stopped working a few days ago.

I never run long horizon tasks. So Composer 2.5 is great.

NetNewsWire is part of my daily routine. The best feed reader out there with support for iCloud Sync and macOS native.

I understand Jeff's argument, but he is missing the fact that one of the features of the Framework 12 is the modularity of the components. So if that is not a valued feature in this scenario, sure it's hard to justify.

I love building and upgrading stuff as well as paying (much) more for tools that will last. But this is a laptop not a socket set, paying (a lot) more for worse performance up front makes absolutely no sense. Seems like the argument should be the Framework 12 just shouldn’t exist.

> he is missing the fact that one of the features of the Framework 12 is the modularity of the components

He does explicitly make that point.

> The biggest win is the modular ports.

I think what makes the perspective in the article interesting is that buying individual components a la carte isn't a good value in today's market. Sure you can upgrade the RAM and SSD in the Framework, but 16 GB of laptop DDR5 is $200 and a 1 TB 2230 SSD is another $200. The question becomes, is it worth it to spend 40% more for a laptop with 40% less performance (as well as worse build quality, a worse screen, worse speakers, worse battery life, and running hotter) so you can have the potential to spend half the price of the laptop to upgrade it in the future?

Brave Search is the best alternative because of their independent index, the AI results are still fairly good as well. And their browser rocks because of the built-in ad-blocking including for YouTube, and the ability to filter out all YT Shorts on the mobile app.

Been loving Zed for the last few months. The Dev Containers were the last thing I needed to switch over and been steady ever since.