81–90 of 296 posts
All the praise for SvelteKit in this thread peaked my interest. But where would I start? I'm currently looking around to find a modern "toolkit" to write a new webbased frontend for our traditional desktop app. It's goin to be more or less CRUD with lots of tables/grids. Currently I favour vue.js with Quasar because of all the tooling and ressources it offers. It looks like it's easy to start with a traditional "navi…
This is how I move fast and break nothing. By having fullstack type-safety from database all the way to the frontend with auto-completion. My current stack: + SvelteKit (could be Next, Nuxt, Solid or any other TypeScript framework) + tRPC (typed calls between frontend and backend, https://trpc.io ) + trpc-sveltekit (glues SvelteKit and tRPC, https://github.com/icflorescu/trpc-sveltekit ) + Prisma (ORM, https://www.pr…
Anyone know a good way to solve Prisma memory usage? We use a database per customer and each one adds 50-80mb of RAM. Our backend currently consumes 2-3GiB of RAM to host less than 100 customers... App server has GC pauses of 50-80ms as a result of the memory usage.
There is a plethora of javascript frameworks: React, Vue, Svelte, Remix, etc. If I know nothing about front-end development, and would like to learn one that is: - Intuitive - Suitable for small projects as well as large projects. - That is here to stay, i.e. either adopted by many companies, or its adoption curve is going up. Which one should I pick? Would Svelte be a good choice?
- It works in small and large applications
- Used in majority of new frontend apps
- Reasonably intuitive, especially if you stick to functional components and aren't messing around with low-level renders
That said, I enjoyed building a web app with Svelte. It's extremely simple and powerful. Its only disadvantage is it is new and has a fraction of the market share.
This is how I move fast and break nothing. By having fullstack type-safety from database all the way to the frontend with auto-completion. My current stack: + SvelteKit (could be Next, Nuxt, Solid or any other TypeScript framework) + tRPC (typed calls between frontend and backend, https://trpc.io ) + trpc-sveltekit (glues SvelteKit and tRPC, https://github.com/icflorescu/trpc-sveltekit ) + Prisma (ORM, https://www.pr…
Personally, I'd remove tRPC on down (and choose a database). It's nice there's type-safety between layers, but that's a bunch of layers that you don't even need. (A layer that doesn't exist takes is 100% type-safe and takes 0 hours to develop and maintain).
Great news and looking forward to how this goes. I still find it incredible that the frontend javascript ecosystem hasn't found a reliable and productive contender yet, a framework you could be sure would be still relevant without too much changes in 5+ years, something like what Ruby on Rails is for full-stack. Some would say that React/NextJS has this role but I have to disagree. When using React/NextJS you still h…
> you still have to rely on third party library for routing, state management, querying etc. Take a peek at Joystick [1]. I designed it as the exact antithesis to this. It's full-stack (Node.js back-end with a fully ssr'd component framework designed for the long-term) and doesn't introduce any special paradigms (i.e., vanilla HTML, CSS, and JavaScript—no attribute tricks/compilers needed). Everything is designed to…
Earlier quoted context omitted.
> It embraces your knowledge of plain ole CSS/JS/HTML Yeah that's one of the things I love about Svelte. It tends to enhance fundamental web knowledge rather than forcing you to think in convoluted ways. Features like actions put you close to the bare metal so to speak. I use those all the time.
This is what they said about React when it first blew up. Compared to Angular, it's much closer to being *just javascript™*. But I wonder how well that'll hold up after the honeymoon phase The builder.io team made a tool called Mitosis[^0] that lets you input a component written in almost any framework and automatically recreate that same component in any other framework including Vue, React, Qwik, Angular, Svelte, R…
FWIW, I think that "it's much closer to being just javascript™" statement has absolutely held up over time (Angular, esp. the old angular.js is painful compared to React for this reason), AND that Svelte is likely a step closer again (which they can do because they have an entire compiler and aren't reliant on embedded a DSL inside of JavaScript).
There is a plethora of javascript frameworks: React, Vue, Svelte, Remix, etc. If I know nothing about front-end development, and would like to learn one that is: - Intuitive - Suitable for small projects as well as large projects. - That is here to stay, i.e. either adopted by many companies, or its adoption curve is going up. Which one should I pick? Would Svelte be a good choice?
If you're learning a new framework and you're concerned about stability React is still definitely the way to go. I think both React and Svelte are good at depending on generalizeable front-end skills that can carry over to other frameworks though so if your career doesn't depend on it, I don't think learning Svelte is a waste of time at all
There is so much "let me just see if this works..." tap tapppy tap .... "no... NO WAY... OMG IT WORKED!" with Svelte. Very little surface area. It embraces your knowledge of plain ole CSS/JS/HTML and empowers you with reactivity and a means of if being able to add motion to your ui. Newbs and Pros alike can build fast with it. That speed + reactivity allows your software to better keep up with your converstaions that…
There is a plethora of javascript frameworks: React, Vue, Svelte, Remix, etc. If I know nothing about front-end development, and would like to learn one that is: - Intuitive - Suitable for small projects as well as large projects. - That is here to stay, i.e. either adopted by many companies, or its adoption curve is going up. Which one should I pick? Would Svelte be a good choice?
The headwinds against React are strong.