Varun Srinivasan pfp
Varun Srinivasan
@v
AMA - anything you want to know about spam after reading this thread
16 replies
12 recasts
133 reactions
Pichi 🟪🍖🐹🎩 🍡🌸 pfp
Pichi 🟪🍖🐹🎩 🍡🌸
@pichi
If we interact with people who are labeled, does that help the algorithm see them to be reevaluated?
1 reply
0 recast
7 reactions
Mikado pfp
Mikado
@mikadoe.eth
About tagging there a myth going around that tagging more than 3,4 per cast is spammy. Is there any truth to that? If I'm analyzing sth and tag users to compare sth and not all of them interact with the cast is that bad?
1 reply
0 recast
0 reaction
KMac🍌 ⏩ pfp
KMac🍌 ⏩
@kmacb.eth
1- Thoughts on publishing the output for other app devs to utilize? Or is this a strategic differentiator for wc? 2- Any plans for reputation scores from the perspective of an account vs the entire network? Your spam might be my friend saying hello to me in an inside joke fashion of one letter at a time
1 reply
0 recast
0 reaction
Ben Adamsky 💭 pfp
Ben Adamsky 💭
@ba
How do well intentioned high cast volume bot accounts like @survey get categorized? For reference we’re looking to move most of the comms to DCs and make our trending posts more high signal, is that enough to reverse any spam markings? And how long would the WC algo take these changes into effect?
1 reply
0 recast
3 reactions
Jorge Pablo Franetovic 🎩 pfp
Jorge Pablo Franetovic 🎩
@jpfraneto.eth
what album comes to you when you think about tackling the spam problem on farcaster?
1 reply
0 recast
3 reactions
kugusha 🦋 pfp
kugusha 🦋
@kugusha.eth
How to deal with this? screenshot from /art around 6 hours ago
1 reply
0 recast
0 reaction
Boo-ca🔵🔴 pfp
Boo-ca🔵🔴
@luca1111
I've noticed a few friends who took a break for exams have been labeled. Could you clarify if this counts as spamming(Level 3)? Knowing that contacting support would help might encourage those who left to come back.
1 reply
0 recast
1 reaction
‍‍ pfp
‍‍
@git
crazy how much things have improved. From a point where every cast I posted received upwards of 100 chinese replies to now, where it's better than X and bluesky. with the knowledge you have today, what would you have done differently when you were building the protocol?
1 reply
0 recast
0 reaction
jj 🛟 pfp
jj 🛟
@jj
How do you label if something is spam or not? I’m assuming you have a ground truths and running your training sets on these labeled data sets. If it’s community sourced how do you ensure data quality? Do you have a contingency matrix for your spam models?
1 reply
0 recast
0 reaction
Sarun Vrinivasan pfp
Sarun Vrinivasan
@dristov
I am almost 100% sure that the system will also identify regular accounts as spam. For me, this is quite relevant, because it is difficult to distinguish spam from post-irony, jokes, delusional messages, which are the main purpose of maintaining your account. It is the type of humour, not spam. Of course, this will happen with such accounts, because they are checked by the system, and not by a living person. What should alt accounts, meme accounts and other trolls and comedians do?
1 reply
0 recast
0 reaction
tim/vortac pfp
tim/vortac
@vortac
Unrelated to the content but a question regarding the form of this cast. Hope it’s ok to put it here… Have you ever considered a native, dedicated cast type for threads? Instead of having a main cast and replying to it yourself for the consecutive messages? So it would be easier to create one (create the thread at once before releasing, saving the thread as draft) and maybe even have special features? I. e. an indicator it’s a threaded cast, maybe an accordion view, definitely keeping the order, index numbers amm I believe it would align well with having more interesting and in-depth content. I miss a dedicated innovative solution to tackle it. Thanks for considering and have a great Sunday 🙏
0 reply
0 recast
0 reaction
Alemac pfp
Alemac
@alemac
Yesterday I saw a lot of new friends getting the $5 reward for being a top /warpcast caster. Dan told me it was because the spam model improvement. That's mean that these people were labelled as spam or it's not related? Thanks in advance.
1 reply
0 recast
3 reactions
chillyazz.base.eth🎩 pfp
chillyazz.base.eth🎩
@chillyazz.eth
are there any correlation between raising spam activity and large projects doing airdrops on Farcaster or smth
0 reply
0 recast
0 reaction
Jacque(she|her).⌐◨-◨ pfp
Jacque(she|her).⌐◨-◨
@jacque
One can still hack or pay for fake validity on these sites IE Twitter, Instagram, TikTok Etc and be a spammer but obtain a high follower count trying to look legit. So why choose to use Twitter as a verification over other means? Personally, I only post mints there and have been trying to sunset it for a while. Back when digital NFT art sites made us verify with it I had to create a new account because I had deleted one from 2004 or so which some jabroni took from me as soon as my time to get it back lapsed. Could there be a new way vs stringing us into web2 again?
0 reply
0 recast
0 reaction
Pornsoup 🎩 pfp
Pornsoup 🎩
@pornsoup
What all behaviour can expedite the process of changing from Level 2 to Level 1? how often does the changes happen between these levels?
0 reply
0 recast
0 reaction
frodo.base.eth 🔵 pfp
frodo.base.eth 🔵
@frodoisbased.eth
is there an opportunity for a user who has been assigned level 3 to move to level 2 or 1 if they have started posting interesting content?
0 reply
0 recast
0 reaction