500 Featured Carts

Yes, we have just crossed that milestone! Have you ever wondered what kinds of carts don't make it into the featured section of the BBS? Call me crazy, but I have.

So I looked at the 500 top-starred carts (searching for tr:2015-2025) and checked which ones aren't featured. It happens there are only 30! Here they are (with links, in case you want to go check any of them):

I do somewhat understand why these carts didn't make it. We don't want the featured section flooded with Celeste mods and SMB ports.

On the other hand, it's not like we don't already have some featured Celeste mods, tools, demos, demakes and ports...

But there is hope. Walker would be on the list, but it has been featured a few weeks ago!

Anyway, what are your thoughts on this?

I consider my cat proof you don't need a lot of stars per se (28-ish at the time). I don't think the featured list is an automated system purely based on playcounts and stars. It feels more manually curated to me, but I could be wrong. Also I've seen carts get featured years after their creation.

There seems to be a bit of luck involved too.

> I consider my cat proof you don't need a lot of stars per se (28-ish at the time).

Pet the cat is awesome and truly deserves to be featured, @Doriencey !

> I don't think the featured list is an automated system purely based on playcounts and stars. It feels more manually curated to me, but I could be wrong.

I agree, and I think I can provide even more conclusive evidence of manual curation.

Have you tried STAR SPLORE? It's a launcher for all featured carts, ranked by stars. If you scroll all the way to the right, you will find featured carts that still don't have that many stars to this day.

However, I believe we can not dismiss the fact that, between the 500 featured carts and the top 500 ranked carts, the overlap is 470 carts! That's 94%... Stars seem to be a very strong factor in that manual curation.

Or is this the EFFECT of these carts being featured and not the CAUSE? Food for thought...

> Also I've seen carts get featured years after their creation.

Again, we agree. I've been keeping an eye on this recently... For example, Combat Chopper, Bubblegum Spin and Sphero are all carts released years ago that have been recently featured.

> There seems to be a bit of luck involved too.

That is something we can never dismiss, right? :)

I think the large overlap is mostly because carts that are good, are also likely to get a lot of stars in the first place.
And being featured makes a huge difference in playcounts and likes too.
I suspects it's more like the "eating icecream causes more people to drown" thing.
A corralation that happens because of a 3d factor that influences both (not counting extra stars gained by being featured).
But I don't pick the list, so we'll never truly know.
It's interesting none the less.

I have tried star splore, yes.

Thanks for trying STAR SPLORE... I hope you liked it, @Doriencey!

And thanks for sharing your views on this subject, it's been a very intresting read indeed!

I do like star splore. I discovered some carts with it I hadn't found before, because I didn't want to search every featured page.

Also thanks for petting the cat. it is NEVER satisfied.

Another example, @Doriencey: Touhou demake has just been featured. It has 58 stars right now.

In the end, I think the featured section does a great job. It gives great carts a month to shine on the PICO-8 front page.

The only problem with stars and that kind of system is the same thing that is inherent in every popularity contest especially when searches become ranked as a result...whatever people are reacting to is what people walking in blind will first see, syphoning new eyes on other projects.

The other problem is that there literally are too many games to play, and that's AFTER you filter out games that are just little tests and learning experiments out of the equation. And THEN you have to filter out stars that are people just playing the game for 3 minutes and not absolutely hating it vs. someone who never gives out stars EXCEPT for games they believe have the highest level of whatever their arbitrary standards are (Mine for instance involves whether the game has a cow for a protagonist or not...if they don't, NO STARS FOR YOU)...and so the ultimate gist is that the stars, if nothing else, promote looking and not hating what was looked at and often does favor great efforts, but sometimes, again based on my arbitrary factors, really duff stuff makes it through. Even on the featured carts, there's some stuff I find is technically impressive and very unfun to look at, listen to or play. So I wouldn't have starred those productions, but if forced to make a decision I would due to some sort of uniqueness or technical excellence on display...again, there's no referee hanging out and grabbing the player's hand if they're about to click a star and ask them, "Okay, player, before you click that star, we gotta find out if you REALLY oughtta be doing that according the rule book."

Sorry for the long winded post.

I guess the TL:DR point I'm trying to get at an possibly failing at is that I've come across 'HOLY SHIT THIS IS BADASS' stuff with no stars here. And...I blame the stars for it. :P I also think search on the forums is janky and messes up a lot of would-be discoveries. I searched the same thing 6 times and only on the 6th time did a new, extremely impressive result show up...

Also, I think a 'jukebox' of demoscene productions would be super dope...I don't know if you'd have to edit the actual presented carts to autoplay the next one, though...

> Sorry for the long winded post.

Don't mention it, @hseiken. You made a very good point.

As in the Heisenberg principle, the more you try to precisely measure something, the more uncertainty you will have on related quantities.

Ok, I know this is not quantum mechanics, so let me stop this analogy right here! :)

Anyway, as imperfect as the stars metric is, it's the one we have... No one has time to try all 10,000 carts released since 2015, so we have to use SOME metric to ATTEMPT to find the best carts without playing them all.

+1 to hseiken's thoughts. I see lots of stars on carts with impressive visuals, or demakes of popular games, or carts that are immediately fun+easy+understandable. and that's natural, I'm not blaming people for starring things the way they feel like, but the star system doesn't seem like a great way to find the sort of games that I personally want to play

Plus these forums have a push-to-front thread ordering thing and not-very-good search, which makes it really easy for stuff to get buried

Just as a data point, I have a game that is "immediately fun+easy+understandable", hosted both here and on itch.io. On itch it has 100,000+ plays(!), but here it has only 10~20 stars. I'd chalk that up to simple visuals + no marketing + buried by forum sorting + people who spread the game using the itch link. I prefer it this way, I'm just bringing it up as an interesting data point. and I'm talking about a conventionally fun game!

I think the star system + push-to-front forum sort order just isn't very good at sorting games, despite presenting itself that way. You see a "likes" number and you assume you know what that means, but it doesn't mean the same thing here because of the specific dynamics of this site

I think it'd be nice if everyone made curated recommendations of things they liked, maybe even with little blurbs attached, but that's a lot of work to ask for. but if everyone felt able to do that then you could follow people whose tastes you liked and see what they have to say

Oh actually, you can search for peoples favorites and stars (e.g. fav:pancelor) but it's a pretty hidden feature, and you have to opt-in to make those stats public (good!)

Solid arguments, @pancelor.

I also like checking people's favorites (when they are public), the curation is definitely better then simple star counts. The downside is that finding people with similar tastes can be tough...

> Just as a data point, I have a game that is "immediately fun+easy+understandable", hosted both here and on itch.io. On itch it has 100,000+ plays(!), but here it has only 10~20 stars.

What game is that? I'd like to try it.

I know that the correlation between likes and plays is not always strong. Here are a few examples of mine...

Tiny Tactics, my first cart, is the only game I made from scratch up to now. Stats: 1800+ plays and 27 stars in 2 years.

Seinfeld Simulator: Fast-iary Edition is a fork of Seinfeld Simulator with a few QoL improvements. It has gathered 7 likes and 400+ plays in 10 days.

Caroline: Cartdata Edition, my most recent cart, is a simple fork of Caroline that saves your progress. It has only 2 likes, but already got 600+ plays in a week...