Hacker News new | comments | ask | show | jobs | submit login
Yahoo blocked emails about Wall Street protest (cnet.com)
180 points by danso on Sept 21, 2011 | hide | past | web | favorite | 49 comments

I'm not saying this was blatant censorship; but the overwhelming consensus in this thread of 'this is obviously a case of dozens of people hitting `mark as spam`' or some other coincidence is really surprising.

Twitter blocked #occupywallst from trending; people switched to #takewallstreet and it was trending again.

I don't see any really good reason to think of yahoo or internet in the USA as some bastion of freedom; especially when it can be 'blocked' by such murky means as a spam filter (plausible deniability allows people to act out just a bit more).

Our government has shown again and again that it is willing to tap all our means of communication (to the point of potential constitutional infringement) and mega corps have been generally willing participants.

To dismiss some phrase or website being manually dropped into a spam filter as an outrageous conspiracy theory makes me wonder if I am insane; because it's my perception these mega corps, banks, and government agencies will go to whatever extreme they can get away with; rather than up to some arbitrary point of what is 'right or wrong'; and this certainly seems like something they can get away with.

Nobody seems to have picked up on the "external" spam filter part of the response.

My guess is that Y! uses some somewhat dodgy external providers of spam/content filtering. Historically, external providers that block inappropriate sites have considered such lists to be their proprietary information, but when those lists have been revealed, they tend to have a pretty heavy political agenda.

Peacefire has some reports that detail this:


(So, I think the reason for the Y! block was likely an unaccountable third-party, but I think the Twitter block is pretty odd.)

I'd be a little surprised if Yahoo uses third parties for spam filtering. The Twitter thing is no mystery, though; Twitter routinely hides controversial hashtags.

Not to put words in your mouth, but I am asking myself this question:

Yahoo! and Twitter care enough about between 200-5,000 protestors marching on Wall Street to commit acts of censorship? Why? What does this accomplish for these companies? They're protecting the tight-knit oligarchy which they are a part of?

I'm struggling to see why this would even be on Yahoo!'s radar.

It's not Yahoo's radar. It's individual(s) who work for Yahoo that have their fingers on the censorship button and also happen to be part of a tight-knit oligarchy.

Some New York government official phones up his old Harvard buddy at Yahoo and asks him for help in avoiding civil unrest and an outbreak of violence. In exchange he promises undying love and future favors. The Yahoo guy realizes he has plausible deniability and so is happy to indulge.

How do you climb the status ladder in America? You do favors for powerful people.

Who wants to avoid the scary possibility of a massive viral wall street protest that inspires the nation to rise up? Lots of powerful people.

Powerful people are by definition capable of getting shit done and influencing others. This is why the cookie crumbles in their favor so often.

In order to have plausible deniability, the guy at Yahoo has to have direct, personal access to ban this domain. This makes him an ops person. Pretty much anyone else needs to go through a chain to get anything like this into production, denying them deniability.

If there's a conspiracy, it's an ops guy showing off to an oligarch friend. I don't see how any member of a thigh-knit oligarchy could get this done without leaving a trail that could get him fired for cause and publicly humiliated.

VP of whatever at Yahoo shoots an email to his pet nerd in ops and asks him to add a few domains to the spam filter. A list of 20 who personally spammed an angry customer.

VP of whatever says, "Nerdbro please add these domains to the filter before noon. An angry advertiser wants immediate action on this before closing a big deal. By the way I've recommended you to the division head for that position. Thanks for helping me out on this."

In the corporate world almost everything that happens has it's genesis in an exchange like this. If you need a citation: Robert B. Cialdini

Awhile later there is some PR consequence to this decision. Some PR guy in yahoo goes to ops and asks what happened.

Immediate answer: "they were mistakenly added to the spam filter." "how did thus mistake happen?" "its technical; and anyway I'm not authorized to reveal the workings of the filter because hackers/spammers want to know how it works. "

PR guy: "say no more. I understand. I'll make sure to let the public know it was an accident"

both of you make excellent points, and either could be true. But the sad fact is, we've had so many examples of government and big corporations behaving so badly that distrust has become systemic in our society. If Yahoo's actions were intentional they were probably unsure of exact amount of protesters but were taking preventative measures. Could have been a tech mistake, possibly. But coupled with the fact that twitter was also actively blocking the same topic, it seems highly suspicious.

