First, on behalf of @imaqtpie, @Seraph089 and myself, thank you all for choosing us to help run the community. We’re all really excited about the possibilities of both this instance and of The Agora community. We’re look forward to working with everyone to make this a great community. Feel free to reach out with any concerns or comments!
Ok, on to the announcement:
Today, I’m excited to share with you some pivotal updates set to streamline our interaction and decision-making processes within The Agora.
The first of these updates is about enhancing transparency. We have established a new and convenient way to track the outcomes of our community decisions. Simply visit this link: https://rentry.co/the_agora. This site will serve as the hub for all voting results, updated at the conclusion of each vote.
Next, let’s discuss the changes regarding the use of our existing [Discussion] and [Vote] tags. To foster clarity and improved interaction, all new posts should now carry the [Discussion] tag.
Regarding the [Vote] tags, we’re introducing a more structured approach here. Going forward, the [Vote]s will be initiated by the moderation team based on the week’s [Discussion] posts and will be posted each Friday and run to the following Friday. This gives ample time for each of us to participate in the decision-making process. Once a vote concludes, the corresponding thread will be locked and the results promptly updated on our new voting results webpage.
For [Vote] posts, your vote should only be cast as a top-level comment. To streamline the process, we ask that you refrain from responding to other votes in the same thread or making non-voting comments. Each [Vote] post will contain details on how to format your comments, and our moderation team will be available to ensure all comments are formatted correctly before the final vote count is tallied.
This is by no means the final process and we’re depending on your feedback and discussion to keep improving things going forward.
We understand the concerns about vote manipulation and the discussions around alternate voting methods (like ranked choice). Use this thread to discuss the changes and any concerns or suggestions that you have.
As of now, the tentative plan is to run with this for the first week, see how many issues exist that require voting, generate the vote threads, complete a round of votes and then iterate on the process once we can all see what works and what doesn’t work so well.
Suggestion for the vote:
From what we’ve seen pretty much everyone was in favor of restricting the vote to this instance’s users. To prevent flooding from people from other instances singing up here just to vote on issues that might indirectly affect them, maybe only votes by people who were already members of the community at the time the vote or discussion was opened should see their vote count?
For once we agree on something
This is a great step forward. I’m unsure how we should handle voting but I do want to express my support for the week-long opportunity to vote on important matters, and the recording of votes in a transparent and publicly-auditable fashion. I do think there should be a practical maximum to the number of topics that come to a vote weekly, I think more than 2-3 would quickly become unmanageable for everyone.
I do think there should be a practical maximum to the number of topics that come to a vote weekly, I think more than 2-3 would quickly become unmanageable for everyone.
Yeah, I agree with that.
We just couldn’t come up with a number that didn’t feel arbitrary, so the idea is that we’ll run a week to see how many topics come up and try to pick the ones who’s discussion threads seem to have run their course.
I don’t think it’ll be a big issue right now since the community is pretty small but I can see that things could quickly get unmanageable. This is definitely an area that I think we need to iterate on.
I would like to use star.vote, if at all possible, but I understand it would be harder to control who is eligible to vote on an external site.
It’ll be interesting to see how the current plan works and what refinements are made.
@Difficult_Bit_1339@sh.itjust.works - I read your post in the “beehaw user’s thread” and had some clarifying questions for the tally:
- Does your tally exclude any votes that have already been removed by a moderator?
- Why have some non-local votes been removed, while others remain?
- Does non-local solely indicate accounts not on this instance, or does it include sock puppets of the same? What is the full list of criteria used to determine this?
- Are there any other criteria you are using to exclude votes?
I’m specifically asking because I’ve been trying to follow along and audit the current vote, but my tally isn’t copacetic with yours. I assume I’m working with imperfect information, so I think it would be worth adding this level of clarity on voting qualifications to the Agora sidebar as well. A true democracy should be well codified; It would be a shame for the disparity to be down to me missing one buried comment somewhere.
Thanks for the work you’ve been doing here.
Does your tally exclude any votes that have already been removed by a moderator?
Why have some non-local votes been removed, while others remain?
