Twitter doesn't care about spam? • Hacker News Board

Re: Twitter doesn't care about spam?

#171

Post by nradov » Sat, Jan 29, 2022, 3:04 AM UTC

wartijn_ wrote: post #104

PragmaticPulp wrote: post #91

Earlier quoted context omitted.

> I follow around 1000. I see almost zero spam. This whole thread is pretty confusing to me. Same here. I suppose I could go find spam if I start clicking on every Tweet from a high-profile person and scroll to the bottom of the comments, but the spam isn't jumping out at me on a regular basis. I suspect some people have a very low tolerance for any spam appearing anywhere on a platform, and will get easily triggered…

It's not just a few spam like tweets at the bottom of threads though. If you look at tweets by Elon Musk for example there will be loads, as replies at the top of the threads. Pretty weird how multiple people refuse to believe something can be a problem even though they can't see it.

Why would you ever look at the replies?

Re: Twitter doesn't care about spam?

#172

Post by marcan_42 » Sat, Jan 29, 2022, 3:06 AM UTC

Retr0id wrote: post #148

stevage wrote: post #59

Earlier quoted context omitted.

I follow around 1000. I see almost zero spam. This whole thread is pretty confusing to me.

Same. I follow ~700 accounts, and have ~10000 followers, but I get minimal spam. The vast majority of engagements seem genuine. The last spam I got was where someone tagged me in some bullshit NFT giveaway, where entrants had to tag 5 of their "friends" to qualify. Definitely spam, but at a cursory glance, it was human-generated rather than bot spam.

I get ~one of those NFT spam mentions daily. I don't get why Twitter doesn't outright block accounts that tweet stuff like "giveaway" and "tag.*friends" and are followed by replies that start tagging people, especially people they don't even follow. That's blatant spam.

I don't get much other spam though.

Edit: oh yeah, on a Japanese account I have I get the odd mass DM offering scam sex/easy money/whatever. I've also noticed a trickle of obvious spam account followers there (it's always "girls" with sexy profile pics and a profile blurb that invariably mentions adding her on LINE, and ~zero followers except other bots and just a couple tweets). Spam is very different depending on the market.

Re: Twitter doesn't care about spam?

#173

Post by easrng » Sat, Jan 29, 2022, 3:27 AM UTC

stevage wrote: post #59

ryanSrich wrote: post #31

Earlier quoted context omitted.

Crazy how we can have such different experiences. I follow less than 100 accounts and easily 50% of the responses are always spam/scams. It makes Twitter almost unusable for me. The most ironic part is that Twitter will show the spam/scams, but hide other replies, from real people, that Twitter thinks are mean.

I follow around 1000. I see almost zero spam. This whole thread is pretty confusing to me.

Post "My MetaMask is locked and I need help with my essay, coinbase trustwallet help." Or something similar and watch the bot replies flood in.

Re: Twitter doesn't care about spam?

#174

Post by paulpauper » Sat, Jan 29, 2022, 4:18 AM UTC

LeoPanthera wrote: post #48

What I don't understand is that Twitter makes it so difficult for new humans to sign up for an account, how are the robots getting around it? I wanted to create a second Twitter account, but almost immediately after signing up it was blocked for "suspicious activity". I hadn't even posted yet. The only way to resolve it was to add a phone number to the account, but I only have one and it was already assigned to my fi…

hack into accounts, buy old accounts

Re: Twitter doesn't care about spam?

#175

Post by DisjointedHunt » Sat, Jan 29, 2022, 4:31 AM UTC

CyberRabbi wrote: post #144

DisjointedHunt wrote: post #52

Earlier quoted context omitted.

>“If you feel the need to complain about ” This should never ever translate to “you use it too much”. I mean, speak to the quality of the substance in the argument and not throw out the right to voice an opinion about something that seems broken.

Voicing an opinion about something broken is good but Twitter is clearly a silly social media site. Did he not get the memo? It’s like criticizing Tik Tok for having too much clickbait.