Yahoo's spam filtering is terrible and I know for a fact that they maintain a list of URLs that are blocked even if they're just mentioned in the body.

Occam's Razor (not to mention Hanlon's Razor) suggest the issue is not deliberate.

I have a yahoo mail account and I can confirm that their filtering is terrible. About 1 in 500 e-mails or so never make it to either my inbox or spam box, without any error going to the sender.

Is Yahoo's spam filter really so quick to (automatically!) start filtering a very recent keyword (#occupywallst) based on just "dozens of people"?

This makes me think about an interesting attack vector ;-)

If you want to silence emails in email provider Y about subject X, you send email to many accounts of Y about X. Then you have the accounts report the messages as spam. Now every email to an account of Y about X goes into the spam folder...

Anyone know what the legality of that is? What would my recourse be if someone forged an e-mail from me that caused me damages? Is it just a civil matter, or would this be criminal?

Not to say that this is malicious, but if it was, they always make excuses like that.

"It was inadvertent, and we're working to fix the problem".

And by the time the problem is fixed, the damage has been done and the desired censorship in this case has been achieved.

The biggest indication this is not malicious is the benefit/risk ratio. If someone at Yahoo decides that talking about Wall st. protests should be silenced, they will, first of all, only silence a small fraction of the traffic - this is the no-single-point-of-failure nature of the Internet. Second, when they get caught (any subject large enough to be meaningful to censor will have enough tech-savvy protractors to detect this) they get hit by both the Streisand effect and the humiliation.

The very fact that the Streisand effect is a well known phenomenon suggests that people running large corporations can't always evaluate benefit/risk ratios very well.

The very fact Streisand is a person and not a company suggests people can't always evaluate benefit/risk ratios very well, regardless of the size and nature of that ratio.

I'd add that the benefit/risk is being evaluated by an individual or a group of individuals, as opposed to the company at large, so what might be insanely terrible for the greater good of the company can also provide a ton of benefit for a specific individual or a group. In that case, the individual or the group would act contrary to the benefit of the company.

It's the classic agency costs argument. I'm actually rather sad that we tend to jump on the company, forgetting that it's individuals that make these things happen.

Of course we can only observe the outcomes where the Streisand effect kicks in - who knows what the risk ratio actually is if one doesn't know the number of outcomes successfully suppressing info...

Not to say that this is malicious, but if it was, they always make excuses like that.

"It was inadvertent, and we're working to fix the problem".

The problem is, most of the time any IT-related issue rears its ugly head, the problem is caused by incompetence and not by malice.

Given the usual bugs, don't you think that a system like this would catch enough false positives that people would notice? Unless... they're censoring such complaints as well! ;-)

I would have sided with you before the news that showed US and EU companies are building net censorship tools for governments. They have been exposed in Syria, Libya and Tunisia but how could they not sell services to democratic governments too ? Saber-rattlers have been talking about cyber-warfare for several years now. Some of the millions spent on this buzzword was probably spent on equipments that can block messages with a deniable excuse.

But one doesn't need that type of infrastructure to block an URL. Yahoo has already spent millions building an infrastructure to block emails that look like spam.

It's "dual use" technology.

Yahoo was always blacklist happy in my experience of working on projects that send large volumes of email. This always impacted users of BT's online services too (btinternet etc) as they use Yahoo's mail servers. It was always a massive hassle to get emails through.

Which clearly indicates to me that the fault lies with consumers at-large. We, the proverbial and collective we, are far too forgiving of the mistakes committed in the provision of services on which we rely. Relatively few people will be spurred to action over this and will dismiss it out of convenience to themselves.

Also not to be overlooked, Yahoo knows that the pain involved in migrating one's email address is enough to retain all but the most idealistic of customers.

Sure, and if there really was an alien spacecraft detected at Roswell, don't you think the government would try to make it look like a harmless weather balloon?

You can justify pretty much any crazy theory with that sort of wonky logic.

If this happens in China, it would be immediately declared an obvious case of censorship and mass control. But it happened in the very "house of the freedom", so it is obviously a simple error and anyone thinking it was on purpose is a lunatic conspiracist. I read the comments of people that know nothing about (place the subject -antispamming algorithms-) but they are more than ready to defend (place the corporation or government -Yahoo-), and just an image comes to my mind. Sheeps. Blind sheeps.

