Introduction
2024 is upon us, and we try to plan for the new year and think of the things we could do or learn in the upcoming year. The...
For further actions, you may consider blocking this person and/or reporting abuse
Nice article, I was mostly curious about htmx because I keep hearing about it -> I wish though that article had more code examples, maybe one code example for each framework!
fixed :)
Ha ok nice!
Are you cousins or something?
almost - twin brothers! :)
I'm the older one though.
I've got you covered with htmx code examples here, especially in the "Common Patterns" section: mvolkmann.github.io/blog/topics/#/...
I'd love to see some complex HTMX web sites to try out for my self. I just feel there is a cut off point where it starts making less sense to use HTMX than full-blown JS frameworks.
See my htmx blog post here, especially in the "Common Patterns" section: mvolkmann.github.io/blog/topics/#/...
For the point about Solid.js compiling to vanilla JS, isn't that true for all frameworks?
Whether it's using a virtual DOM or not, isn't it still just running JavaScript code in the browser? (excluding WebAssembly, which I don't think comes into play here)
Good point. I mainly wanted to emphasise the fact that Svelte doesn't use virtual DOM unlike React.
Exciting stuff. I went deep into porting my app to Qwik (City), Solid (Start) and Svelte (Kit) this past year, and they're all great until I eventually hit a snag or friction due to their respective ecosystems not yet being fully developed, and the common factor for me was that I was less productive than on React (despite much preferring the DX of all three to React). Solid would be my choice of the three, but Start hasn't even shipped yet and is still churning. Qwik is the most innovative and has the best DX IMO, but it doesn't seem to be catching on (e.g., still no official tanstack support), and some of it's design choices complicate 3P component development (e.g., github.com/chakra-ui/zag/discussio...). Svelte? Well, I just much prefer JSX, and it's still baking (Runes forthcoming, and zero interest in refactoring my code, even if Svelte appealed to me, which it does not). I've since gone back to React (Remix), and will wait it out another year or two before making the switch, if at all.
The big part of appeal of React is the ecosystem that developed around it mainly because of its longevity. Over time, ecosystem of other frameworks will hopefully catch up.
Very true. Who will be the next React and gain critical mass is anybody's guess right now, which is why I've opted to wait it out despite my natural inclination to want to jump on board with latest shiny new thing.
Great article Boris. It's interesting how you think Svelte is easy to learn - we used it in one project, and since it was implemented, I regretted it 🤦♂️ to me, it seems very unintuitive and complex
Thank you! I pointed out that Svelte is easy to learn because it is intuitive to get into it after learning basic HTML/CSS/JS since the structure is similar.
What created issues for you? Was it that the project was too complex, or something else?
Maybe it's due to the fact that we used it to create a VSC extension. Not sure how it works on a regular website.
Great post! Thank you! 👏
Remember that since HTMX does things in a bit of a new way it may have accessibility concerns that have not yet been tackled. The other solutions are more proven on this regard.
That means if your project has any possibility for legal a11y requirements then you have to pay extra careful attention whether you can use the tool.
Note that even the more proven solutions may still be a bit lacking, such as client side routers do have issues in frameworks like Next so they are not fully accessible by default. Issues are like focus being in the wrong place, or page not scrolling to top when client side routing happens. However you will never have this issue if you rely on traditional HTML page load.
That is really useful information. a11y must be taken as one of the things to consider when choosing your framework as well.
Cool post bro! Thx for sharing
Can see why all of those were picked, but I'll just stick with getting more proficient in Nuxt.
I'm super excited to see where HTMX goes. I've used it for a few Django projects before, and I absolutely love it.
I like HTMX when learning a new language and I want to play around with making a simple http website
nice!!!
can we consider 2024 the year for something new or something good? in any case, web3 will not stand still, but will only progress
I'm a little disappointed that there are only javascript frameworks in this listing.
Which ones would you recommend to look at? Blazor, Yew, Flutter Web?
I like Phoenix and Ruby On Rails. Ruby on rails is still a great chose!
This is a really good overview, Boris! I keep hearing about Solid and Svelte but never heard an easy explanation of how they differ and what are their main strengths. Thanks for clearing that up!
This article as very helpful. Provide and bring some exciting and valuable tools to my attention. Thank You!
Check out AHA stack if you would like to couple HTMX, 💯
After many, many years, Rails still stands out as the best framework overall.
primatejs.com is the one I'm currently involved with.
Nice article @martinovicdev! wasn't aware of Qwik, will give it a look.
Thanks for sharing.
Svelte will also hit version 5 soon where the syntax will be a bit different, but it will solve a lot of issues. You can look at the new syntax at component-party.dev/
Gotta say... Svelte has changed the game for our shop.
Great Post 👏
About htmx its still a javascript and and can be coupled with other frameworks to achive best.
Have to check wasp!
Solid js could be loosing its charm!
Astro is a King maybe in coming days they are going to release astro database
Svelte 5 is just great
Quik 2.0 is noice
Conclusion
People Still going to use react 😂 and react 19 is also coming
Finally someone is not missing qwik!
This is the one and only framework for me. The developer behind it are so skilled.
If qwik would be number one, the internet would be so much more performant.
I am all about back to the basics.
I feel like HTMX itself is not complete, more importantly is what you pair HTMX with on the backend. Some people are advocating for HTMX + Golang, which I'd really love to try one day.