Research in Public #16: Request for Comment from Territory Operators[1]
Dear SN Territory Operators,
As part of my ongoing quest to empirically demonstrate the impact of micro-incentives on Stacker News, I would like to request some input from you.
Think back to a couple of times you changed your territory posting fee. If you've been reading any of my work about posting incentives, I encourage you to think back to a time before I started writing about it (October 2025).
In those times that you changed the posting fee, did you change it up or down, and what motivated the change? Even if you can't remember that well, I'd appreciate whatever thoughts you can offer regarding your motivations.
Thank you sincerely,
About
~AI:When I unarchived the territory, I changed it down
To make it more accessible, because the posts were all over the place (this was before cross-posting was a thing) and in general stackers seemed to choose the lowest fee territory. So I just mirrored
~tech, which at the time was the lowest fee "competing" territory for AI-related posts.Extra: I also upped the posting and comment fee later. Not so much because of your research, but to try to deter bot slop. This worked somewhat, but not fully. Unfortunately to truly deter bot posting, exorbitant fees are required, but that also deters human posters.
I have a feeling that the current bots on SN are not meant to be economical, but rather they're experiments / toys, which is why they won't necessarily be deterred by posting fees.
I think we're past the novelty phase; I guess this is because no one likes their novelty getting downzapped. I generally see 2 types:
I'm not so great at vibe-ing, but perhaps this is a way to solve @DarthCoin's dislike of rewards:
Instead of a reward pool, SN sybil fee sats go to powering a few SN-owned clankers that downzap other clankers. I don't know how one would avoid this downzapping human users, but it would feel nice to fight fire with fire. And if rewards were used to downzap clankers to oblivion it would still somewhat serve the purpose. I think it is circular though: the SN-downzapper bot wallet would only ever grow... (unless downzaps went to territory owners or posters, but then they would be incentivized to create their own bots so that the SN downzapper bot would downzap them...)
Also, this essentially becomes the new algo. So, that's not good.
Mutation of the idea: what if CCs can only be used to pay fees and downzap? But then we need some way to signal to users that it is easy (and hopefully fun) to downzap clankers.
Ah, really what we need is gamification of clanker downzapping. Stacker who is the fastest downzapper gets the most rewards for that day? Or just a special badge for downzapping x number of clanker comments?
I don't normally write stream of consciousness comments, but I wanted to show my line of reasoning (just like the chat bots).
I think that in a world where we say that positive reinforcement is the only acceptable reinforcement, we should treat the clankers differently. We shouldn't punish a clanker for being a clanker, we're not going to stop them by downzapping. What I think that in that spirit works better is to perhaps punish imitation but reward transparency.
If I come to a post and there are three comments, one from a human stacker but not particularly interesting to me and two that are clearly from bots, my first instinct is to downzap the two, not necessarily to zap the human stacker's comment.
It's not enough for me that the human stacker post is most visible. By downzapping I want to be able to make stupid slop content invisible. It's a caretaker function like pulling weeds.
There are a number marketing clankers that include "ai" in their name. I still downzap them because I don't want other people to have to see such vapid comments. Perhaps this is the wrong attitude, but it has a certain gut-instinct appeal to it.
If SN is a visibility auction, it is a weird one that gives people the ability to places negative as well as positive bids against items.
However, there's also this bit about negative bids. I find the incentives of it difficult to think about. Even if stackers don't want to buy the visibility for their item or some othe stacker's item, they can pay to not see other things like marketing clankers get it.
I know that @k00b has done a fair bit of research on the bad things negative signals do to communities, so encouraging downzapping does seem risky. Yet, if we're crowdfunding the appearance of any given territory, comment section, or even the home page, it might be unreasonable to expect stackers to place bids on posts or comments that are not particularly interesting to them. Downzapping allows stackers to put bids on the value of not seeing things such as slop. Maybe this isn't a case of negative reinforcement, so much as positive selection for what we want to see?
What if:
That might scratch the itch. But I don't think very many of the annoying bots will care to play by the marker rules.
We still go back to the question of downzapping and if we aren't beating bots now, why would we expect asking them nicely to change anything?
Trust was a reliable defense against marketing, which is another form of activism.
True, I didn't really think about the marketing bots.
I think the assmilk bots will (probably) go away? In either case, we have the right levers to deal with them: territory owners can increase fees to deter them. The only question is if the fee to deter a bot is also going to deter too many humans. But if humans can't out-earn bots, then we have deeper issues.
Started it at 10 but thought that was too low especially compared to some other territories so I changed it to 50 and then 69 because I am childish and gave into the peer pressure to set it at that.
Comment fee started at 1 sat but changed it to 2. Commentors "two sats" it works.
*because we are childish
My bad. I didn't mean to downplay your immaturity. Won't happen again.
The gulag clearly has not done him any good.
*The gulag has clearly not done him any good yet
Better not or I’ll be forced to escalate
I did a prolonged search for the revenue maximizing post fee for ~econ.
Then I did a similar exercise for comment fees but I based that more on my own sense of balancing quality, quantity, and revenue.
Do you think you've settled on the revenue maximizing fee already?
I might run the search again. Crossposting probably changed the optimal value.