Slashdot Mirror
US Secret Service Wants To Identify Snark
beschra (1424727) writes "From the article: 'The U.S. Secret Service is seeking software that can identify top influencers and trending sets of social media data, allowing the agency to monitor these streams in real-time — and sift through the sarcasm. "We are not currently aware of any automated technology that could do that (detect sarcasm). No one is considered a leader in that,'" Jamie Martin, a data acquisition engineer at Sioux Falls, SD based Bright Planet, told CBS News.'
Why not just force Twitter to change TOS to require sarcasm tag?"
Why not just force Twitter to change TOS to require sarcasm tag?"
that's gonna work
Search the text for /sarcasm or #sarcasm.
Done. Where's my paycheque?
They hunted till darkness came on, but they found
Not a button, or feather, or mark,
By which they could tell that they stood on the ground
Where the Baker had met with the Snark.
In the midst of the word he was trying to say,
In the midst of his laughter and glee,
He had softly and suddenly vanished away—
For the Snark was a Boojum, you see.
This will TOTALLY work. Detecting sarcasm will be DEFINITELY not be hard to do in software, seeing as how it's SO EASY for humans to do already.
that's a real useful invention.
Ha ha CBG, Frink gets the last laugh!
With all the sarcasm, don't you all realize you're creating the perfect dataset for them to train their algorithms on?
"[Regarding the 'cloud,'] ownership was what made America different than Russia." -- Woz
A significant part of their job is deciding whether threats made against said VIPs are serious or not.
How can I believe you when you tell me what I don't want to hear?
Sorry, you want the dismissive banter department. This is sarcasm.
Post may contain irony: discontinue use if experiencing mood swings, nausea or elevated blood pressure.
Just feed a bayesian analysis with postings from /. and use that to match against twitter. Should be able to get within 10%...
Some conspiracy insight - the government's quantum based AI big brother master computer of the NSA is up and running in order to completely surveil the internet in real time.
The problem is, it is nearly useless because it can't determine if someone is being sarcastic / snarky or not.
Flappinbooger isn't my real name
If they required a sarcasm tag, I'd put it on everything just to be safe. Or would I?
For all intensive purposes, "whom" is no longer a word. That begs the question, "who cares"?
Because people would instantly start using them sarcastically.
[sarcasm]It's like ten thousand spoons when all you need is a knife[/sarcasm]
The automated detection of sarcasm and derision will be one of the fastest growing segments in the new economy.
Already at least 3 startups have begun with this included in their mission statement, along with a stated goal of relieving venture capitalists from the burden of their investment money.
It is of vital national importance that we identify who is merely being a dismissive arrogant douchebag, and who is at risk of inadvertently hurting someone else's feelings so we send them for re-education and thought alignment. It will also help to identify people who haven't yet fully swallowed the kool-aid and don't believe that the government is, in fact, here to help us, uphold the Constitution, and defend our rights.
The US Secret Service is going to aggressively fund a second mandate to decode the mysteries of eye rolling, sneering, and derisive snorts.
This should further embolden the usage of widespread warrantless collection of our personal information with the knowledge that law enforcement agencies will be able to accurately detect sarcasm and redirect scarce resources to fondling young children and old people in airport security lines.
A spokesperson for the Treasury Department indicated that popular internet forum Slashdot, as well as Digg and 4Chan will be used as exemplars for this technology, as these have been identified as the single largest sources of snark on the interwebs since Al Gore invented it.
It has been further reported that Hamid Karzai of Afghanistan has said he welcomes this new initiative in the spirit of cooperation between the two countries, and that Kim Jon Un is hoping this will lead to a normalization of relations as it will allow the US to realize that North Korea was only kidding.
Lost at C:>. Found at C.
Or, why not just allow free speech? Why do we have to identify sarcasm? Maybe part of the expression of the message is its ambiguity.
Automated propaganda. Sheep herding shouldn't have to be work.
Well, as much as I don't like to defend it ...
Imagine that the Secret Service is, oh, I don't know, responsible for assessing threats to various people.
Now, some random internet loon says "grrr, I'm so angry I want to air drop a million pounds of used condoms, Snickers wrappers and Depends undergarments onto Capitol Hill in protest" -- now, you have two possibilities:
1) The random internet loon is blowing off steam, but in no way has either the desire or the resources to actually do this.
2) The random internet loon has just made a tangible threat.
And, since we've already seen stories about people who get arrested for making what they thought was an obviously flippant remark, this becomes a problem.
Of course, as with any automated system ... the false positives and false negatives will be what really kills it.
Because you will have instances where someone makes a genuine threat, and it is flagged as sarcasm. And, you will also have cases where what should be clearly interpreted as sarcasm will be interpreted as real.
