Critique of the Unity Feedback System

I feel that there are some problems with how the Unity feedback system (the page where you can vote on community suggestions) currently works. I feel it could be made more effective than it currently is at prioritizing changes.

Consider these two observations about the current feedback system.

• There is no way to estimate the cost of implementing a proposed feature.
• You have a finite number of votes.

Why is this a problem? It inevitably leads people to voting on huge features, and the simpler, more practical (and perhaps more useful) ideas are more likely to be ignored.

Suppose you wanted to propose a "small" change. Perhaps something like giving a more accurate breakdown Physics.Simulate in the console. I understand that due to some technical reason this might not exactly be a trivial change - but by comparison to a feature like "Voxel-Based Terrains", or "Physically Destructible Meshes", it is by far a much simpler task.

Having a more accurate readout on physics performance, I'm sure we could all agree would be an improvement. Perhaps you yourself have encountered ambiguous "Physics" performance issues that left you scratching your head.

But, why would you spend your only 10 votes on it?

Why wouldn't you want to get "the most bang for your buck" and spend your only 10 votes on something that would have the most impact? To spend them all on smaller change would be a waste.

Solution

As others have suggested, a "Like" system has proven to be a reliable way to gauge what people are interested in. This is because, even though there is unlimited supply, it takes some conscious effort to find something you're interested in and deposit it.
On the other hand, imagine only being able to have 10 likes on Facebook. You probably wouldn't spend them on that funny cat picture someone sent you, even though you actually do like it.

In addition to that, however, I feel it would be useful to have some way of estimating how expensive a feature would be. Perhaps when you post an idea, you have to describe the scale of the idea, from minor improvement, to features that would be considered a big selling point. Perhaps the sorting criteria for priority could be something like, the demand for this feature, divided by the cost of implementing it.

This way, small improvements would float to the top, because they're easier to implement, and people wouldn't be afraid to vote on them.

 

That's not at all comparable. You aren't voting on a potentially important feature being developed, you're voting on an entry someone has made on their account. One way to solve this would be to simply give everyone an unlimited number of votes but allow them to only vote on a single entry once.

Not that the feedback system is even worthwhile when someone can create multiple accounts to drive up the votes. It is possible this has been happening with the Linux editor entry.