Hacker News new | ask | show | jobs
by least 480 days ago
It's not about the meta-characteristics of a user base. By allowing anyone to create issues, you are creating additional noise. Even if the signal to noise ratio were to be higher, you're still increasing total noise.

There are limits to how practical it is to allow for more and more feedback and that threshold for a solo developer is quite low. Restricting your user base by charging for your work means that there is less noise because the only people sending bug reports are paid users.

The quality of these reports are probably lower than if you had an open issue tracker, but you are substantially reducing the mental overhead and you know the people that are sending feedback are doing so with their own interests in mind.

1 comments

At any given time when I'm at a restaurant or grocery store, there's more food around and on the menu than the threshold for how much food that I as a single person can eat.
At a grocery store or restaurant, there’s no expectation that I examine every item individually, nor am I responsible for organizing the options categorically. Those are already handled for me by the establishment. My only responsibility is to navigate the choices and make a selection, and even then, there’s no 'wrong' choice, per se.

An issue tracker, on the other hand, requires active engagement from the developer. Every issue, even low quality ones, require some form of processing, be that responding, closing, or categorizing. While tools can assist a person in these tasks, the developer is ultimately still responsible for it.

I'm not saying people should only create closed-sourced paid software, but I strongly disagree with the idea that it's negatively affecting the quality of the software because there's no open issue tracker for people to post to.

It's not just github. It's every single issue tracker where users can submit feedback, some of which are almost entirely opaque, like Apple's feedback system. Look at Mozilla's issue tracker, or look at the mailing lists for linux. It's a lot of effort which simply is not worth it for a lot of people in a lot of cases.

> An issue tracker, on the other hand, requires active engagement from the developer. Every issue, even low quality ones, require some form of processing

Nope.

Yes, it does. The only way it doesn't is by not looking at the issue tracker at all. How do you figure it doesn't?
Do you have to read the entire restaurant menu before ordering?