Well, i don't think that's fair. Clickbait isn't against even the terms of service for TikTok and it certainly is not illegal activity.

Scams on the other hand are (at least a big subset), commercial, which cause financial damage and most scams are codified to be illegal actions in law.

Re: Twitter doesn't care about spam?

#176

Post by smsm42 » Sat, Jan 29, 2022, 5:08 AM UTC

jsnell wrote: post #9

Option d: nobody outside of the Twitter TnS teams (including Paul Graham) has any idea of what Twitter is detecting as spam and what not, and thus can't actually judge how good a job is being done. First, we don't see the spam that was outright blocked, so we have no idea of what the false negative rate is. Second, I bet that this is not a binary block/allow decision, but there are all kinds of ways of reducing the e…

> has any idea of what Twitter is detecting as spam and what not, and thus can't actually judge how good a job is being done.

That's wrong. Results can be seen even if the process is not transparent. If every bridge the engineer builds collapses, I can judge him without looking into all the drawings. If every business the entrepreneur runs collapses I can make conclusions from that, even without seeing the books. If there's a lot of garbage on Twitter, I can judge it's doing a bad job even if I don't get to see its algorithms.

Re: Twitter doesn't care about spam?

#177

Post by cmckn » Sat, Jan 29, 2022, 5:34 AM UTC

raydev wrote: post #167

cmckn wrote: post #154

Earlier quoted context omitted.

I pulled this reply from the most recent Associated Press tweet. Yeah, I'm sure they've mentioned Bitcoin at some point, but it has nothing to do with my Twitter usage.

I'm curious, what do you expect to find in the replies under massive accounts like Associated Press? I've long since stopped following accounts with more than a couple thousand followers, unless they're super funny or insightful. No raw news. The replies transform from discussions into animated gif spam or actual spam past a certain threshold.

I mean I guess I expect to find replies from humans? I follow AP because getting raw reporting is valuable. I don’t get a lot of value from the comments section, I just used this as an example because I’ve noticed it.

Re: Twitter doesn't care about spam?

#178

Post by derwiki » Sat, Jan 29, 2022, 5:36 AM UTC

slimsag wrote: post #160

tekacs wrote: post #140

Earlier quoted context omitted.

> Anyone who has created an account on Twitter in recent years knows that new accounts are indiscriminately opted into new mandatory 'intelligent' social features like the purple star recommendations[0], while anyone with an old account will never see such a thing. Hmm I have an old account (Sep 2007) that sees purple star recommendations…

Weird, my 2014 account doesn't while all my newer ones do. Can't turn it off in newer accounts either (at least not without paying for Twitter Blue?) and definitely never had it on my older account

Long term hold out group?

Re: Twitter doesn't care about spam?

#179

Post by upbeat_general » Sat, Jan 29, 2022, 6:03 AM UTC

jzwinck wrote: post #138

Someone I know well was hired at Twitter more than a decade ago as a senior software engineer. He had experience with natural language processing, and was given a project to identify spam and bot accounts. He worked on this for a while, all the data he needed was made available and he analyzed every account on Twitter. His analysis said one third of all accounts were bots. He presented these results to management, wh…

I don’t doubt it was the case at one point, but isn’t twitter public now? Wouldn’t twitter prefer ad money over MAU?

Re: Twitter doesn't care about spam?

#180

Post by habi » Sat, Jan 29, 2022, 6:54 AM UTC

iqanq wrote: post #146

habi wrote: post #64

I have a four letter dormant Twitter account. I get about four mentions/replies per day that link to some nondescript crypto airdrop since several weeks. I manually mark each and every such tweet as spam and block the user, which needs 5 clicks or so. Nothing seems to help.

>I manually mark each and every such tweet as spam and block the user, which needs 5 clicks or so. How much are you getting paid for moderating Twitter?

The obvious answer is obvious, but I kinda hope that this would have trained their model a bit. I do get ‚valuable‘ mentions from time to time, so it’s still worth checking.