One thing I wonder, if you look at the comments here it is clear the moderation and the general hostility of the site was a big reason why a lot of people left or didn't engage with it. Yet, I can't recall SO doing anything about this, and its confusing to me why. Was it: a) They don't acknowledge this was a problem. b) They are unwilling to tackle this problem. c) They are unable to tackle this problem. I'd love to…
What AI did to stackoverflow in a graph
491–500 of 615 posts
Re: What AI did to stackoverflow in a graph
#492Earlier quoted context omitted.
The gamification was a key reason the site became popular, but as the site grew, the rules and game mechanics did not evolve. An edit that made a response worse should have knocked the mod down so that they were unable to mod any more. The quality of the edit should have been determined by the original author. "Did this edit make your question better?" Moderators should have been ranked and scored based on their abil…
> An edit that made a response worse should have knocked the mod down so that they were unable to mod any more. Literally everyone on the site is permitted to propose an edit, and everyone with at least 2000 reputation can make unilateral edits. The proposals are approved by a 2 out of 3 majority of random unilateral-edit-privileged users. None of this is considered "moderation" and is not done by "mods". Of millions…
Message received I guess, seeing as how the overwhelming majority left.
Re: What AI did to stackoverflow in a graph
#493If someone wants to get help, and they end up on Stack Overflow, the structure looks like a Q/A site, and they treat it that way. The people "in the know" then accuse them of being "bad" for posting a question to a documentation site. Combine that with the gamification, and it had to turn into the toxic mess that it did.
When I first heard of the gamification method of moderation, I thought it was a really good idea. But, over the years, it has proven to be a means to enforce an echo chamber, rather than a way to filter out bad posts. I have yet to find a site (HN included), where that isn't true.
Re: What AI did to stackoverflow in a graph
#494Earlier quoted context omitted.
Re: quis custodiet ipsos custodes... Slashdot has a meta-moderation setup where random users (with at least a minimum tenure and rating on the site) would get to vote on the quality of the moderation for randomly selected posts. I still think that this has a lot of potential for improving moderation, even if it's just used as a way of ferriting out problematic moderation.
Yeah, I really like Slashdot's approach to both regular and meta moderation. I don't understand why it never caught on elsewhere.
Re: What AI did to stackoverflow in a graph
#495Earlier quoted context omitted.
I reverted nearly every single edit to my answers other than obvious typos. If I’d have meant something other than what I said, I’d have said it. You want to see your own words on the page? Write an answer of your own. Lordy, that use to piss me off most fiercely. I don’t want someone else’s words attributed to me.
It is taken very seriously in Stack Overflow editing policy that edits are not supposed to go against the author's apparent intent. But it does involve removing a lot of things that you might want the answer to say, but don't meet the guidelines for what answers are supposed to contain. When you join the site, you agree to Terms and Conditions that, among other things, grant a Creative Commons license to the communit…
Re: What AI did to stackoverflow in a graph
#496Earlier quoted context omitted.
Yeah, I really like Slashdot's approach to both regular and meta moderation. I don't understand why it never caught on elsewhere.
The big downside of the moderation system on Slashdot is that you can’t both participate in a discussion and moderate at the same time. I want people who are interested and well informed on a subject doing both in the same discussion! Meta-moderation was an interesting idea, though they seemed to have stopped promoting it much by the time I stopped posting there often. I’m not convinced it’s as effective as having a…
I get that, but moderation points carried a lot more weight than they do most places because a randomly code chosen subset of users would only get like 5 points… so you had to choose carefully where you used them. Since it wasn’t super common to get mod points, it’s not like you were limited often.
And the users on that site, frankly, could be downright feral in its heyday, let alone now. I think it needed more structure, not less. If there was even one person that would clean up the garbage, maybe the discussion would be civil enough to allow that. They can’t even be arsed to remove swastika ascii art half the time. That’s what happens when you address fundamentally human problems with purely technical solutions.
Re: What AI did to stackoverflow in a graph
#497Re: What AI did to stackoverflow in a graph
#498Earlier quoted context omitted.
Their rules, (I believe unintentionally) give iron-fisted fiefdom rulers a toolbox of justifications to control and alienate under the guise of protecting the quality of the site data. I honestly don’t even think most of the control freak mods could objectively judge the propriety of their actions because it was all encouraged by the rules. (And I do not think this was universal among the mods, but it was certainly e…
> Their rules, (I believe unintentionally) give iron-fisted fiefdom rulers a toolbox of justifications to control and alienate under the guise of protecting the quality of the site data. That isn't what happens. I know many, many people believe it to be what happens, but I know from years of seeing the process on the inside that it's absolutely not what happens in the overwhelming majority of cases. The "quality of t…
I was the only one with formal education and a professional background consistently answering, and frankly, that site needed my expertise a hell of a lot more than I needed to share it— and they failed to provide a reasonable forum to do that despite being its sole purpose. If the point system actually represented expertise, it might have worked. I’m fine with being scrutinized by a peer or superior… But it didn’t go down like that.
Re: What AI did to stackoverflow in a graph
#499Earlier quoted context omitted.
As someone who came into the industry in college, the problem with SO was simply that it was too hard to ask a question. They were up your ass about minutia that really didn't matter. Good riddance and can't wait to visit the site and see an EOL static page.
> As someone who came into the industry in college, the problem with SO was simply that it was too hard to ask a question. That's because you were intended to use the site like GP describes, and not by asking simply because you want to know something. > They were up your ass about minutia that really didn't matter. What you consider "minutia" were critically important, because the entire point was to optimize for GP'…
Re: What AI did to stackoverflow in a graph
#500Earlier quoted context omitted.
> Their rules, (I believe unintentionally) give iron-fisted fiefdom rulers a toolbox of justifications to control and alienate under the guise of protecting the quality of the site data. That isn't what happens. I know many, many people believe it to be what happens, but I know from years of seeing the process on the inside that it's absolutely not what happens in the overwhelming majority of cases. The "quality of t…
Mod vs non-mod-user-with-edit-privileges is a semantic difference from a user perspective. Nobody gives a shit what the internal labeling system looks like or hierarchy among the people with edit privileges. I’m not going to litigate my case in front of a clique of other officious hall monitors just for the privilege of making that site better. I don’t care if it was against the rules for me to be annoyed by their ob…