3 ms·

The last time I worked on a react native project my team and I wondered all the time if we were doing something seriously wrong or not. It seemed like very pop

The last time I worked on a react native project my team and I wondered all the time if we were doing something seriously wrong or not. It seemed like very popular and commonly used dependencies had poor community support. Upgrades were often painful. Breakage during routine package upgrades occurred far more than in our non-RN typescript projects. We did a lot of research and asked around but it seemed to be par for the course. Overall I’d guess that we still saved time over learning and developing with native solutions for iOS and android, but it left me feeling like the ecosystem could be so much more productive still. I’ll admit I recognized the need to contribute to libraries I saw were lacking support, but the project wasn’t my main focus and I don’t think my employer cared much. In the future I should definitely complain less and help more.

bromuro 5y ago
Putting a javascript layer over the native OS API never seemed a clever idea to me. I tried React Native years ago but it was like suffering twice - the usual dependencies-hell from npm by plus the inconsistencies 9f each platform. (also: react native apps never “feel” truly native. or is just me?)
- steve_adams_86 5y ago
  No, you’re right - they rarely feel native. I also agree about the layers of suffering. This was in line with our experience as well. The whole thing felt like cutting a log with a dull chain saw rather than using a really sharp pull saw. Sure we got there and it took a lot of labour away, but traded for another kind. It was a little faster. The end result wasn’t as nice. So… Was it worth it? I’m still not sure. It’s certainly worth it in cases where you’re well-versed in react native and ready to build and deploy quickly. But then I have to wonder. What if you just spent the same time and energy learning native tools in the first place? There is still the fact that transfer of business logic and user interfaces is trivial with react native. That is a great help. I’m sure this is even more relevant on larger teams with little to no native experience, too. In any case, it’s a bit of a suffer fest.

bromuro 5y ago

Putting a javascript layer over the native OS API never seemed a clever idea to me. I tried React Native years ago but it was like suffering twice - the usual dependencies-hell from npm by plus the inconsistencies 9f each platform. (also: react native apps never “feel” truly native. or is just me?)

steve_adams_86 5y ago
No, you’re right - they rarely feel native. I also agree about the layers of suffering. This was in line with our experience as well. The whole thing felt like cutting a log with a dull chain saw rather than using a really sharp pull saw. Sure we got there and it took a lot of labour away, but traded for another kind. It was a little faster. The end result wasn’t as nice. So… Was it worth it? I’m still not sure. It’s certainly worth it in cases where you’re well-versed in react native and ready to build and deploy quickly. But then I have to wonder. What if you just spent the same time and energy learning native tools in the first place? There is still the fact that transfer of business logic and user interfaces is trivial with react native. That is a great help. I’m sure this is even more relevant on larger teams with little to no native experience, too. In any case, it’s a bit of a suffer fest.