Today Twitter’s CEO said they may in the future “edit out any…clearly offensive [trending topics].” He also said “we edit out any [trending topics] with obscenities.” Thanks to @rachelsklar for finding this tweet and @AbigailCollazo for pointing it out to me:
This is major news from Twitter.
Probably they have not already begun removing “clearly offensive” trends because they are harder to identify than those with obscenities. After all, you can make a list of obscenities that are not allowed, but it’s harder to list topics, because topics can be presented from different perspectives (for/against/nuetral) and wordings can vary.
Dick’s choice of words is interesting too: “I’d like to see…” He doesn’t elaborate whether this means Twitter plans to, or why what he’d like to see hasn’t begun happening. Twitter could also allow users to choose to filter trending topics, as opposed to outright censorship, although this would mean a change to their infrastructure.
Why Twitter needs to clear up their policy, and fast
The problem is: If Twitter does what the CEO wants, where will they draw the line? It already seems inappropriate that Twitter has not taken a clearer approach, e.g.
One of the reasons this is such major news is that Twitter allows accounts where all their tweets are about clearly offensive topics. Twitter’s terms of service doesn’t allow child porn, but it does allow a lot else. Many people have been frustrated in Twitter’s unwillingness to remove offensive accounts, but Twitter has always come down on the side of free speech, and against censorship.
Clearly Mr. Costolo feels that allowing offensive accounts is one thing, but amplifying offensive topics is another.
Twitter shouldn’t have policies that are not publicly stated.
They at minimum need to change their public policy page on trending topics to reflect what they are actually doing, such as by publishing a list of obscenities that are removed from trending topics.
Prior to this admission from the CEO, all Twitter policies said about obscenities was:
- You may not use obscene or pornographic images in either your profile picture or user background.
- Twitter does remove obscene or pornographic images in either user profile pictures or user backgrounds.
Do you think Twitter manipulated a trending topic?
If you do, you need to at minimum read Twitter’s explanation of how topics trend, and check the volume of other trending tweets before and after the time you think one is being suppressed. If you don’t know how tweets trend, and you don’t know how to compare the volume of one versus another, they you are, by definition, ignorant. I don’t mean this as a criticism, but you can’t know nothing about trends and have no data and expect to be taken seriously when you say you think Twitter is manipulating trending topics.
On the other hand, Twitter does in a manner of speaking manipulate trending topics over time by changes to their algorithm. As Twitter once explained to Mashable:
“Our Trends list is designed to help people discover the ‘most breaking’ breaking news from across the world, in real-time. The list is generated by an algorithm that identifies topics that are being talked about more right now than they were previously.
“There’s a number of factors that may come into play when seemingly popular terms don’t make the Trends list. Sometimes topics that are popular don’t break into the Trends list because the current velocity of conversation (volume of Tweets at a given moment) isn’t greater than in previous hours and days. Sometimes topics that are genuinely popular simply aren’t widespread enough to make the list of top Trends. And, on occasion, topics just aren’t as popular as people believe.”
Don’t believe everything you read
“Blocked” Trending Topics
Again, topics do NOT trend only because people a lot of people are tweeting about them. They trend because new people recently began tweeting about a topic in sufficient volume. But volume of tweets alone isn’t enough.
The key word is “recently.”
Many topics trend on Twitter, then disappear, even though they become more popular later. But Twitter’s algorithm considers it “old news” once it has been trending for awhile. If it didn’t, the trending topics would be overwhelmed with things people are talking about all the time (such as Justin Bieber).
The exception is if new people begin tweeting a lot about something.
The key is that something has to be “new” to be considered “trending.” Either new people, or a new topic. Having the same people tweeting about the same things won’t make a new trending topic, or keep an existing one going.
So popularity isn’t enough. To trend, a topic must be new, or an older topic tweeted by people who haven’t tweeted about it before. Otherwise the algorithm won’t place it on the trending topics list.
Frustratingly, some news topics trend and then disappear, even though new and important things are happening. But if the same words are used in tweets, Twitter’s trending algorithm considers the topic to be getting old (despite whatever may be happening with it in the real world) and unless new people begin tweeting about it, or new words are used, it will fall off the trending list.
Suspensions
I regularly see people saying Twitter has done something wrong, censored something or someone they “shouldn’t have.” I used to get involved, but now I just wait a few days for the inevitable correction to appear.
- Example: Guy who says he got an email from Twitter saying his Twitter account was suspended for “suspicious activity, including anti-government propaganda.”
- Inevitable Correction: Guy says he was affected by a Twitter bug, and fell for a fake email.
Most of what I’ve read about Twitter suspensions are incorrect. Most commonly from people who were smart enough to file a ticket with Twitter requesting information on the suspension, but haven’t heard back right away and decided to write an angry blog post speculating on the situation. But now, with Twitter’s CEO admitting they have an unpublished policy to censor trending topics, and that they may do more, raises the question of what other unpublished policies they may have.