In general, the country with less of a constant trend towards mass censorship gets the benefit of the doubt. And for all the United States' faults, they have a significantly better censorship record than China.

"Sheeps" is a bit shrill, but yeah, the double standard used on China and Iran in reporting from FreedomLand(tm) is a bit rich.

I expect this hypocrisy will eventually be dropped. Though people expecting the better of the two standards to win out are... rather optimistic.

Looks like a typical spam filter block. Go to your spam folder copy a url and send an email containing it and you should get the same response.

There's the question of how the url got on the spam filter blacklist, but it could easily have been an automated process.

Plenty of startups have issues with their invites getting marked as spam. Doesn't mean it's a conspiracy.

Conspiracy theories make better headlines.

Agreed. A little too early to conclude conspiracy. Verizon uses Yahoo as their email backbone. Did anybody have a problem getting a similar message out via Verizon (fron an account like xxx@verizon.net)?

Care of Wikipedia :A Yahoo is a legendary being in the novel Gulliver's Travels (1726) by Jonathan Swift. Swift describes the Yahoos as vile and savage creatures, filthy and with unpleasant habits, resembling human beings far too closely for the liking of protagonist Lemuel Gulliver, who finds the calm and rational society of intelligent horses, the Houyhnhnms, far preferable. The Yahoos are primitive creatures obsessed with "pretty stones" they find by digging in mud, thus representing the distasteful materialism and ignorant elitism Swift encountered in Britain. Hence the term "yahoo" has come to mean "a crude, brutish or obscenely coarse person".[1]

Anyone remember the HBGary hack. After what came out about government and corporate dirty tricks during that episode is it so unreasonable to think that somewhere in the US a consulting service was working for some Wall Street billionaire to derail the protests.

They are responsible for the service that they provide, even if they farm some of it out to third parties.

Those conspiracy theories are actually hurting the credibility of the movement. Right now the protest simply does not have the critical mass to threaten anyone.

Hmm, I could see it as a spam filter which notice a really high volume going into one account and flag it as suspecious, but it would seem much more suspecious if the emails were going out of a new account at a very high rate.

Personally this seems fishy to me. Moreso since it is comming from Yahoo.

A spam filter should act when one account sends several similar mails to numberous accounts unrelated to the sender.

A spam filter should not act when a hot topic is `trending' -- being send around from multiple accounts mostly to accounts related to them (with existing history of communication).

A spam filter acting on mere global volume of a keyword is a prime example of solution more troublesome than the original problem itself.

Doesn't it seem possible (likely even?) that one or more overzealous "99ers" sent numerous similar mails to many accounts to try to spread their message?

I will not be surprised if this is by oversight. I am regular yahoo mail user and their spam filters suck. I frequently need to look into the trash and spam folders to check for wrongly filtered emails.

Yahoo has always been overzealous about spam. I can't email anybody on Yahoo because my entire hosting company is on their block list, and has been for years. So this doesn't surprise me in the least.

Overzealous and yet incompetent. Yahoo mail became unusable to me due to spam very quickly (I have a 4-letter Yahoo handle), and I'm not sure it's much better now (I don't look at my account there often). Whereas Gmail blocked spam pretty efficiently from the beginning.

I see this kind of statement all the time from people and its just not a fair test. If you want to do a real test of the spam rates you must use the accounts in the same way.

its a known fact that older accounts get more spam than newer ones. And who knows what services you have signed up before with your yahoo account that you didn't with your Gmail.

Fair enough.

But my Yahoo account spam filtering appears to get suffer roughly as many false positives as false negatives which actually makes it worthless.

Occam's razor: A few hundred people getting e-mails about an event hit the 'this is spam' button. How is this news?

At least this time the headline doesn't says "blocked" instead of "censored", but still, it's not like this was a willful malicious act. An automated system hit a (debatably) false positive. Happens every day. Isn't a headline.

Algorithmic censorship?

I bet it's just the machines' way of getting back at us for CAPTCHAs.

Inadvertent my ass. I hope Anonymous retaliates.

Yeah, who needs evidence? Vigilante justice every time!

And how exactly do you expect the evidence to appear? Will the FBI investigate Yahoo for this? ;)

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact