3 ms·

For programming projects, I personally find that the ritual of setting up all of the infrastructure is very likely to distract me from actually building the thi

For programming projects, I personally find that the ritual of setting up all of the infrastructure is very likely to distract me from actually building the thing. I can be obsessive about doing things the right way, but my most successful projects have always been the ones where the code looks like I dragged it kicking and screaming over the finish line and the infrastructure is basically nonexistent. And sometimes I clean it up afterwards, once I am motivated by the success of the project. By infrastructure I mean things like CICD or even tests. If I have to choose between thorough testing and being able to finish the project at all, testing can wait.

xmprt 5y ago
I can relate and I think that's because with the projects that I'm most passionate about, I can't wait to set up the infrastructure and do it right. I want to do it now! But generally, if there's a project that I'm struggling to find motivation with, I'll start the long process of setting up the infrastructure, hoping that I'll get motivation in the meantime and it sometimes helps a little but often it just ends up with me constantly pushing it back.
toast0 5y ago
> If I have to choose between thorough testing and being able to finish the project at all, testing can wait. The way I look at it, you can't test anything until you know what it's supposed to do. You won't know what it's supposed to do until it's at least close to being done. No sense in testing until you get there. If you want to have a head start on testing, making sure to write your code so it's easy to test is probably fine, even if it's not what I'd choose.