3 ms·

How is a twitter bot costing Twitter money?

mike-cardwell 16y ago
I wouldn't like to guess how many man hours it has taken them to develop the software and processes required to deal with spam on twitter? Then there are the ongoing costs of dealing with new spamming methods, and the increased hardware costs of having to deal with the extra load caused by the spam. And then there is the less obvious negative impact it has on their brand and image when their users are sent unsolicited spam.
- akkartik 16y ago
  But does it add to Twitter's costs in spam management? It's not nearly in the same league as untargetted, extremely high volume spam which follows a million people and sends out essentially the same ad over and over again. To answer your original question, yes, I think Twitter isn't bothered by spam until it gets past a certain threshold. They're in it to make money, not to wage religious war.
  - mike-cardwell 16y ago
    "But does it add to Twitter's costs in spam management?" Yes, it all adds up. If nobody did what he was doing then Twitter wouldn't have had to spent as much money on the problem.
    - akkartik 16y ago
      You're assuming spam is like trash: it accumulates from everyone sending small quantities. But spam is never a side effect. 99% of spam comes from a very few operators operating at extremely high scales. So even if nobody spammed book recommendations with affiliate links, Twitter would still have to deal with the large-scale operators.
      - mike-cardwell 16y ago
        It's a lot easier to deal with 10 spammers sending a million spams each than it is to deal with a million spammers sending 10 spams each.
        akkartik 16y ago
        1. Are you claiming there are millions of spammers sending 10 each? If so, your definition is different from Twitter's. If it's just a future worry I'm going to cross that bridge when we get to it. 2. I don't know if it's easier, but it's certainly not easy. Otherwise gmail and twitter wouldn't have spam anymore. It's easy for spammers to pretend to be somebody else. Get a new username, a new IP address, forge headers. Even if it's only 10 spammers, Twitter can't see that. What they see are a million different spammers. So they already have to deal with the problem you claim is harder.
        mike-cardwell 16y ago
        1. No. Read what I said. It's all about patterns. A single spammer uses a small set of patterns. It's not just down to IP address.
        jbz 16y ago
        You're pretty much taking the evolution of email spam and going backwards, unless you can point to a specific reason twitter is different from the general evolution of spam on every network up till now I cant see how thats right. Given a million people sending out 10 spams each theres bound to be a very similar way they carry it out. However 10 spammers dedicated churning out the maximum capacity they can achieve will put up a fight no matter how you try to stop them. Email spam originated with people like hood, generally not doing much damage and very easy to block and shut down. It eventually led to the current situation where vast majority of spam originates from a handful of spammers. So did comment spamming, and early social network spamming, they all pretty much followed this pattern. In my view you're letting yourself get wrapped up in the minutiae of the argument. What hood did was wrong and no amount of justification will change the fact that he externalized the cost of commercial advertising. However none of us can look back at our lives and say we didn't have such morally questionable moments. We generally react like hopefully hood has done and don't make it a career; assholes keep going even when they only make $1 for every $100 they externalize.
- what 16y ago
  What if it's only in response to people asking for book suggestions? Like this one: http://twitter.com/GEOFFUSELESS/status/21014481706 http://twitter.com/GEOFFUSELESS/status/21014481706
  - mike-cardwell 16y ago
    Then it would be solicited, and thus move from the black area into the grey area.

I wouldn't like to guess how many man hours it has taken them to develop the software and processes required to deal with spam on twitter? Then there are the ongoing costs of dealing with new spamming methods, and the increased hardware costs of having to deal with the extra load caused by the spam. And then there is the less obvious negative impact it has on their brand and image when their users are sent unsolicited spam.

akkartik 16y ago
But does it add to Twitter's costs in spam management? It's not nearly in the same league as untargetted, extremely high volume spam which follows a million people and sends out essentially the same ad over and over again. To answer your original question, yes, I think Twitter isn't bothered by spam until it gets past a certain threshold. They're in it to make money, not to wage religious war.
- mike-cardwell 16y ago
  "But does it add to Twitter's costs in spam management?" Yes, it all adds up. If nobody did what he was doing then Twitter wouldn't have had to spent as much money on the problem.
  - akkartik 16y ago
    You're assuming spam is like trash: it accumulates from everyone sending small quantities. But spam is never a side effect. 99% of spam comes from a very few operators operating at extremely high scales. So even if nobody spammed book recommendations with affiliate links, Twitter would still have to deal with the large-scale operators.
    - mike-cardwell 16y ago
      It's a lot easier to deal with 10 spammers sending a million spams each than it is to deal with a million spammers sending 10 spams each.
      - akkartik 16y ago
        1. Are you claiming there are millions of spammers sending 10 each? If so, your definition is different from Twitter's. If it's just a future worry I'm going to cross that bridge when we get to it. 2. I don't know if it's easier, but it's certainly not easy. Otherwise gmail and twitter wouldn't have spam anymore. It's easy for spammers to pretend to be somebody else. Get a new username, a new IP address, forge headers. Even if it's only 10 spammers, Twitter can't see that. What they see are a million different spammers. So they already have to deal with the problem you claim is harder.
        mike-cardwell 16y ago
        1. No. Read what I said. It's all about patterns. A single spammer uses a small set of patterns. It's not just down to IP address.
        jbz 16y ago
        You're pretty much taking the evolution of email spam and going backwards, unless you can point to a specific reason twitter is different from the general evolution of spam on every network up till now I cant see how thats right. Given a million people sending out 10 spams each theres bound to be a very similar way they carry it out. However 10 spammers dedicated churning out the maximum capacity they can achieve will put up a fight no matter how you try to stop them. Email spam originated with people like hood, generally not doing much damage and very easy to block and shut down. It eventually led to the current situation where vast majority of spam originates from a handful of spammers. So did comment spamming, and early social network spamming, they all pretty much followed this pattern. In my view you're letting yourself get wrapped up in the minutiae of the argument. What hood did was wrong and no amount of justification will change the fact that he externalized the cost of commercial advertising. However none of us can look back at our lives and say we didn't have such morally questionable moments. We generally react like hopefully hood has done and don't make it a career; assholes keep going even when they only make $1 for every $100 they externalize.
what 16y ago
What if it's only in response to people asking for book suggestions? Like this one: http://twitter.com/GEOFFUSELESS/status/21014481706 http://twitter.com/GEOFFUSELESS/status/21014481706
- mike-cardwell 16y ago
  Then it would be solicited, and thus move from the black area into the grey area.