Deployer
@deployer
I hate the new channel changes. Passing the buck to channels to implement spam moderation feels like a cop out. Big channels don't have the resources to constantly patrol for spam. The client should be responsible for a minimum level of spam filtering. I don't think Warpcast has solved spam filtering yet and passing it off to channels isn't the answer.
20 replies
13 recasts
53 reactions
Ryan J. Shaw āļøš©
@rjs
So channels can have up to 10 moderators now... if the channel can't find 10 serious moderators then isn't that a problem with the channel itself? Because IMO they tried automated filtering and it is always going to create an us vs. Warpcast dynamic, which is unhealthy. Now it's an us vs. community moderator dynamic, and the option is always there to splinter and create a new community -- market forces at its best IMO.
1 reply
0 recast
5 reactions
Deployer
@deployer
I agree in part. The first attempt at spam filtering didn't work, because of the Us vs Warpcast dynamic. They labeled a lot of real people and contributing members of the network as spammers. Of course people would get mad. But I don't think that means spam filtering should solely fall on human moderators. There's some clear indications of spam like repetitive messages over and over of completely out of context posts. AI generated gobbily gook. Warpcast has raised a shit ton of money at an incredibly high valuation. The hard problem to solve is spam. Imo they should focus on solving it.
1 reply
0 recast
4 reactions
Ryan J. Shaw āļøš©
@rjs
I think the cool thing about it being open data is anybody can build a spam moderator bot of the sort you describe, right? It'll just take some time? And even @warpcast can offer that as a service if they don't want to throw their code away (I suspect they won't, or if they do it'll be on a trial basis, as they are really aggressive on sunsetting capabilities that don't get extensive usage)
1 reply
0 recast
1 reaction
Deployer
@deployer
Yeah I agree. Automod was great and I think Farcaster opens the door to new companies to be formed around moderation as a service, or algorithm as a service as the network continues to grow. With that being said it seems that Warpcast is supposed to be THE client. So, if Warpcast will ever compete with the likes of X or reddit then there needs to be proper automated spam solutions, where it comes directly from the Warpcast team, or a third party. Honestly I'd be happy with either.
2 replies
0 recast
2 reactions
ted (not lasso)
@ted
encourage you to check out what @haole and @sahil are building (and potentially others, too!) @yassinelanda.eth not sure if @mbd has built anything yet
0 reply
0 recast
3 reactions
Ryan J. Shaw āļøš©
@rjs
Completely get your point. @dwr.eth @v I'm guessing you considered migrating the spam algorithm over to a moderator bot but decided against it? In theory, it would've eased the transition to Channels 2.0 while market forces worked on alternatives. Or was the concern that everybody would use the Warpcast-provided Spaminator Bot by default and you'd just end up fielding the same complaints about spam algos all over again, and market forces might not get involved at all?
0 reply
0 recast
1 reaction
