3 ms·
I'm surprised that no one has mentioned how this relates to spam fighting. One could speculate that by switching to this "shadowban" system, spammers have less
by newhouseb 13y ago
I'm surprised that no one has mentioned how this relates to spam fighting. One could speculate that by switching to this "shadowban" system, spammers have less data to figure out whether their accounts are being banned/blocked or not. This is pretty common amongst other large web properties -- Reddit and Yelp are two that come to mind. Even if the banning is on a per-user basis, one could imagine that twitter uses ban counts as a metric for spam detection and so if a spammer cannot measure this metric themselves, they won't know when to taper/accelerate their spamming to game the detection algorithms.
- icebraining 13y agoA spammer can just access the spammed accounts' pages and see if it has new tweets that the spammer hasn't received. Seems easy to me.
- newhouseb 13y agoThat's how it originally worked, but with the proposed changes, turning blocking into effectively muting, a spammer can still see tweets from the spammed account, but the spammed account will no longer see tweets from the spammer. Note: twitter has backtracked on this change anyway, so it's back to how it originally worked.