But regarding suspensions, though I’ve never experienced it myself, I know Del Harvey, head of Twitter’s Trust and Safety team, has a lot of compassion for folks who get suspended. If you did something wrong, @delbius is very forgiving. Once you understand the problem and promise not to violate Twitter’s policies again, your account is almost always unsuspended. Some folks have even been suspended and unsuspended more than once. As I say, @delbius is very forgiving 
{ 32 comments… read them below or add one }
← Previous Comments
Well, the people need a place to talk what they want, without the censure.
BUT! when somebody , send a milion e mails with the same obcene content at once, than the Twitter’s contributor, he can see, there is not everything O.K.
Well, I am not sure, if my mail find a way to Twitter, will see….
Well, the people need a place to talk what they want withoth the censure.
BUT! when somebody , send a milion e mails with the same obcene content at once, than the Twitter’s contributor, he can see, there is not everything O.K.
Well, I am not sure, if my mail find a way to Twitter, will see….
So…Twitter is independent of its own algorithm? Twitter defines what the algorithm is, makes changes to it, it is part of their policy and therefore, if any ‘manipulating’ of the TT goes on, Twitter IS responsible for it. To say that ‘the algorithm did it’ is facile. It’s a cop-out.
Trending Topics is a means of directing conversations – once a topic falls off TT (doesn’t mean the hashtag is gone), conversation can and will dwindle. Trending Topics is easily polluted with spam. ‘Offensive’ matter is easily used to remove ANY topic from the TT. Twitter’s policies surrounding Trending Topics is, at the very least, just annoying, at its worst, the means to manipulate conversation (whether ‘intentional’ or not.)
In previous versions of Twitter, it was easy to ‘hide’ Trending Topics.’ In Twitter’s latest iteration, you cannot, it is placed front and center with no means to ignore it. This suggests how important TT is to Twitter itself – and anything that can be so easily manipulated and thus confuse others is NOT a good system, however you look at it.
Your point is completely valid, and yet there is a sliding scale of intention. What people complain about is Twitter’s intention to remove a specific trending topic. Twitter’s explanation of its algorithm is that it does NOT single out specific topics by their content, except in some cases of profanity. If they singled out topics by content, that would be the manipulation they refer to, regardless of whether it was done manually or by modifying the algorithm.
It’s true that #TT’s are very important to Twitter—it’s a place that ads appear, and Twitter is committed to getting the majority of their profit from ads. I’m definitely NOT saying that this is a good thing, but it’s understandable from Twitter’s point of view.
Can twitter not include a non-idiot Trending Topic option that filters all fan stuff out?
That would be a great idea!
I agree with John Smith. There are obviously some things that should never even be considered as search terms or hash-tags, etc. But language and expressions differ from country to country and what may be offensive in one country is perfectly acceptable in another. For example, the coloured in South Africa is totally acceptable and refers to a specific group of people. It is in no way considered offensive or destructive. Yet used in America or the United Kingdom, you are considered racist. Twitter and twitter users should be aware of peoples motives and national expressions before complaining or taking down a term.
“Twitter takes down requests all the time.”
Yes, Twitter’s algorithm does, all the time. “Twitter” does not. There are quite a few posts about this, lots of data. Take a look at, for example
• http://blog.socialflow.com/post/7120244374/data-reveals-that-occupying-twitter-trending-topics-is-harder-than-it-looks
• http://blog.bufferapp.com/five-twitter-secrets-about-censored-trending-topics
• http://blog.tweetsmarter.com/twitter-rules/how-i-uncovered-twitters-trending-topics-secrets/
No, the above angry Rachel knows you aren’t Twitter, she’s just mad that you are fronting for them. She’s given you a clear example of Twitter censoring, and you ignore it and joke about something else.
Twitter takes down requests all the time. The Patriot Act doesn’t even let our librarians tell us when the U.S. government has accessed our library cards, so Twitter is going to have a conscience? Google, “do no harm,” has released lots of email records of activists to many countries and gotten activists imprisoned too. They are companies and profit and survival is the bottom line.
The way oppression works in America is that activists are charged with criminal charges and politically imprisoned, thus, “no censorship!” and “no political prisoners!” Wikileaks was clearly exposing war crimes, but no, Twitter says it was merely a criminal act and it is censoring criminality. Of course, IF that was true they’d be a lot more transparent about what trending topics they’ve changed and when, but then the U.S. govt regulatory agencies would be turned loose on them.
No reason on earth why the name Troy Davis would be censored, its a name.
but the word nigga, niggers, bitches, hoes, sluts, and rape just effortlesssly become trending.
as a black woman what do you think you are protecting me from?
how dare you block anyones NAME but allow clearly offensive words to trend? make sense of this to me?
who looks at the word NIGGA or BITCH in your employ and says oh thats ok…
are you kidding me???
you realize that kids use twitter..so Troy Davis is offensive but its ok to let a 12 year old see obscene words but not learn about a historical case?
just get rid of trending topics and stop insulting the intelligence of BLACK people, this isnt the first time you’ve censored “black” hot topics, we SEE you.
Just for clarification, in case you’re not being rhetorical: we don’t work for Twitter. See http://bit.ly/WhoIsTweetSmarter for more info.
So why it’s troy davis blocked? It’s not offending anyone. This is a supporting trend. I don’t understand and I don’t think it’s right.
