I propose a new way to do contests

MakerWorld
1

I was giving some thought to how the contests are conducted on Makerworld, and though there’s nothing technically wrong with it, I have an idea, and I’m curious what the community thinks. Obviously I do not have the power to implement these ideas, but maybe if enough people like it, there would be some consideration by the staff and judges at Makerworld.

Here are some challenges that I see currently:

1 - There’s a whole lot of entries. I don’t care how many judges they have, it’s a big task to give equal and fair assessment to every entry in a contest. For example, both the school supplies and refrigerator magnet contests had over 1300 entries. Lower participation contests have maybe between 400 and 600 entries. That’s just too many to be judged fairly.

2 - There are always at least a small percentage of entries that are either significantly or completely off topic. These entries don’t belong in the contest, and only distract from the on-topic entries, making it an even more difficult job to judge than it already is.

3 - Entries that are submitted late in the contest cycle, when coupled with the sheer quantity of existing entries, cannot possibly be given the same consideration. I’m not doubting that the judges do their very best, but those who submit earlier entries will always have a greater exposure, and therefore a greater chance at winning.

My proposal:

Contests don’t run from a start date to an end date, they start on a date, and the first n entries to gain x likes then become the only entries in the contest. Once that criteria has been reached, there is a period of days in which everyone can continue to boost/like their favorites of those n entries. In that same period of time, the judges can begin printing and assessing, knowing that those are the only entries they will need to judge.

FOR EXAMPLE: The first 100 entries to reach 20 likes will move to the next phase of the contest, which runs for 10 days. During those 10 days, no other entries are considered, but everyone is welcome to boost/like their favorites. The judges then focus on only those 100 entries. The other entries that did not make the cut are no longer in the contest, and their creators can start thinking about the next contest.

Here is how my proposal addresses the problems I outlined above:

1 - Less entries to judge…WAY less entries. The judges can focus on the popular subset. 100 is still a lot, but maybe it’s 200, or maybe 50.

2 - In theory, off-topic entries won’t get the likes/boosts. Of course they could get likes/boosts unrelated to a particular contest, but it at least mitigates the issue a little.

3 - If the contest has been running for, say, 20 days, and there are still 10 more slots, even though your entry is “late”, if it’s good, it gets the likes/boosts and makes the cut. Then, it gets equally fair consideration by the judges, even though it was submitted well after the contest opened.

Ok, that’s…all for now. I would love to hear your thoughts. Bad idea? Great idea? Problems with it? Things to add?

Thanks

2 Likes
2

I didnt have time to read it all but i think for starters only 1 entry per person would trim down alot of the low hanging/ low quality spam designs. 5 entries per account is way too much!!

Most contests only allow for one design, and even if they made it so you can replace your design with a better one, it would improve the actual quality of the desings rather than overload their system with “hueforge” type junk.

Im guessing they are in it for the “quantity” aspect rather than the “quality” one.

I dont think the “like” system is a good gauge for contest winners as it can and is being exploited and doesnt give designers with low end skills(such as myself) a chance to participate… this one point I would disagree with you on…

1 Like
3

I agree in general and support your idea.

However: I have been a contender for a very long time on some other platforms.
Then I gave up and shortly after also gave up on sharing my models.
Your stated issues with contests are spot on.
And these problems were discussed in other places before, sometimes even with resulting changes.
Here is a concept that was meant to change a non long gone make site:

Every month there is a contest, sometimes two, like for adding xmas or halloween.
The criteria for models were clearly outlined and several examples with pictures were provided to make things easier to understand.
Users wanting to enter a contest had to be registered for at least one month while also having shown MEANINGFUL activity in the forum section.
And only ONE, with rare exceptions TWO models per user were allowed to enter.
If you entered with a another model or variation you had to decide which one you want to keep.
NO entries were shown on the website, only how many users already entered.
All entries had to be genuine and created for the contest - no previously listed models allowed, no copies / theft either.
Of course 4 weeks is not always enough, so more demanding contests were running for 2 or 3 months to give creators time to, well, create.

When the contest was closed for new entries and changes the judging started in two ways.
First of course through the appointed judges.
But also by opening the contest to view the entires.
Now registered users could the entires in random order and pick the favourites in several categories.
Like how unique, usable or just in line with the contest a model might be in their view.
But, and that was the best thing about it IMHO: You could select and rate as many entires as you wanted to.
But once you click on submit you had to limit it to just three model from place one to 3.
This user provided ‘ruling’ was then used by the judges to provide an additional set of winners.
The user rating so to say.
Not always with big prices always promoted the same way as the other contest winners.
Sadly the whole thing went down when the two main sponsors just pulled the plug without warning.

Our problem however is that these contests, in their current form, generate far too much money and far too many clicks.
A real change for a better and fair system would mean to not just invest more money into contests but also to miss out on money.

