My blog has been attracting trolls recently. Or more likely, I suspect, one troll. They comment most days on my latest blog post under various different names, using a different email address and from a different IP address each time.
Despite the different names, email addresses and IP addresses each time there are consistencies that betray a connection. The domain name of the email addresses are always either that of a throwaway email address provider or a large website. Mostly not an major email provider.) The first part of the email address is often five characters starting with “n”. The name is usually a short form of a male name such as bob, dan, matt, fred, and often has a number like dan 50. They always post the same kind of message – that people won’t work because they’re better off on benefits, that Britain is broke, that British people are lazy and that’s why foreigners have taken all the jobs, that I’m a scrounger. Sometimes they post a link to the daily mail.
You can probably agree that I have reason to suspect this is all the same person. But more than that, I wonder if this person is using an online persona management service like that known to have been requested by various US government agencies, and produced by HBGary among others and sold to governments and corporates. This kind of service and software allows one operator to handle dozens of identities which are used to leave comments anywhere their employer wants a message promoted. It is used to give the impression that there is a lot of disagreement with an idea, or to try to discredit someone by pretending to prove them wrong or just calling their ideas into question. This software is known to exist because one of the companies that make it was hacked in 2011 and had their communications published. A request for this kind of service by a US government agency was also published openly, possibly accidentally on a website for suppliers of services to governments.
Daily Kos writes:
“According to an embedded MS Word document found in one of the HBGary emails, it involvescreating an army of sockpuppets, with sophisticated “persona management” software that allows a small team of only a few people to appear to be many, while keeping the personas from accidentally cross-contaminating each other. Then, to top it off, the team can actually automate some functions so one persona can appear to be an entire Brooks Brothers riot online.”
Who would use this kind of service? And why would they bother with me? I think there are a fair few interested parties who would want to discredit my message about welfare and benefits. Big insurance companies, for one, and companies involved in administering benefits now and in the future. Various think tanks too. We know that insurance company UNUM has been “advising” government for years, and has actually boasted in writing that they influenced welfare policy. Atos has an interest in taking a cut for assessing people. I’m not claiming that any organisation has done this, but these are exactly the kind of customer that persona management services are marketed to.
I don’t think this is paranoia on my part, it’s just an idea I’m pondering. I’m still not sure if my troll is a lone person or several. I’m not claiming to be stalked by the military, I’m talking about software that is sold to military, government agencies, and corporates alike. I’m not talking about being targeted by spy agencies, I’m talking about being on a list of blogs and social network users that have ideas that are vaguely troublesome to the customer for this software. I’m not talking about anyone spending time on me, I’m talking about a person who has dozens of targets and a short brief about what kind of message to post there every so often under a new identity each time. Commenting on my blog is a tiny fraction of their time.
The troll on my blog doesn’t usually read the blog posts, they just scan for keywords like “benefits” and then post their message. They don’t see messages like “your comment is awaiting moderation”, they just try to post four times over a couple of hours. When I block their IP address they don’t come back immediately, they appear the next day on a new address – as though they waited for software to move on. They reply to people who reply to them but they just say more of the same without addressing questions. They are repetitive, saying much the same thing each time but written slightly differently, like someone has a brief to convey a certain idea but has no awareness of context.
I’m probably way off the mark here, but it’s an interesting idea that we could be seeing the work of persona management software in the wild. The messages left on my blog could have been left by people using this kind of system. But they’re probably the work of a particularly witless nasty troll. Thanks for the inspiration for a blog post, dan / fred / wes / dave / ozzy / mark / jane!
I’ve been checking the troll’s IP addresses at BotScout.com and all but two that I have checked have been known sources of automated comment scripts. It appears that the system being used might not be as sophisticated as the persona management that I talked about, but is instead pre-written comments posted automatically by bots on blogs with welfare-related keywords.
The Register: HBGary ‘puppets’ FAIL to convince
The Guardian: Revealed: US spy operation that manipulates social media
Daily Kos: The HB Gary Email That Should Concern Us All
Daily Kos: HB Gary Federal CEO a Daily Kos Member?
Pastebin: An archive of the troll comments left on this blog – with email and IP addresses
Pastebin: IP addresses used to post troll comments