4 ms·
I've recently tried this and got burnt. Spent a week of my own time doing a (what I believe to be) pretty serious improvement to both code structure and runtime
by ThalesX 5y ago
I've recently tried this and got burnt. Spent a week of my own time doing a (what I believe to be) pretty serious improvement to both code structure and runtime behavior for a very green-field project. Showed it to Product Owner "This looks great to me, but you should talk with X see what he has to say". Talk to X, total subjective dismissal.
Now I could go on the offensive and push my vision slowly but certainly, but I won't. Ultimately, I am not responsible for the product, nor for the product owner or colleague's (ignorant?) attitude, and truly I don't want to spend my time improving more than what is *wanted* in the project and my own craft.
I would have better spent the time doing Leetcode than trying to improve this and I find it hard to see why I would engage in something 'slow but certain', against odds, for something that doesn't truly benefit narccisistic me.
- majkinetor 5y agoIts not easy as that. You need to wait for good timing. Once things depend on you totally, you do it your way and use that opportunity to further pitch your story. Because before that its just a theory, and after that its battle afterhour. Its also true that this can not succeed always, but most of the time, given enough time, I will get what I want, the way I want it.