Skip to content

Twitter Censorship – Network or Message?

October 21, 2009

Is Twitter a network facilitator, or a participant in message and meaning? Should the environment be open or should it be sanitized, moderated and censored? I don’t know. I also haven’t seen anything formal or significant on policy or approach that provides clarity for the Twitter community.

So why is this an important issue?

Lets take an example. The #Iranelection hashtag (hashtag is a deliberately flagged search term, preceded with the hash character – ‘#’) was a trending topic on Twitter for many months. The topic and Twitter posts covered a host of nasty, sensitive and confronting material based around the election, election violence, democracy and media transparency during this period. The conversation still continues. Twitter was one of the few open outlets that allowed apparently unfettered comment to escape Iran, as traditional media and international visibility was shut down and journalists ejected. Twitter it seemed, let all of this content into the public arena, and there would be no doubt that certain forces were trying to prevent such material from entering certain geographies or to shut down and divert certain conversations completely. Twitter played an active role when they decided to delay a planned network maintenance outage, due to the importance of the dialogue and the volume of traffic. Perhaps no one sees this as an issue, but positive interference is still interference, and changing systematic behavior because of the nature of the prevailing message extends into territory that is much more than network provisioning. I’m not suggesting that it should have been handled any differently, just that it showed clear involvement with the message.

Twitter Intervention and Stewardship

Twitter ‘suggest’ people you should follow when you create a new account, or any time you go back into the ‘Find People’ function (under the section ‘Suggested Users’). Twitter regularly remove inter-account linkages in occasional clean-outs and also make decisions on the suspension of accounts. If you use Twitter, you have probably come across the ‘suspended account’ graphic. This is all generally accepted and welcomed maintenance and hygiene practice. Suspension for example, would normally be the result of a breach of the Twitter ‘terms of service’, including falsifying an account, obscene material and so on. Twitter suspends accounts, such as the publicized closure of the Dalai Lama account, because of misleading ‘spoofed’ operation but chooses to leave thousands of others, with lesser profile, open, despite their clear violation of the terms of service. Obscene content is also on a continuum and based around moral values as well as laws that vary considerably by state and country. There is plenty of pornographic, violent, adult and offensive content on Twitter (as there is online), so for better or worse, Twitter makes a determination. Clearly there is censorship on Twitter! This is not surprising, and it is welcomed intervention by the majority of users, probably even necessary for the effective functioning of the network. There are always going to be people on the ‘thank-you Twitter for getting rid of (insert topic, content, person or view here)‘ and those on the other side of the fence.

Guidelines and Policy

It would of be good however to know what Twitter’s policies and processes are. Is there review, escalation, appeal, audit and accountability? If the #Iranelection stream was instead focused on US domestic politics, a certain political party, or something ‘closer to home’, how would that have gone down? What happens to Twitter streams that relate directly to Twitter – the competitive environment, new social media plays, investment, dirty-linen and so on? How much pornography is too much? How are under-aged Tweeps (Twitter People) handled in comparison to adults (by the way, what is underage in the Twitter world, 16, 18, 21 – how do we know)? In Australia, Indigenous communities have very strict rules relating to the discussion of the recently deceased, known as ‘Sorrow’ material. What happens to the accounts, posts and comments relating to Tweeps who die? Maybe most of the time we don’t care and we expect and assume that Twitter (the people running the network) will act fairly, honorably and in so far as possible without heavy-handed cultural, national or other subjective bias. Can we really expect this? Is this what is happening?

Probably most Twitter users believe that the ‘trending topics’ are determined purely on some mathematical algorithm. Some manipulation is of course needed to group the same conversation where the #hashtag or keyword has identical or very similar meaning to others (which would divide the conversation). Should Twitter remove some inflammatory, sensitive or offensive ‘trending topics’, or weight them in some other message-aware fashion. When would this be appropriate, and when would it not?

