There are a lot of good things I could say about Super Mario Maker. In fact, to someone who's been making Mario levels for years already using the fan-made Lunar Magic, having a Nintendo-endorsed level editor packed with awesome features and built-in sharing capabilities is like a dream come true.
However, there is one thing that continues to frustrate me, and that is the starring system present in this game. Rather than getting into another argument on Reddit about it, I thought I'd try to explain my views as clearly as possible on here. So, without further ado...
The starring system is a sensible idea in principle. Initially you can only upload 10 levels, and as people "star" your levels, your upload limit increases. This prevents poor level designers from uploading hundreds of terrible levels, which not only clogs up Nintendo's servers, but makes it harder for players to find good levels.
First, I want to stress that, at least for me, being able to upload levels is a crucial part of the game. I enjoy making levels, but if I spend hours making a level and I can't share that level with others, then to me, that time has been wasted. This makes the upload limit extremely significant, as the game essentially ceases to be fun for me as soon as I hit that limit.
The problem with this system then, is my upload limit, and thus my enjoyment of the game, is inherently dependent on other players, who are often needlessly stingy when it comes to starring levels - perhaps because the starring system is never explained in-game, leaving the criteria for starring levels ambiguous and down to personal interpretation.
Another concern is that, as the game ages, the player base will decrease, and it will become harder and harder to accumulate stars.
I've heard a lot of arguments supporting this system, which I will try to respond to below.
As far as I'm aware, there are 3 ways that people will typically find your levels:
Obviously, if your levels don't even get played, they are not going to be starred.
Thus, many players resort to advertising in order to boost their star count, often to great effect. I've advertised several of my levels, and seen a noticeable increase in my star count. In fact, I would posit that anyone who advertises their levels enough is likely to receive some stars, even if it's just "for effort".
I have two problems with this:
Firstly, it is unfair (albeit inevitable) that people who advertise their levels should get more stars. Their levels aren't necessarily better than other people's levels. They don't deserve more stars just because they've advertised their levels.
Secondly, we shouldn't have to resort to advertising our levels online for them to get played / starred!
This is a very common argument, and it's true, to a degree. I'm sure if you make a truly exceptional level, it will naturally accrue stars. But most levels aren't exceptional - often they're just "good". And good levels don't accrue stars so easily.
I'm sure some people out there would argue this point because it's been true for them, which is understandable; I'd probably be making the same argument if I was sitting here with thousands of stars. But in my experience, this is not true for everyone. I have come across plenty of good level designers who have not had much luck with stars. Ultimately, I believe that luck can play a large part in the popularity of a level.
I'll stop ranting now, and try to suggest some actual improvements to the system:
Arguably some of these suggestions could mean that "bad" level designers are able to get stars more easily, thus undermining the whole purpose of the system. Perhaps that's true, but I think the system needs a bit of undermining. Besides, I think the rate of obtaining stars through these methods would be slow enough to discourage all but the really serious players.