Slashdot Mirror
The FBI Got Its Hands on Data That Twitter Wouldn't Give the CIA (theverge.com)
The FBI is using a tool called Dataminr to track criminals and terrorist groups on Twitter, according to documents spotted by The Verge. In a contract document, the agency says Dataminr's Advanced Alerting Tool allows it "to search the complete Twitter firehose, in near real-time, using customizable filters." However, the practice seems to violate Twitter's developer agreement, which prohibits the use of its data feed for surveillance or spying purposes. From the report:"Twitter is used extensively by terrorist organizations and other criminals to communicate, recruit, and raise funds for illegal activity," the FBI wrote in a contracting document. "With increased use of Twitter by subjects of FBI investigations, it is critical to obtain a service which will allow the FBI to identify relevant information from Twitter in a timely fashion." [...] Earlier this year, Twitter revoked API access to a tool called Geofeedia, citing the same clause in the Developer agreement, after a reports showed the tool had been used by police to target protestors in Baltimore. Facebook was also a Geofeedia customer, and used it to catch an intruder in Mark Zuckerberg's office. This isn't the first time Dataminr has run up against Twitter's anti-surveillance clause. In May, Twitter revoked CIA access to Dataminr, a move that was taken as part of a larger ban on US intelligence agencies using the product.
It's worth noting that the two organizations have different legal restraints. So it is possible for the FBI to have access legally via a court warrant which the CIA, not being a law enforcement agency couldn't get. This also indicates that illegal cooperation between agencies is a serious risk.
I think part of it is that dictionary bots have already reserved any normal sort of name you can conceive of. So you can get tumblr.com for cheaper than tumbler.com and so on.
Unless you sincerely advocate for abolition of the CIA and FBI, you should help them do their jobs.
Yep, no middle ground between either helping them do their jobs or abolishing them completely. I know that we have a lot of programmers here but remember that most of life is analog, not binary.
The data is too sensitive and the potential for privacy violation is too high for the FBI or CIA to be given access, but it's perfectly fine for hedge funds, advertisers, newspapers, and, well, any other asshole with a checkbook?
Really?
Do you also think that companies outside the US should provide data about their customers, including Americans, directly to non-US intelligence agencies, or do you think only US companies should do that for US intelligence agencies?
I'm asking because one reason to not have special "Orwellian" data links from private companies to intelligence agencies within the US would be that the US should lead by giving a good example, so e.g. US authorities can argue convincingly to other countries that they shouldn't collect massive data on US citizens but only on select targets under reasonable suspicion. I'm worried that by giving intelligence agencies too much leeway, the US could gamble away more of its moral authority than it already has for only moderate gains. (Of course, Russia has recently forced companies to install such links under a new law, but maybe they aren't giving a good example anyway and Russia is way further outside the sphere of US influence than other countries.)
Talk about a tempest in a teapot - - - Twitter is so totally open, with it's datastream virtually wide open, that this issue is basically just a 'news bleep' that just isn't news.
Give me a break - so the CIA (or whomever) can access the 'firehose' at Twitter - well, so can just about anybody else.
Besides, Twitter not only gives access to this data, they also data-mine the stream for advertising and sales purposes.
ANYBODY using Twitter (or any other 'social media') that expects any kind of effective data security is so totally out of touch with reality that there is just no realistic communication with them. With the openness of the social media craze, I have very little sympathy for any of that crowd that gets hammered with loss of password / name / credit card data / etc from breaches in the social media's systems - since the social media orgs are basically setting themselves up as targets for any script kiddie that wants to 'give it a whirl' using any of the vast number of hacking / cracking tools available off the web. When you get to the level of state sponsored intruders, the social media orgs are just plain old 'low hanging fruit'.
redneck geek
"The FBI signed a contract" is not the same as "The FBI got the data". You must be new to government contracting.
There are places in the world, where the government is evil and remains in power not thanks to the sincere will of the governed, but by force of arms and other coercion. In such countries law enforcement organizations (such as the already-mentioned KGB) are used for oppression and should, indeed, be abolished. No cooperation with them is ethical. Even there, however, cooperation with institutions charged only with investigations of real crimes (murders, theft, rape) would be Ok — if it were possible to discern these institutions from each other, which is a separate topic...
The US is not (yet?) such a country in my humble opinion. But, if Twitter's management disagrees, then they should, indeed, call for abolition of the FBI and/or CIA to be self-consistent.
Orwell's Big Brother watched people in private. We are talking here about public data — stuff people willingly and enthusiastically post on Twitter.
How is our "moral authority" threatened in this case? I'm not saying, FBI should be able to compel firms into cooperation. My point is, it is foolish for the companies to deliberately cripple authorities whose work they do not oppose in general.
In Soviet Washington the swamp drains you.
Do you also think that companies outside the US should provide data about their customers
There are places in the world, where the government is evil and remains in power not thanks to the sincere will of the governed, but by force of arms and other coercion.
How about Germany or Italy then? Would you mind if they give your data away with free technical aid and without court order to secret German and Italian data collection authorities? Where do you draw the line?
Orwell's Big Brother watched people in private. We are talking here about public data — stuff people willingly and enthusiastically post on Twitter.
Maybe the position of Twitter and similar companies is that a company should not provide bulk data collection to intelligence agencies unless compelled to do so by law. That's consistent with thinking that the CIA is important for the US and maybe even should be supported, and even consistent with providing such bulk data links to other authorities (or advertisers, say). Bear in mind that automatic mass surveillance != publicly available data, as anybody can confirm who puts a robot.txt on his website with the hope that it will be respected. The fact that some data is publicly available does not necessarily justify that it should be used for mass data collection and heuristic evaluation.
German and Italian governments today are alright. Though I'm appalled by their position on the so called "hate speech", it is still Ok in my opinion to cooperate with their police.
No "maybes" here — it obviously is their position. And I argue, that it is a foolish one.
Whether it should may be for the FBI, CIA, et all to decide. But it is certainly not unethical for the companies to cooperate in easing police access to the already public data.
The test is very simple: could millions of police officers watching millions of Twitter feeds in their browsers and taking notes legally collect the data? Yes. Therefore, there is nothing wrong in automating the same job. The same test applies to license-plate readers and other creative use of video-cameras, BTW — sad, but true.
In Soviet Washington the swamp drains you.
and they just bought it from a company who is selling them and fbi is stupid enough to pay for it.
getting "realtime analysis" isn't all that great that it's meant to be.
also, twitter is kinda weird - providing developer access to firehose and then .. well. damn, the only use of it is realtime analytics AND REALTIME ANALYTICS IS SURVEILLANCE GODDMAADNFDSAF.
you ever wondered how twitter managed to be so big without managing to make a cent of profit while their service logic is 100x simpler than facebooks is? because it was always headed by idiots.
world was created 5 seconds before this post as it is.