pull down to refresh
0 sats \ 5 replies \ @purplereign 12 Aug 2023 \ on: Swarmstr - Find answers to your questions. Assist others in resolving theirs. nostr
top 3 off the top of my head:
- bounty / reward mechanism - at least promise rewards for best answers, beyond organic zaps
- filter by topic (or do you intend that we run one of these per tag, didn't quite get that part..)
- filter responses by social graph (ie follows/fofs) - else its just gonna get unruly (swampstr more like..)
- I'm going through answers often while testing swarmstr and I'm zapping the best answers. Perhaps I should mention that. In the future, I'd like to put aside a certain budget and automatically match zaps up to specific amount.
- yes the idea is one instance per tag {['t': 'TAG']} (can be only a server instance). You can include any instance to get the search results from there.
- mute lists are one of the thing I want to put my hands on next
reply
how about the person asking the question puts a (zap budget) and whoever provides the best answer (measured by likes) gets the Zap Budget OR another way the person who asks a question choses who gets the zap budget
Whatever the option is, users will compete between each other to provide best answers which will lead to more users and then this will lead to more people asking questions as answers are quick and/or close to an accurate answer ..... REAPEAT....
reply
Reactions are unreliable so I was thinking the person asking the questions chooses the best answer by marking it as accepted answer (by labelling, NIP-32). Perhaps the answer could be marked as accepted automatically when the user zaps it. Either way I don't think there's a good way to 'enforce' the person asking the question to always zap the best answer (am I wrong), so perhaps swarmstr could reward answers marked as accepted regardless.
reply
its also a good way to think about it ....
I was thinking more on how to generate traction (adoption) .... let me try to give some perspective from different angels
-
Questioner (person asking): the intention of any questioner does vary (looking for social interactions, truth, entertainment etc...) so we can never tell what that is, therefor letting the system decide which answer is best is probably a good way. at the same time, its also useful to have the questioner decided which one is the best answer as it can be purely subject and personal (how the questioner feels about the answer)
-
Answerer (person answering): intention is also unknown and varies but in general the answerer wants provide input .... a mechanism to reward/punish needs to be in place to encourage answers the questioners like ....
Later on, once you have a bigger pool of questioners & answerers ... you can create profiles linked on topics/subject matter experts .... such profiles within a scoring system would provide insight to other users on believability/credibility
I can go on and on but I dont want to make it very long :)
reply
Will take these into account. Thanks!
Also if you'd like to, you can raise an issue at https://github.com/ptrio42/swarmstr.com 😉
reply