Identifying a key variable (resources) helps but doesnt really solve the optimisation problem. Resources depend on the context and business model and that migh

Identifying a key variable (resources) helps but doesnt really solve the optimisation problem. Resources depend on the context and business model and that might not be under the control of the developer. Resources may also be variable over time, e.g when the funny money runs out. In the post and comments you see the anxiety about having to support diverse codebases. Thinking longer term you also need to address how platforms and stacks and the application requirements might evolve over a few years. You dont want to compromise your future options. This is even more critical for open source projects. Further, it is not even a binary choice as there are multiple types of "native" and "hybrid" per platform. E.g. on android you could have pure java, or some variable amount of kotlin (or even qt/c++, depending on your definition of native) and desktop is much worse. The bottom line is that its much more of a gamble than might have been in the past.