Balance – The ‘No God’ / ‘Know God’ Example

On the date of this post, there was a trending topic – Know God. Somewhat earlier there was a trending topic – No God. It is no surprise that this philosophical battle would be waged on Twitter. Religion, spirituality, politics, and sex are stock-in-trade topics for discussion in many social networks. It would appear that ‘No God’ as a trending topic was either removed, or was merged within the ‘Know God’ trending topic. Both clearly have different and largely opposing meaning. The issues that arise are both practical (which trends are the same) and political (what should be manipulated). It would appear from other traffic trend analysis, and from simply watching the Twitter stream through windows such as Twitter Fall, that ‘No God’ should have remained in the trending topic list on a mathematical basis. I think it is fair to say that this is an area with strong opinion and it is unlikely that any 140-character message is going to solve the debate – I am certainly not pitching any view! Surely if Twitter are intervening with the removal of one trend, then they should remove the other counterpoint trend at the same time?

[I am not making an argument for either trend, so please, no hate mail or ‘un-following’. This is not a forum for comment on the No God/Know God debate and any such comment will not be published here. Comments on social media censorship, social media policy and intervention practice and related topics will be most welcome and published.]

My view, and hopefully the reasonable one, is that intervention is a necessary and largely unfortunate component of platforms such as Twitter. It must be a pain in the proverbial for Twitter itself to manage. The intervention should however be based on clear policy that is balanced, equitable and non-partisan. Can Twitter currently claim balance and how would the community know?

Please contribute to the conversation, make comment here, and post on the social media network we love – Tweet away!

8 Comments leave one →
  1. Andrew Mitchell permalink
    October 21, 2009 4:53 AM

    I was one of those particpating in the knowgod/nogod fun. You state:

    “It would appear that ‘No God’ as a trending topic was either removed, or was merged within the ‘Know God’ trending topic. Both clearly have different and largely opposing meaning. The issues that arise are both practical (which trends are the same) and political (what should be manipulated).”

    Yes, these are largely opposing sides but both streams had many tweets duplicated so it seems to make sense to merge them. It seems to me that the chosen combined name was a political decision to reduce offence to Christians and selection of a neutral name would have been better policy.

    Yes Twitter nees to learn from this. They need a clear and comrehensive policy and then they need to consistently act within that policy.

  2. Andrew permalink
    October 21, 2009 9:58 PM

    My 2 cents.
    Twitter is a computer program….it will pick the trending topics based on stats.
    But twitter also has a limitation – 140 characters.

    So maybe the programmers built “Slang/txt speak/etc” into the trending topic engine.
    Hence it know and no could be used within a tweet to mean the same thing (no being used for know) but the no used because it’s shorter and gives the extra 2 characters.

    So when you take out the words following or preceding the Know/no words. Maybe this should be tested?

    Maybe there are other words that this works on?

    • Andrew Mitchell permalink
      October 22, 2009 12:00 AM

      @Andrew: Maybe it was not clear in the article but the issue was that there were TWO trends “no god” and “know god” (with overlapping content) that were merged together.

  3. October 22, 2009 12:02 AM

    Andrew, I see the thinking. However why would both have been in the trending topics at the same time earlier in the day? I think you are looking for a ‘systematic’ reason, when the evidence clearly points to human intervention. I don’t think anyone is supporting the idea that intervention doesn’t or shouldn’t happen – the question is – on what basis? My post is really asking for clarity and transparency as well as even-handedness. Thanks for the comment.

  4. Andrew permalink
    October 22, 2009 12:23 AM

    Arr yes, my mistake.
    I thought one was there, and then the other came along, and sort of merged the stats or something of that nature.

    Thanks for clearing it up. :).

    I wonder what its like to work at Twitter HQ…and what you would do there? and what the employees think of this sort of thing?

  5. October 22, 2009 2:15 AM

    Would be very interesting. Would be great for them to give a ‘day in the life’ (as well as other vision, process and principles info). Would make a good story. Happy to do the video if anyone at Twitter wants an ‘interest piece’ done 😉

  6. Ron Jackson permalink
    November 12, 2009 4:52 PM

    You can easily and quickly find out who, what or which tweets are being censored by Twitter.

    To see WHO Twitter Is censoring – Search for the user’s name in Twitter Search. If that user only show ups in RTs, THEY ARE BEING CENSORED from Twitter Search and therefore from the Twitter Timeline.

    You can also use this same method to see if Twitter is censoring YOU: Tweet then look for yourself in Twitter Search. If your tweet (or you) can not be found, or If you (or that tweet) only show up in RTs, YOU ARE BEING CENSORED from Twitter Search and therefore from the Twitter Timeline.


  1. uberVU - social comments

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: