It's my personal opinion but I think these days, unless you're web app doesn't require any JS, manipulating DOM manually is a very risky approach. It's very easy to get into a spaghetti call stack if you e.g. need to fetch and interpolate some data into DOM (especially when constructing lists). Having said that, I see why shifting towards roots using pure JS + HTML + CSS might look attractive. And I see how CV-driven development became so popular, and that over-engineering and using fancy tech-stack takes precedence over the product/feature being implemented. Luckily there are still minimalistic rendring libraries, micro frameworks, starter boilerplates or even entire self-contained tools which purpose is to make you productive.
It's my personal opinion but I think these days, unless you're web app doesn't require any JS, manipulating DOM manually is a very risky approach. It's very easy to get into a spaghetti call stack if you e.g. need to fetch and interpolate some data into DOM (especially when constructing lists). Having said that, I see why shifting towards roots using pure JS + HTML + CSS might look attractive. And I see how CV-driven development became so popular, and that over-engineering and using fancy tech-stack takes precedence over the product/feature being implemented. Luckily there are still minimalistic rendring libraries, micro frameworks, starter boilerplates or even entire self-contained tools which purpose is to make you productive.
Thanks for your opinion on this. I think it is safe to do simple DOM manipulating without a framework.