We were removing posts of people that were not instance members. However, since users cannot see what the post said, having a [Vote] post full of ‘removed by moderator’ can look like unfair vote manipulation on the moderator’s part and, in addition, it was a lot of work to manually remove each post (the mod tools are… primitive). Instead of removing them we’re just excluding them from the count. Each user’s account has a ‘local’ flag that is checked by a script and if they’re not a local user it doesn’t add their vote to the tally.
Does non-local solely indicate accounts not on this instance, or does it include sock puppets of the same? What is the full list of criteria used to determine this?
Are there any other criteria you are using to exclude votes?
We’re having to check for sock puppet accounts manually at the moment. We have not excluded any local votes. There are not a HUGE amount of suspicious accounts (maybe 3-4 accounts who were created two weeks ago and have no comment or post history except for one vote) so we don’t expect it to affect the final result.
We’ll probably need to have a discussion in the future as to what objective criteria we should use to determine if a person is an active community member in order to prevent any perception of bias.
I’m specifically asking because I’ve been trying to follow along and audit the current vote, but my tally isn’t copacetic with yours. I assume I’m working with imperfect information, so I think it would be worth adding this level of clarity on voting qualifications to the Agora sidebar as well. A true democracy should be well codified; It would be a shame for the disparity to be down to me missing one buried comment somewhere.
The final vote count will likely be done by hand with the script acting as an audit. If the script and the hand count match then we’ll have a bit more confidence in it going forward. It’s a very manual process currently.
I think that after this vote is done, we should open a new discussion thread on voting so folks can give their feedback on the process.
Thank you for the comprehensive reply. It means a lot to me as someone invested in the success of the instance.
I would like to suggest that we consider open sourcing the script, since it is a practical encoding of the instance’s voting bylaws. I’m interested from a purely technical perspective, but I am also of the opinion that open source software is a tool of democratization.
I would like to suggest that we consider open sourcing the script, since it is a practical encoding of the instance’s voting bylaws. I’m interested from a purely technical perspective, but I am also of the opinion that open source software is a tool of democratization.
This is imminent. TheDude just wants to track down a few issues and then polish it up before it’s subject to public scrutiny.
I think this is an excellent starting point. Of course, anything and everything can and should be improved but as a baseline this should work pretty well.
I was a pretty big advocate for mandatory discussion before any vote, as the OP can dominate the narrative as people start voting before people have a chance to mull it over and figure out the ramifications. I’m really glad that a form of that has been decided on, and hope that any future iteration of the rules keeps that as a core idea.
Coming over to this thread as I watch the votes roll in on defederation for exploding heads.
I am noticing a few accounts voting that registered today. That does seem problematic. I know we had a discussion post 12ish days ago about some criteria to make votes count and I continue to think thats a good idea.
I’ve only clicked 10ish names in the thread, 8ish ayes and 2ish nays (theres just not many nays yet) - but of those I clicked 1 was freshly registered ( an aye vote) and 1 was an account that had only posted in exploding heads before their vote (nay vote)
edit: just wanting to be clear I don’t think any of this info invalidates the vote today, I just think we should use it as a good test case to tweak going forward. If nothing else we should notice if the <5 dayish old accounts differ wildly from older accounts in opinion
Yeah, probably a rule that says your account has to be at least a week old and maybe some other criteria to quantify what an ‘active user’ is.
I believe the account should be a week old, and with a certain minimum number of comments. I’ve also seen the suggestion that the account needs to be active in more than just the Agora, and I agree with it. We want to ensure people aren’t making an account just to vote, and by having a minimum number of comments, we make it harder for one person to create multiple accounts.
And to combat nonsense or chatgpt comments made to get past the comment limit, we’ll all have to keep our eyes open and report them.
It takes a lot of wargaming to see what would work.
If you make it a minimum number of posts then we’ll get users reposting memes or content from other social media websites. If it’s a certain number of comments then a person could use LLM generated comments or, like on Reddit, just steal comments from other threads.
Having a classification system that could look at all user activity and then sorting them into buckets (normal user, possible alt, spam, bot, etc) based on training data would be nice but that’s a very complex project to embark on.
On the other hand just leaving it up to the moderation team to decide what is or isn’t an active user isn’t sustainable either. Who’s to know that we’re not disqualifying people who’re voting against our position or some other shady activities?
It’s a tough problem and definitely one to keep talking about.