So, then you either get actual attacks happening nobody took seriously. Or the men in dark sunglasses hauling you off in the night for questioning because they're 100% convinced that your threat to drop the condoms, Snickers wrappers, and Depends on Capitol Hill was real.
Lost at C:>. Found at C.
Nobody seems to be asking WHY.
Why would the Secret Service, in particular, want to tell sarcasm apart from other speech? Think about who they are.
They want to be able to distinguish sarcastic political speech, from sincere political speech. Of course both are protected speech.
Now, they might have a benign purpose, but from the description in TFA it doesn't seem so. After all, the administration would look pretty foolish if they tried to harass or jail someone for being sarcastic.
Oh, that's a really useful invention.
http://m.youtube.com/watch?v=m...
Dear Slashdot: next time you want to mess with the site, add a rich-text editor for comments.
Which scenario do you think is more likely? Furthermore, if anyone is a real threat, there will be much more intelligence (as in evidence of a threat) surrounding that individual than their tweets. Arresting people based solely on their tweets or FB posts will very rapidly devolve into an outright ban of saying anything critical of government officials or policy -- AKA fascism.
sig: sauer
Comment removed based on user account deletion
The reason the Secret Service wants sarcasm detection is because of the bad PR they get every time they harass someone for being sarcastic. The problem is not sarcastic political speech vs sincere political speech; it's sarcastic threats vs sincere threats.
Support Right To Repair Legislation.
Well ... gee ... let me think.
This kid:
It's a real thing, it has happened already. No evidence of a crime (or even the actual intent to commit one). But someone sees it and goes "eep", and then you get dragged off to jail.
You seem to be under the impression this isn't happening already.
It is.
So, ask me again if I think what I said is a plausible scenario. Because I said it with the full knowledge it has already happened.
Lost at C:>. Found at C.
Actually, what they could do would be to correlate English language postings with equivalent German language postings. As Germans are known world wide for their fun-loving sense of humor and sarcastic wits, the difference should yield accurate, non-snarky posts in English.
John
If you have nothing to hide, you have nothing to fear. /sarcasm
I am certainly glad to see my tax dollars spent on worthy high tech and expensive efforts by the Secret Service and NSA. They should continue to spend with no concern for rationality. If you don't want to pay $100 billion a year to watch what people are doing you clearly have something to hide. The Tea Party should also continue hate big government, but also support unlimited spending on anti-terror efforts. #snark
Conspiracy insight level 2: The government already has a sentient AI, but it's an Aspie like Data on Star Trek, so they're crowd-sourcing its emotional development. *dun-dun-DUN*
your thin skin doesn't make me a troll
No, that wasn't sarcasm, that was brilliant.
Also, you need to look up 'sarcasm.'
So, do you not think it is a patriots duty to overthrow a tyrannical government?
If so, then the question is, how much Tyranny is too much Tyranny? Not that we shouldn't act, but when should we.
But if not, then that means you support Tyranny, making you part of the problem.
Agent K: A *person* is smart. People are dumb, stupid, panicky animals, and you know it.
You keep using that word. I don't think that word means what you think it means.
Faster! Faster! Faster would be better!
so they're crowd-sourcing its emotional development. *dun-dun-DUN*
...and they're seeding it with all the data from Twitter, Reddit, Slashdot, and 4chan. So this is how the world ends
If you haven't been paying attention lately, having any kind of civil servant in the loop to determine if a threat is credible or not is also going to have significant problems.
Keywords: Swatting, ATHF Movie/Boston, NSA/FISA Court, Hello Kitty "bubble gun" and/or pop-tarts, etc.
Have gnu, will travel.
What they are doing is trying to hunt down and substantiate those quietly influencing the whole of the internet, not accidentally but purposefully via memes http://en.wikipedia.org/wiki/M... hidden in sarcasm, those being political and social memes not kitten memes. So what they will be doing is backtracking ideas and changes to existing ideas from larger more spread elements of internet media and attempting to track them back to original sources based upon date of occurrence and looking for repeated patterns. Then they will endeavour to silence or force influence over those quiet originators via one method or another. Likely they already have some suspects in mind from around the world and just want to confirm for action whilst expanding their search for others. The will also be looking be looking for collusion between originators to try to prove conspiratorial associations, this is likely where their ideas of substantiation will collapse as that consensus of thought between originators is 'Anonymous' to each other or of such loose circumstantial nature as to be legally negligible. Elements of government are basically pissed off at losing their propaganda war on the internet and looking for people to blame.
Chaos - everything, everywhere, everywhen
To be more precise, the quantum computer can determine if a comment is snarky or who made it, but not both. That's why they can't find out who's being snarky.
"When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes