4 ms·

Why developers hate new JavaScript frameworks?

4 points by Raed667 4y ago

rektide 4y ago
I chalk 50% of it up to general anti-js and anti-web attitudes. It's another easy target, a sign that, oh no, there's a place that isn't well groomed, well tamed, well under control, like- as this gist proposes- the safe secure well-patterned language with only a couple frameworks that I use is. It's crazy to remember though that Backbone was introduced in 2010, largely credited as the first JavaScript framework. And there was indeed a manic cambrian explosion, a half decade of more more more. And a lot of the ideas were only part-way baked. There was a thread on Why Devs Cant Sell[1], and I think some of what I said there[2] applies here. (Geeky) devs are high-context creatures, and businesses/industries/fads push us to higher-level, more cookie-cutter modes of production. Geeks should have curiosity & want context. What are the tradeoffs? Where are the weak spots, places we haven't explored? It's murky what we are buying into. Too many frameworks, imo, Sell, and not enough inform; they don't target the alpha-geeks taste-makers that would relish being fully briefed in how it got built, what the tradeoffs were, what actuation points there are under it. A lot of these frameworks look like they're being sold to suckers who are ready & raring to go swallow the marketing grade swill, rather than to intelligent & smart folk who will understand & properly appreciate what they are "buying". From a comment[3] in a thread on new breed of JS state management libraries: > The reality is that the community hypes certain tools, but in practice, they tend to have gotchas buried far beyond the surface level demos and documentation. The problem with that is that you only figure that out after committing to those tools and using them. This leads to tool abandonment, or in some cases, developers taking a swing at their own version. What they come up with is more often than not a rehash of the old ideas but lacking any "why" or long-term vision. Where-as libraries tend to solve distinct & clear problems, Frameworks outline a whole dwelling for your app, and the bounds of use are going to be very wide. It's taken millions of developer-hours to suss out & explore React, with various phases of exploration re-figruing shit out again and again (Render Props, Higher-Order-Components, Hooks, State Management). Having a lot of communal, shared learning time is incredibly important. But figuring out how to get that battle testing, how to understand the real merits & problems with what we create, learn where things do adapt & fit & where they are course or stiff or not fitting: few projects even attempt an earnest portrayal, have material to list, explore what context they do cover, much less indicate, Here Be Dragons. [1] https://growthlab.so/why-devs-can-t-sell https://growthlab.so/why-devs-can-t-sell https://news.ycombinator.com/item?id=31936015 https://news.ycombinator.com/item?id=31936015 (36 points, 1d ago, 15 comments) [2] https://news.ycombinator.com/item?id=31936177 https://news.ycombinator.com/item?id=31936177 [3] https://news.ycombinator.com/item?id=31960694 https://news.ycombinator.com/item?id=31960694

4 points by Raed667 4y ago

I chalk 50% of it up to general anti-js and anti-web attitudes. It's another easy target, a sign that, oh no, there's a place that isn't well groomed, well tamed, well under control, like- as this gist proposes- the safe secure well-patterned language with only a couple frameworks that I use is. It's crazy to remember though that Backbone was introduced in 2010, largely credited as the first JavaScript framework. And there was indeed a manic cambrian explosion, a half decade of more more more. And a lot of the ideas were only part-way baked. There was a thread on Why Devs Cant Sell[1], and I think some of what I said there[2] applies here. (Geeky) devs are high-context creatures, and businesses/industries/fads push us to higher-level, more cookie-cutter modes of production. Geeks should have curiosity & want context. What are the tradeoffs? Where are the weak spots, places we haven't explored? It's murky what we are buying into. Too many frameworks, imo, Sell, and not enough inform; they don't target the alpha-geeks taste-makers that would relish being fully briefed in how it got built, what the tradeoffs were, what actuation points there are under it. A lot of these frameworks look like they're being sold to suckers who are ready & raring to go swallow the marketing grade swill, rather than to intelligent & smart folk who will understand & properly appreciate what they are "buying". From a comment[3] in a thread on new breed of JS state management libraries: > The reality is that the community hypes certain tools, but in practice, they tend to have gotchas buried far beyond the surface level demos and documentation. The problem with that is that you only figure that out after committing to those tools and using them. This leads to tool abandonment, or in some cases, developers taking a swing at their own version. What they come up with is more often than not a rehash of the old ideas but lacking any "why" or long-term vision. Where-as libraries tend to solve distinct & clear problems, Frameworks outline a whole dwelling for your app, and the bounds of use are going to be very wide. It's taken millions of developer-hours to suss out & explore React, with various phases of exploration re-figruing shit out again and again (Render Props, Higher-Order-Components, Hooks, State Management). Having a lot of communal, shared learning time is incredibly important. But figuring out how to get that battle testing, how to understand the real merits & problems with what we create, learn where things do adapt & fit & where they are course or stiff or not fitting: few projects even attempt an earnest portrayal, have material to list, explore what context they do cover, much less indicate, Here Be Dragons. [1] https://growthlab.so/why-devs-can-t-sell https://growthlab.so/why-devs-can-t-sell https://news.ycombinator.com/item?id=31936015 https://news.ycombinator.com/item?id=31936015 (36 points, 1d ago, 15 comments) [2] https://news.ycombinator.com/item?id=31936177 https://news.ycombinator.com/item?id=31936177 [3] https://news.ycombinator.com/item?id=31960694 https://news.ycombinator.com/item?id=31960694