I think it’ll come down to people keeping an eye out for shady things like you described. Just like a lock only keeps the honest out, people will find ways to break any rules we put in place.
definitely a tough needle to thread - we’re all newbies mostly. Maybe there’s an intelligent and fair way to weight votes based on active participation and time as a user. Idk though, because there’s always perverse incentives and I definitely don’t want folks commenting garbage just for the sake of pumping the weight of their vote. There’s always a way to game the system, that when rules to prevent such attempts are put in place, make it worse for good faith users.
yeah anyone who is dedicated to gaming the system will do it. But we could make a 48 hour old account rule or something to prevent basic last minute brigading
Since this post was linked from another thread, @Difficult_Bit_1339, you’ve unfairly characterized @socialjusticewizard as a beehaw.org user coming here “trying to stir up shit” as you’ve phrased it. Their sh.itjust.works account predates their beehaw.org account by two weeks. This post, the one we’re commenting on right now, is NOT clearly labeled as being the rules for vote posts. It’s just named “changes.” You should consider putting the rules for vote posts in the sidebar and in the vote posts themselves.
If you want to label me as a beehaw.org user coming here “trying to stir up shit,” too, so be it. At least my first account was from beehaw.org, and I came here looking to see sh.itjust.works and lemmy.world posts. At this point, I’m entirely done with this instance based on the overall handling of this situation and will be using my @Quill7513@slrpnk.net account for the purposes this account was originally meant to serve. Here is some advice I have, as a sort of exit interview.
- Be nicer. Come on. The way you moderate this community will influence the way this entire instance operates. Your rudeness and dismissiveness sets the tone for the entire instance and how people will perceive users with
@sh.itjust.worksas part of their identity - Define an executive process for defederation, just as you already have an executive process for moderation. Defederation is part of moderation and 1 month is not a fast turn around for this sort of situation
- Increase the transparency of the audit scripts you’re using to tally votes by linking a link to a git repo containing the script. I think it’s fair to say that your automated script for what the vote talley is and what someone reading through the vote sees as being the vote results are quite different
I referred to them as a Beehaw user because their posts and profile both indicate that they’ve moved to Beehaw as their primary instance.
Be nicer. Come on. The way you moderate this community will influence the way this entire instance operates. Your rudeness and dismissiveness sets the tone for the entire instance and how people will perceive users with @sh.itjust.works as part of their identity
That’s fair.
However, I don’t believe I was being rude or dismissive.
- The user’s question, asked in the title, was answered in the stickied announcement post.
- They changed their display name to ‘Leave This Instance’
- They announced in their profile that they were leaving the instance and accused the admin team of not acting in good faith.
- The body of the post is simply discussing the voting system. We already have a post announcing and discussing changes to the new vote system… creating multiple posts about the same topic only fracture the discussion.
The fact that the user was leaving the instance, encouraging others to leave and also accusing the admins of acting in bad faith doesn’t really mesh with the idea that they’re just a user attempting to have a good faith discussion about the voting system. Instead it reads like a concern troll by a user who wanted to get one last jab in before they left.
The thread is up and available for anyone to read, it’s just locked so that conversation about voting changes will be placed into the correct discussion thread.
Define an executive process for defederation, just as you already have an executive process for moderation. Defederation is part of moderation and 1 month is not a fast turn around for this sort of situation
I think a discussion about defining a policy for de-federation is a great idea and is probably more sustainable than having to have votes on every individual instance. In this case, the issue was fairly contested and had already been submitted for a vote so we used it as the topic of our first week’s voting. That doesn’t mean that all de-federations require a vote. There was another instance that was allowing content that, under Canadian law, is consider child pornography and it was de-federated immediately.
Increase the transparency of the audit scripts you’re using to tally votes by linking a link to a git repo containing the script. I think it’s fair to say that your automated script for what the vote talley is and what someone reading through the vote sees as being the vote results are quite different
It isn’t my script to share and the final count will be done by hand with the script acting as a check. If it is accurate to the hand count then we may depend on it going forward but currently it’s just a way to check the progress.
We didn’t want to have the thread full of comments that only say ‘Remove by moderators’ as that could be construed as vote manipulation since nobody can see the content of the removed messages so we decided to leave the comments and instead wrote a script to check to see if a user is flagged as ‘local’ or not.
deleted by creator
What are you even talking about?