4

I can only imagine the daunting task of a MakerWorld team to make their selections as there are so many amazing designers and models to choose from. The idea of selecting the first to hit a certain number of likes will not achieve the outcome I believe, but overall I like that you shared your idea and I have a few thoughts as well.

Rather than the traditional method of judges reviewing every entry, a small group of designers on the platform could be invited to participate in the process. These qualified designers, who would opt out of submitting their own models, would help curate a shortlist of the best entries that align with the contest’s theme. Their task would be to identify creative, high-quality designs, particularly focusing on spotlighting talented designers who may not have a large following yet.

Once this shortlist is created, MakerWorld employees would then have the final say, selecting the winners from this pool of design, as well as overwriting this selection with others they find of value.

  • With qualified designers filtering the entries, MakerWorld staff could focus on a smaller, refined set of submissions, ensuring that each one gets the attention it deserves.
  • Encouraging Fairness: By selecting qualified designers who agree to be unbiased and focus on quality over popularity, we could help bring visibility to new or lesser-known creators. This approach could also reduce the influence of popularity on contest outcomes, which often gives an advantage to those with bigger followings.
  • The curated shortlist means that off-topic or low-effort submissions are less likely to take away from the attention given to high-quality entries. This would streamline the entire process and allow more thoughtful judging of the best work.
  • Every entry, whether submitted early or late, would have a fair shot at being included in the shortlist, as top designers would review the entire pool with fresh eyes.

This could help build a more supportive and engaging community by promoting undiscovered talent. Thanks for bringing your suggestion to the table and excited to hear what other folks think.

2 Likes
5

Just a couple of thoughts.

1 - A contest influenced by like or download totals will be manipulated 100% of the time.

2 - Telling established designers that they’re designs won’t be judged equally is a great way to alienate established designers.

3 - If a design deserves to win, it should win, regardless of whether it was made by John Doe or Fotis Mint himself.

6 Likes
6

I fell there’s many valuable points in the long replies above, but I think it it’s a bit of wishful thinking to change the contest scheme so fundamentally.
My suggestion is to offer minimal changes that would benefit the experience massively or at least considerably.

  • Limit submissions to 1 per user I think is very easily implemented and will reduce spam as well as incite more participation as there is not such a huge number of entries to compete against.
  • Reporting irrelevant submissions sounds like a great idea but it hasn’t been managed correctly, I personally have reported a lot of entries that were clearly irrelevant and stayed in the contest.
  • Featuring entries from a contest in the front page is very problematic as it gives those submissions an unfair advantage with massive exposure, if it were changed to a random selection, I think it would be more fair while keeping the invitation to join or browse.
  • Maybe the contest timeframe could change in the sense that the topic, rules and such are published way in advance and submissions are limited to only a week, then selection and prizes could be done by gauging the performance of the submissions after the contest has closed.
  • Maybe the analytics of the submissions should be hidden while on the contest page, to keep things fair.
  • Ban contest spammers from joining future contests

I think any of these points could be more or less easily implemented and would benefit the experience, without doing a complete overhaul of the system.

Also I think Bambu Staff could engage with us in the forums running discussion threads or surveys for more acuracy on the changes.

1 Like
7

Defining the system based on user-supplied Likes is ripe for trouble.

If you think curating 1000 user-entry models in a category via a bambu human is a daunting task to vet - you are asking for a literal tire fire to require that same bambu human to manage something purely open to WAN user vote manipulation.

Ask Facebook how their process of combating automation in astroturfing is going. (I’d say ask Twitter, but they waved the white flag like the cowards they are long ago - and see how that has affected quality)

It’s human nature to try to exploit a system that has a real reward, especially if its material reward. Make your systems resist that.

I think you will only get good results via community management, which involves visible rulesets and very timely moderation, so that authentic users see that the community is actively enforcing its rules against the astroturf. My criticism of the current setup is that it doesn’t seem to be someone’s job to curate the content, because it only happens at the end of the contest, if ever at all.

8

Do you think Reddit has a better system of up / down voting by the community than the other platforms?
Genuine question, because I do believe it does, but I’m not very knowledgeable on the topic or if my perception is actually true.

9

Thanks for sharing your thought, but I will say that an upvoting and downvoting system like Reddit will have negative consequences. Imagine you work tirelessly to make a great model, you post it to MW, and then you get a snowball of downvotes. That would be enough to make some people feel like quitting. The great thing about likes is that even if only a few people like your model, you still know that it’s liked, and maybe you can make changes and try to improve next time.

We should ensure the system encourages positive behaviors and fosters a sense of community. Cheers,

1 Like
10

I agree with you but feel we need a way to punish bad actors. Especially people gaming the system. And thought in my ignorance, Reddit might shed some light.

11

Having MW follow their own rules every time they receive a report would go a long way to helping.

