News for hackers. Stuff that matters.
Re:The Follower Factory by sgk284 on Saturday January 27, 2018 @07:02PM (#16248845)
I used to work at Twitter, so not sure I can get into specific numbers, but they remove hundreds of thousands to millions of accounts every day (literally every day). It's particularly tricky because bots aren't inherently against the ToS (e.g. the earthquake bot cited in the article), so you can't just ban based on "Is this account behaving like a human?".

Many "bots" though are actually giant networks of humans being orchestrated and they share things that are compelling to a subset of real users (e.g. Hillary did this bad thing... lock her up), so now real users are retweeting and muddying any signal that was there to detect a bot network. It's an incredibly challenging problem at scale and you basically only see the < 1% that Twitter missed.

They've got a pretty sizable team that works on it so hopefully in a few years they'll have solved it much like Gmail has mostly solved spam.

Re:The Follower Factory by Retra on Sunday January 28, 2018 @01:46PM (#16252641)
Maybe it's the "one click to share" behavior - an extremely low-resolution action - that is the problem? Perhaps things would be easier to fix if the human engagement component were more than "can you press a button?"
Re:The Follower Factory by djsumdog on Saturday January 27, 2018 @08:31PM (#16249175)
> much like Gmail has mostly solved spam.

Gmail has not solved spam. Gmail and Microsoft dump so much legitimate e-mail that it's basically made e-mail totally unusable:

http://penguindreams.org/blog/how-google-and-microsoft-made-...

Re:The Follower Factory by fjsolwmv on Saturday January 27, 2018 @09:40PM (#16249451)
No they don't. They, like most abuse sensitive recipients, don't accept mail from small indie servers, so people know not to use small indie servers to send mail.
Re:The Follower Factory by fyfy18 on Sunday January 28, 2018 @02:16AM (#16250239)
I’ve read this a lot, often as a counter argument to running your own mail servers, and I believe it to be mostly FUD.

I run a small SaaS and it sends out quite a lot of notification emails to customers (they are paying me for these emails, they aren’t spam or newsletters). I was originally using SendGrid but their deliverability rates were pretty bad and I was about to hit a ceiling for my plan, which would meant the monthly cost would more than double, which was unaffordable for me.

I decided to try setting up my own mail server. Everyone said don’t do it, you emails won’t be delivered, it’s too much work, etc. I setup a VPS with OVH as I’ve heard their policies regarding email spam are pretty strict (I contacted them before to clarify - if they find out you are spamming they cut your off ASAP to protect their other customers). The IP I got was actually on one blacklist, but I got that resolved in a few days.

The deliverability rate was better than SendGrid and it was costing me 1/10th of the price. One thing that I would assume helped is the domain had already been in use for a few years, and I setup proper DKIM and SPF records.

The server has been running for a year and a half now, and the only bounces I get are due to invalid or suspended email accounts. Most of my customers are small to medium sized businesses, a lot of which run their own Exchange servers - with SendGrid those were usually the ones who were bouncing my mail due to being on a blacklist. Now admittedly just being accepted doesn’t mean it is ending up in a users inbox, but I haven’t had any complaints from paying customers.

Re:The Follower Factory by bzbarsky on Sunday January 28, 2018 @01:25AM (#16250121)
It's not just small indie servers. They do "reputation scoring" of senders that penalize them for things like sending plaintext mail. See https://groups.google.com/forum/#!msg/mozilla.governance/WWK... for more details of one example.
Re:The Follower Factory by cookiecaper on Sunday January 28, 2018 @01:27AM (#16250124)
Right! We can't trust those "small indie servers" on the world wide web. Who knows where they've been?! Make sure you use mail servers furnished by Google, Microsoft, and other NSA-approved facilitators. They only cost $10 / mo / user, that is easily affordable for anyone who is not a Russian agent tricking the unwashed masses into preventing Hillary Clinton from ascending to her rightful place as ----- errrm, uh, from "corrupting democracy".

Any non-Google, non-Microsoft mailers will be blocked, lest users be asked to renew their VIAGRA prescription from a Canadian pharmacy, or a non-compliant political message be received. It's for your own good, citizen!

Thank you, wise and great Google! Thank you, wise and great Microsoft! I wouldn't know how to use this free, non-controlled computer network without your benevolent work!

unsnark: Seriously, the way that Google et al have captured "anti-spam" mechanisms and converted them into "pay a big company for your mail" mechanisms is extremely sad, and speaks to the true forces at work behind all the warm and fuzzies about an "open web".

Re:The Follower Factory by lucb1e on Sunday January 28, 2018 @05:02PM (#16253852)
Even if that were true, then your parent's statement still holds: "basically made e-mail totally unusable"

If it's down to a few big senders and the rest is rejected, they might as well get rid of SMTP, close it off, and just talk a proprietary protocol among each other.

Re:The Follower Factory by pilif on Sunday January 28, 2018 @02:21AM (#16250256)
I'm not convinced about this generalization. Case in point is the mail relay server of our company which definitely counts as a small indie server and whose email is accepted and not marked as spam.

Of course we have configured SPF, DKIM and a proper reverse dns lookup on both ipv4 and IPv6, but that’s just being a good email-citizen anyways.

Re:The Follower Factory by djsumdog on Saturday January 27, 2018 @08:30PM (#16249169)
I think it was the 99% Invisible podcast that did an interesting report about whole networks of people who were paid in Mexico to spread election messages; and these networks are often reused by gangs or other organizations to promote stuff or harass opponents. So in this case, they're not bots. They're actual humans, because it's cheaper to do that than pay people to write scripts.
Re:The Follower Factory by amsilprotag on Saturday January 27, 2018 @10:53PM (#16249708)
I don't remember that 99pi. Was it Reply All?

https://gimletmedia.com/episode/112-the-prophet/

After Andrea is attacked by a stranger in Mexico City, she just wants to figure out who the guy was. Investigating this question drops her right into the middle of one of Mexico’s biggest conspiracies.

Re:The Follower Factory by djsumdog on Sunday January 28, 2018 @09:43AM (#16251457)
Ah yes, sorry you're right. It was Reply All.
Re:The Follower Factory by kakarot on Saturday January 27, 2018 @10:55PM (#16249716)
Additionally, you can train a human to become better at astroturfing over time, and train your network as a whole to be resilient to partial takedowns.

These things are much easier to orchestrate than bots, especially if your specialization is not CS but scamming / manipulating opinions, and in many cases, especially with scale, human labor can actually be more expensive.

Of course, subsidizing the labor to impoverished countries definitely helps; these networks tend to lack the sophistication used in more sensitive political or industrial topics.

Sometimes for national or global campaigns you'll find a mix of both cheap labor, a small team of experienced astroturfers with a good grip on their persona acquisition and management, and bot networks, each targeting a different set of demographics. If your target demo is sufficiently ignorant or radicalized in their beliefs, it doesn't take much to convince them and bots will do just fine.