To cut back on spam, duplicates, low-value markets.
1,000
3.00Many low-value and poorly structured markets are the result of new users experimenting and figuring things out. More beneficial than raising the barrier to entry would be lowering the cost of being wrong, at least initially. Make it easier to learn from mistakes. Make changes easier. Or put a STUDENT DRIVER sticker on everyone's first three markets.
Mostly, no one creates a duplicate market on purpose. If only there were like a very smart searchy thing that could pop up to say "Hey! Did you mean to make a market similar to this one?" Like Clippy, except useful.
@ClubmasterTransparent I second this, some forums have an inbuilt search engine that shows you similar questions when typing your own. Maybe this could be a future feature.
The downsides of increasing the costs could easily be offset by upping the trader bonus. For example, if the trading bonus was increased to 10 for the first 10 traders, it would offset the cost increase. The creator will make just as much if it is successful, but punishes the low quality markets that don’t gain interest.
Here's a possibility I don't think I've ever seen raised: drop the requirement that markets have fixed creation cost. Start off with a base cost and base trader bonus for all markets equal to what we have now. For each market a user makes in a given day, increase the cost and trader bonuses of the next one by some knowable-in-advance amount: maybe arithmetic (1x, 2x, 3x, 4x..), maybe geometric (1x, 1.5x, 2.25x, 3.8x...), maybe even logistic (1x, 2x, 3.9x, 7.7x... ...20x); whichever you pick, it always resets at the end of the day. Maybe play around with new-trader subsidies alongside this? This would help prevent spam and make people think twice about duplicate or near-duplicate markets while still avoiding penalizing new users too harshly for making 2 or 3 markets in a day - after all, if they're good markets, they'll still make the creation cost back.
Currently makret creation is highly subsidized. The entire 50 mana you spend goes into subsidies, and then for every bettor Manifold mints 25 mana (I think? 20 subsidy + 5 trader bonus). Steps need to be taken to fight mana inflation if Manifold sees Mana as something that should be tied to real value rather than exclusively as play money. It makes a lot of sense to do that to market creation, which is currently usually positive-sum even if you don't consider subsidies. I think the fact that this poll has more than a third voting yes is a good argument for yes, even though it is not the majority. People don't tend to vote for things that would be detrimental for them, so if this was even a neutral change in term of Manifold's long-term health there would be many more against.
I think an increase an price can be counteracted by discounts for new users, like those that already exist as quests and the 3 free questions. The weekly market making quest can be increased to 50 mana.
Seems like this will mostly stop newer users from creating markets. I don't think it would affect people with decent mana pools very much.
Are there other options you are considering to achieve the same goal?
If the intent is to foster market category diversity, I wonder if the solution is to price market creation variably, dependent on category?
I suppose it creates a different sort of moderation challenge (category/name mismatch), or creates a perverse incentive to create markets that have multi-leg bets (for example, outcome of the 2024 election… contingent on an NFL game), but it might cut down on low quality AI/politics markets if you price those market categories higher.
NO Voters - you oppose raising the cost - so, would you recommend lowering the cost? Or you're stating that magically, out of all numbers, 50m is just the perfect one?
My point: cost to create a market is one of the inputs to an equation which has both costs & benefits for manifold:
Too many markets
bad for searchers & high quality market creators, since it's hard to find good markets (unless search team is on point), and good markets get overwhelmed by bad. (classic deletionism viewpoint)
Too costly to create markets
bad for creators, frustrating, losing possibly good creators
Too much friction to create markets
Similarly, proposals to "manage" market creation can create (newbie LLM assistants to creation, required tutorials before creation, time limits, community approvals for new users markets, market creation rate limits etc) => annoying for creators.
Make markets cost a variable amount
this improves judging efficiency & related things, but makes mm much harder to understand for new creators and traders (since markets now are of even more types, judging rules, etc.)
So overall: Manifold has to make a choice which balances all this. Is there any way to gauge these as of now? Is one favored too much, at the expense of the other? I don't think you can just say "the number should go up" or down without explaining why.
It doesn't have to be with numbers, but I'd like to know how you all view the tradeoffs of ALL of the costs and benefits of raising or lowering it.
I haven't voted either way cause I don't know what the costs and benefits of change would be, or what manifold is thinking about in the future
NO Voters - you oppose raising the cost - so, would you recommend lowering the cost? Or you're stating that magically, out of all numbers, 50m is just the perfect one?
Hi, NO voter here. Representing myself and my own opinions.
I don’t think 50M is the magic number, but I’m semi-change averse, and I think what we have today is working - enough so that absent data on why we’d want to double, I didn’t opt for making so a big change.