3 Likes
12

Thanks for responding.

I probably didn’t explain myself very well, but I’m not sure where you concluded that your number 2 or 3 would be issues.

Every person would have a chance to participate in the contest. Nobody would be excluded. The idea is that the final pool of, let’s say 100, entries would be judged equally, regardless of the number of likes/boosts. The likes and boosts are only to determine if they make it into that pool, not whether or not they win the contest.

As for manipulating because it’s influences by likes, I somewhat agree. There will always be a percentage of people who game the system, or are just more popular. The idea, however, is that the pool is big enough (I use the example of 100, but maybe it’s 200) that even if 10% of those that made the cut did so dishonestly, they have no more of an advantage for winning over the remaining 90% who came by it honestly.

13

Thanks to all who have responded so far.
I will say a few things, as general responses:

1 - Currently, if an entry in a contest is good, it will get a lot of likes. Then, if the judges choose that entry to win, we might say their win was influences by likes, but in reality, it was just a good entry. With my proposal, that wouldn’t change. Why would it?

2 - The idea with my proposal is that, once the pool of entries is locked down, due to getting x number of likes (say 20), those in the pool would get judged equally. So, even if an entry remains at 20 likes, they would get just as fair of an assessment as somebody with 300, as long as they’re both in the pool of entries.

3 - I feel that the ability to upvote or downvote (like on Reddit) is fraught with problems, and gaming. It’s a nice idea, but I think it would be a mistake to allow downvotes.

4 - By implementing my proposal, there is much less likelihood that likes would be the primary consideration for the judges, because they have a limited pool of entries. Currently, let’s say there’s 1200 entries - common sense tells me that they HAVE to have SOME way of filtering down to a manageable number, so why not use likes? I think it’s unreasonable to think that they are looking at and evaluating every one of those entries.

14

I do not know if you have noticed that a late entry (7+ days after the contest start) will earn very few downloads or likes no matter the quality of the entry.

Too many overlapping contests act to siphon visitors away from the previous ones.

2 Likes
15

+1 for “Limit submissions to 1 per user”

1 Like
16

There’s a possibility to alleviate that somewhat using combined suggestions here regarding submission limits. There can be at most 5 concurent contests due to contest schedule (weekly). How about keeping the limit of 5 models sent to contests but make it not per contest but per user? So if one would like to have entry in every contest they’d be limited to 1 submission per contest, but if they feel like they have good ideas for single contest then they can invest their time into one single contest.

That could also limit the usage of contest as a promotion of irrelevant models since in worst case scenario a person not caring about rules can get 25 models promoted via contest but after my proposed change it’d be 5 at best.

Also I feel like the reports about contest entry being irrelevant to contest might not lead to removal from contest immediately, but after contest ends, since bambu seems to act rather fast on other reports (at least from my experience).

How about on top of point deduction for irrelevant model entry that would lead to also contest entry limit deduction? Eg: first time the person enters irrelevant model that’s a 100 point deduction and limit decreased by 1 entry (so at most 4 entries in total to all contests), 2nd time 250 point deduction and limit decreased by 2 (so at most 2 entries in all contests) and 3rd time would be 500 point deduction and ban from entering contests.

Also I agree with @user_3026326371 suggestion to allow only established accounts to join contests, so say minimum 1 month in MW is good starting point, but I’d also add a requirement for users to have at least one published model outside of contests. That could be further narrowed down with things like requirement to complete the point challenges: so 20 points for completing personal profile, 50 points for uploading a model and 15 points for uploading print profile… Maybe going as far as requiring reaching Milestone 1 in “High Quality Model” and “High Quality Print Profile”

1 Like
17

I think the main issue with contests is off-theme entries.

Perhaps a prominent notice of the penalty at the Join Contest page can help? Some may not know and are trying their luck to get eyeballs with clearly off-theme entries.

About the point of too many entries (on-theme ones), I fail to see the logic to limit participation. We all got to start somewhere, even if it’s a simple model.

1 Like
18

I like this idea, determined gamer would still get through but my guess is that fewer than how it is done right now.

The increasing deduction in points and decreasing submissions level sounds like a good way to keep the gamers honest.

19

About the point of too many entries (on-theme ones), I fail to see the logic to limit participation. We all got to start somewhere, even if it’s a simple model.

There is no limiting of participation. Anybody can enter. If they have a good entry, they’ll get likes/boosts. If there are still slots remaining in the 100 (or 200, etc), they get in.

I completely agree that even simple models should be able to be entered. I think that the misconception is that every entry is getting equal time in front of the judges. If there’s 1200 entries, they can’t possibly give equal time to each one. Having the perception of “participation” and actually getting judged are two different things.

20

Yep I think reducing the number of entries per user from 5 to 1 (or 2 max) would greatly help IMHO.
At least to promote quality over quantity.

2 Likes