- Be nicer. Come on. The way you moderate this community will influence the way this entire instance operates. Your rudeness and dismissiveness sets the tone for the entire instance and how people will perceive users with
I love the new approach, seems well thought out. I like the website, but I feel like I’m gonna forget to check it haha
I have an adjustment to voting that I’d like to propose, considering how the vote to defederate EH is going.
After 3 days, if the vote tallies show a supermajority of votes for aye or nay, and the vote has passed a certain number of votes cast, then the vote is called early.
We would want to make sure the supermajority is high enough to be clear, say 80% or 90%. And we must absolutely be sure that enough votes have been cast to represent the instance at large. Perhaps 40-50% of the total users? I’m no statistician, so these numbers will likely need to be adjusted.
I think this will help us, especially early on like this, where there may be lots of policies and procedures for us all to vote on.
And this should only be implemented for future votes, as I think it’s bad form to change things mid-vote.
The concern that led us to picking a week as the voting period was specifically to ensure that users who could only be active on certain days wouldn’t be left out of the vote. We wouldn’t want someone to be unable to take part simply because they worked on the weekend and couldn’t find the time to participate.
Ideally, I think that we’d want to have things so that we vote on policies and then the policies are implemented by the admin staff. That way there would never be some issue that would be so pressing that it couldn’t wait a week.
Ideally, I think that we’d want to have things so that we vote on policies and then the policies are implemented by the admin staff.
I’ve been having the same thought, that we can create policies, or “laws” that the admins can then use to decide things like defederation or removing communities.
Seeing how the discussions have gone about defederation, I imagine the Discussion thread deciding what the triggers would be will get a bit spicy.
All sounds good to me! :3
I’m a little late to the party with this. But I like the vibe of everyone I’ve interacted with from here so far, and @TheDude@sh.itjust.works sounds like his ethos runs fairly parallel to my own w/how these online communities should kinda self run.
But what is https://rentry.co? Just a static site for the vote results?
I’d like to offer up the BeyondCombustion.net LimeSurvey instance or help set one up at su.rvey.shi.tjust.works or something, whatever. Real software for serious bidneess.
TheDude has set that up, I’m sure that we could migrate to another system pretty easily.
One issue i have that may seem irrelevant for some but for me as a non-Angloamerican it isn’t: The language used (aye/nay) alienates me very much. It gives me the feeling that “I am not at home here” but rather some USA citizens want to impose their antiquated understanding (bc. antiquated language) of democracy on all. Makes me wish i could be part of a more sophisticated community. … So why not “yes/no”? – still not my first language but at least it wouldn’t make me feel like being in the US senate or the British parliament (whatever language they use idk, it just feels that way). ;-)
Apart from that, i’m still in favour of ranked choice. (Edit to clarify: even in yes/no polar choice there always needs to be a third option, “I want neither of these”, or “I abstain”, so that all voters are counted, not only those who have a polar opinion.)
Don’t just dismiss it because some evaluation methods don’t work well. The issue there is not with RCV itself, but it’s in the (antiquated!) idea that a winning candidate must have absolute majority and therefore the votes need to be reduced until only two candidates are left. – Of course such a method will produce a skewed output because it discounts parts of the votes as “irrelevant”. –What i was already suggesting elsewhere is as i learned, called “Borda-Count”. In that, no candidates or votes are eliminated by the evaluation but the rankings are simply added up. This allows for minority winners and inconclusive polls. So, if opinions are divided in a community, the poll would reflect that – which is the way to go, imo.
So why not “yes/no”? – still not my first language but at least it wouldn’t make me feel like being in the US senate or the British parliament (whatever language they use idk, it just feels that way)
In the UK it’s aye/no, because sticking to a theme is to simple. /s
In reality the real reason is that the whole aye/nay thing is made up. It’s always been aye or no.
And for our part, we only chose aye/nay because that’s what people were doing anyway. It seemed easier for everybody to keep it consistent.
Yes/No, Aye/Nay, it shouldn’t matter too much as long as your position is clear. Using a specific word makes it easier for the preliminary count (using a script) but it’s just as easy to count Yes AND Aye as a yes vote.
