[dmix]

It's only been ~1yr since the election. You have to be patient for big cases like this.

Until then we have to trust the US intelligence community. Their messaging has been consistent that their arch-enemy they've been competing with for decades pulled a fast one on them and was unexpectedly effective at influencing an election.

Government sources have been leaking a bunch of stuff to the press in the meantime. This week it was alleged "Russian-linked" sources supporting the Republican party had bought Facebook ads [1] in the critical swing states:

> A number of Russian-linked Facebook ads specifically targeted Michigan and Wisconsin, two states crucial to Donald Trump's victory last November, according to four sources with direct knowledge of the situation.

Two states which Clinton's $500 million campaign reportedly neglected [2] despite the pleas of her former-president husband and advisors:

> Clinton made no visits to Wisconsin as the Democratic nominee, and only pushed a late charge in Michigan once internal polling showed the race tightening.

The other big leak was that of the hundreds of people the Trump campaign staff met and had phone calls with in 2016, it turns out 2-3 of them had connections to the Russian government. But it's not clear if they had any follow up meetings.

I'm looking forward to the full report showing the "critical role" Russia played in getting him elected...

[1] http://www.cnn.com/2017/10/03/politics/russian-facebook-ads-...

[2] http://wtvr.com/2016/11/11/bill-clinton-strategy/

[c3534l]

> You have to be patient for big cases like this

So in other words, no “verifiable“ proof at all, just rumors, hearsay, and unbacked assertions.

> Until then we have to trust the US intelligence community.

Sorry, no. Their success rate and overall trustworthiness is abysmal, let alone it's foolish to trust secret data even if it weren't the case.

[jejeuei73737]

You are out there every day pushing back against anonymous voting then?

Who voted for which candidate is secret and we rely on counts that aren’t broadcast live

[c3534l]

Anonymous doesn't mean secret. If the governor of Ohio says his people want Trump over Hilary, just trust him, you don't need to know how he knows would you? I mean, I don't trust unauditable electronic voting machines. I certainly wouldn't trust someone who couldn't even point to the machine he got his info from.

[linkregister]

Much of the voting system is an exercise of trust. The most reliable part of it, besides the fact that election officers are publicly known and accountable, is that officers from both parties oversee each voting center. Your trust is that the adversarial nature of the political parties will raise alarms if the ballots are tampered with.

[nl]

It's odd that people keep saying to trust the intel agencies, since much of the best research on the propaganda side is done in academia, and the data is out there.

Here's a Tableau of the influence of just 6 of the ~200 recently banned Russian-run groups on Facebook: https://public.tableau.com/profile/d1gi#!/vizhome/FB4/TotalR...

Here's the raw data: https://data.world/d1gi/missing-fb-posts-w-share-stats

Here's some analysis: https://www.washingtonpost.com/news/the-switch/wp/2017/10/05...

I also like http://dashboard.securingdemocracy.org/

If you want to look at research, Oxford is doing some good work in this area. They have a whole research group on computational propaganda: http://comprop.oii.ox.ac.uk/category/publishing/academic-art...

[dmix]

I read through each of your links and there's some interesting stuff there. Overall it looks like Russia basically ran a full-time campaign marketing team with access to a few million dollars for ad and social media sharing buys.

That said, the breathlessly covered "tens of millions of ad impressions" isn't that much on Facebook. From my experience getting 0.1% of viewers to care would be a significant number.

Being familiar with online marketing makes a lot of this sound less scary.

What I'd love to see is the Russian ad spending in the great context of the entire campaign. Considering both sides spent $1 billion on their combined campaigns it's entirely possible that the ~200 PR stories and 10M ad impressions are a minor blip in the wider scale.

What's interesting is how many people voluntarily shared these posts because it struck a chord with them (although it's equally as easy to buy fake likes/shares). And the fact they were focusing on critical swing states that Hillary's massive campaign failed to hit, basically non-english foreigners outperforming the most expensive American consultants...

The leaked data is another story and Wikileaks will never be proven, which means 50% of the leaked data is very likely via Russia.

For each of these, there were plenty of moving pieces out of Russia's control (the FBI and media's handling of pretty insignificant stuff, the highly receptive audience sharing the propaganda, etc) that all worked in their favour. Even if their contributions were minor, the US political environment played a huge role in amplifying it into something far bigger than they could ever do themselves.

Outside of some future smoking gun connection with the Trump campaign (which seems highly unlikely so far) it's going to be very difficult to measure exactly how much meaningful influence Russia really had on the elections. But it's an interesting lesson for the future regardless and the vagueness will offer plenty of leeway for the Clinton's campaign to sidestep responsibility for both running a bad campaign and for being a generally unlikable person (which matters more in these popularity contests than capability).

And if you didn't notice my entire original comment was satirizing mainstream discourse. I don't think you need or should trust intel agencies nor the media's uncritical interpretation here. I left it purposefully vague for those smart enough to see through the popular narratives.

[nl]

Measurement of effectiveness is pretty hard, I agree. And as I point out elsewhere, the Russians promoted diversive left-wing causes as well.

But I wish people would stop making claims that it didn't happen just because they (reasonably!) doubt the intel agencies. This stuff is trivially verifiable by anyone with the desire to do so.

[spaceseaman]

Wow I wasn't aware this was so directly available.

Thanks!

[nl]

No problem. It's a bit sad how rarely people bother to look into this.

People spend hours arguing on HN about this, and it takes a lot less time to take a look at the data.

You don't even have to be anti-Trump to agree with what the data shows: That set I linked to had strongly anti-Trump groups like "Blacktivists", "United Muslims of America" and "LGBT United".