I’m new enough to remember when spending 50M to make a binary was a big deal, that was like ¼ of my wallet sometimes. I agree that the site is getting spammy now, and I’m not entirely sure if doubling the cost is the most effect fix for the problem.
Reading back the arguments in the thread, I’m closer to a YES now, but still probably a NO on this going to 100M.
@mattyb The 50 mana is NOT A COST. It's an investment. You get paid out from the pool equivalent to how poorly the market closed vs resolving.
If the 'cost' to create a market goes up to 100, I expect the majority of it to go into the liquidity pool, just like it does today.
@Eliza sure but that’s mana that’s locked away for a long time for some markets. For a new and active market creator, raising this to 100M is going to slow their upward mobility. Look at this dumb market I made thinking it would do well a few months ago. Back then, that was a decent amount of mana that I lost and had to make back some other way.
If you double the barrier of entry for new people, you’re making it harder for people to use market creation as a successful strategy for wealth accrual.
@mattyb Fully aware, I have spent my entire balance on creating questions for my entire career, to the point of "not having any mana to bet with" -- see my other suggestion that I think anyone should be able to make a question for less, but I still think that allowing people to make questions with more is a good idea.
@Ernie No voter. IMO, the barrier to entry is sufficient now. I don't see too much spam. Additionally, the M50 is a factor in my not making many markets. (Though having to resolve them and write the text has more weight.)
But not everyone feels that way. Maybe we can add a feature where people can choose what they pay (over some base # like M50) to create the market, but users can filter by market creation price to only see markets that cost more than X. This could give evidence relevant to the best creation cost.
Manifold might have an income stream from users paying real money to get into a specific set of filters.
@Ernie I'd like to see a hypothetical breakdown of effects. Is there some way to quantify any of it? D the founders know something about cost/ benefit here? I don't want to vote without knowing reasons
@Ernie Hello! NO voter here. I think that the cost to create a market should be 25 marbles exactly, so that a new user with a maxed streak could, if they chose, create a new market each day. I don't think that too many markets is a problem, but I remember having so little mana at the start that I was leery of making markets at all, and that scarcity mindset has somewhat stuck with me for better or worse.
In my eyes, this incentives people to duplicate markets even more and reduces the number of fun and unique markets (which is why so many people are on Manifold).
Especially new users won't really be able to create markets, if they start out with 1000 Mana. The people who just want to try out new questions and formats would hesitate even more.
On the other hand, the boring and easy questions will be duplicated like before because they are safe cash cows.
If some question appear to spammy, find a way to improve the homescreen and the questions which are recommended under other questions. (Kind of "If you like this question, you'll also like this one.) I'm rarely using the homescreen (expect to collect bonusses), because there are nearly no questions for me.
Why 100 vs 1000 or 10000?
Is there a weighting of costs vs expected benefits from the policy change?
If he price went up, would there still be rules against duplicates? Or would they still be unsubsidized?
Does judging cost count here? Creating expensive to adjudicate markets, or ones by users likely to quit and add a load to moderation teams, and damage the reputation of manifold with users?
@Ernie classic UGC business question:
"Can we stop users from making so much UGC? It's expensive to handle it all"
@Ernie Long term do you want users to feel like they can easily create markets, or not?
compare:
Youtube: extremely easy, users never think about costs
Roblox: extremely easy (as easy as we could make it, anyway), no costs imposed
Wikipedia; easy, but you get smashed 10 minutes later => nearly nobody adds things
Where would manifold like to end up?
Some cost seems reasonable but if in meetings about market creation rates, the only input is "moderators saying how bad lots of markets are", then the cost will keep going up.
I think it would be great to also have top-level views & principles protecting the idea that mm should be a place where creators can create markets with as much freedom as possible. Is there some guide for the benefit of market creation, or someone focusing on making sure creators have a good setup and that manifold has enough of them? They're a cost, but manifold needs market creations too. Does manifold think there are too many markets now? Does manifold benefit from new creators creating a few bad markets, before they figure it out? If manifold prevents those new creators from doing those first experiments, will that creator instead give up, or will they persevere? Will the overall conversion rate of "new user => newbie creator => good creator" go down? how high does that rate have to be for manifold to sustain itself?
I know lots of people in manifold believe creator experience & freedom is important, but is it written or confirmed anywhere? If moderation comes in and says "too many bad markets are making us spend too much time" will there be a way to fight back? or long term, will content creation get more and more restricted, exactly as it has done on wikipedia?
If in 2 years VCs come in and say "You're spending too much money on moderation of markets, we have to cut costs right now, easiest way is to raise market creation cost to 500m and only allow 1/day for new users", would you be happy? or would this be a failure? I'm worried that if a sustainable automatic moderation system doesn't get created, then manifold might end up in that position