4 ms·
Svelte's philosophy on turning the virtual DOM concept inside out sounds like it has merit, and is very promising. But it's going to take a lot more than that,
by xtagon 7y ago
Svelte's philosophy on turning the virtual DOM concept inside out sounds like it has merit, and is very promising. But it's going to take a lot more than that, in my opinion, before a large number of people consider switching from React, Ember, etc.
I don't see that as a drawback, I see it as an open opportunity for Svelte to keep building out on improvements other than the DOM updates, and catching up with everything else the SPA alternatives provide that have nothing to do with the virtual DOM.
For example, Ember is just a joy to work with, and makes it easy to rapidly prototype reactive frontends in a way that reminds me of Ruby on Rails's initial appeal to developer happiness, and the tooling is very mature. If you could unlock all those benefits while keeping the blazing fast DOM updates, oh boy!
- hatch_q 7y agoTooling is the key word here. Svelte simply doesn't have the tooling needed for any big project. - testing/testability (unit-tests are easy, but what about functional, e2e?) - strong-typing support (flow, typescript) - good IDE support? - i18n? ICU support, etc? They need to redo what ember-intl or react-intl do. Without these things it's simply not viable to start bigger projects with new framework.