Do they censor Dick’s name?
If they use an “algorithm” as they said, then yes. “algorithm” my @ssssss. Censorship doesn’t end there when offensiveness is subjective. Hey, I’m offended by #youknowyoughetto reference. When does it stop or where?
they need to set up a filter for justin bieber.. his name is always trending … twitter is ran by his fans.
His fans are definitely TRYING to run Twitter!
Taking out a trending hashtag is not censorship. If someone wishes to speak about why they beat their girlfriend, let them do so without encouraging a massive amount of people to talk about the same thing.
While I did use “Editing offensive Trending Topics” and not “censorship” in the title, I’m still thinking about this one. I note that you don’t differentiate between censorship generally and content that you personally don’t want to see. It’s important to make the distinction. Encouraging the blocking of content is censorship. Encouraging the availability of filters so you don’t have to see content you don’t approve of is a different matter.
We are in the age of aggregation, and have been for some time. Even many years, ago, when the Associated Press put out an article, it appeared in newspapers around the world. If a newspaper decided not to run it, it wasn’t ever clearly censorship, but it could have been left out by censorship-type motives.
Nowadays, aggregation is automatic all across the web. Trending Topics is a form of automated information aggregation. Deciding to block some information that would otherwise appear automatically due to its content sounds very close to defacto censorship.
Well, thanks for lying about this, Twitter. That’s even worse than treating your users like children who need to be protected from all the naughty words out there.
Im brasilian too, and the episode about Ricardo Teixeira was REALLY suspicious
Ricardo Teixeira is known to be corrupt, powerfull and is frequently compared to a gangster
he has ties with brasilian biggest media corporation(it has interests on the transmission of soccer games), who is itself VERY powerfull too, and won’t air bad news about Teixeira
Meanwhile another media corporation is airing all investigation and accusation against teixeira, people that are not as alienated by Globo(the biggest one and Teixeira’s friend) plus sports fan are trying to create buzz about this, specially since the next world cup is gonna be here, people got #caiforaricardoteixeira(something like ‘get lost, ricardo teixeira’) on trending topics and twitter removed it, later alledging that ‘they thought it was spam’… I find it hard to believe
To be credible, you must provide a reference. Where did Twitter say they removed something because they thought it was spam? You can’t quote a person or company without providing a reference.
Not a fan of Twitter censoring any information. We’ve seen Twitter have an impact on political debates and even some revolutions. This can be a tool for good and censorship even in the name of goodness can be misused. I’ve recently had some success with gaining more followers through twitter.popularfans.com (editor’s note: service appears to violate Twitter’s Terms of Service) and it would be a shame to see Twitter’s desire to make money cause them to lose their focus because that would definitely cause me to reduce how I prioritize Twitter promotions for my blogs and stuff. No sense investing in a network that’s going to be censored in such an inconsistent way. There’s a few things I can understand Twitter doing in an effort to make some money, but censorship is one thing that definitely leaves a very bad feeling.
I’m of Brazil (please sorry my English) kkk
Last week the TT #foraricardoteixeira was excluded from the TT …
Twitter justify span, lie, lie, lie
obs: ricardo Teixeira is the president of CBF (confederação brasileira de futebol) local soccer confederation, and president of commite of world cup.
Rightfully, many people reacted strongly against the #reasonstobeatyourgirlfriend topic.
The interesting thing about this as a trending topic was actually that by far most of the tweets were talking about how offensive/ inappropriate / irresponsible the #tag topic, as well as the act described in that #tag, are…. thereby preserving it’s position as a trending topic for all the RIGHT reasons.
True!
As I tweeted on Sunday, when the offending topic was trending:
“Dear Halfwits, the only reason for a misogynistic trend is that you keep repeating it in your complaints. Go back to FB and forward hoaxes.”
People who are too dim to understand the obvious consequences of their actions are not doing anything ‘for all the right reasons’ they are simply dim.
Twitter is US.
US uses violence and fear to control masses.
Twitter already removed trends, usually those speaking against the government and those related to Wikileaks & Anonymous. However, reasonstobeatyourgirlfriend (although there are many) is violent and was up there for hours. It is how Twitter runs things, censorship in the name of the Government of the United States of America. They have as little conscience as those who kill people in the name of the law or massacre Palestinians because they do not believe in the same version of a story as them. The difference, Twitter has no access to bombs and guns, and trust me, that is a good thing.
Testing my personal rules about censorship, I see…
Partly =)
Quite frankly any censorship is wrong. If some people find a word, topic, or trend objectionable let them use parental filters. I am certain a market is out there for this. Political issues and many other should not be censored period. Please keep the forum free and open.
John
I agree, filtering that users can opt-in to is really the best approach. Twitter’s infrastructure is just not yet up to the job, I’m afraid. Hopefully one day it will be.
Problem is one of your follows re-tweets something offensive (filthy language) from someone not following you. I’m for free speech. But let’s face it, everybody knows when something is offensive. On the web so many take advantage of hiding behind their screens and saying things they would never dare to your face. Some kind of solution has to be implemented.
{ 32 trackbacks }