What’s Considered Spam On Twitter?

I have been using Twitter for a while now, and it is quite a useful tool. I don't know that it is an indispensable utility that I depend on, but it does help generate short conversations and has many other helpful uses.

With any new communication tool that pops up on the web nowadays, there are sure to be spammers and others who exploit the good intentions of others, which is why today's story at TechCrunch concerning spam on Twitter caught my eye.

Twitter has always seemed like a prime target for spammers, but in its early stages there was no concern because hardly anyone used the service. But now, as the number of Twitter users skyrockets and as the service begins to attract everyday web users (as opposed to the early adopters who have been using it) the issue seems more important.

My concern, however, is how does Twitter determine what is spam and what is not? I have discovered first hand that as I begin to follow more and more people, my stream of incoming messages has become more and more cluttered. Some users only post messages 1-2 times a days, while there are others who seem to update their fellow Twitter friends on just about every action they make during the course of a day. Is it the number of updates added from a particular user that causes Twitter to flag an account as spam? Or, do they actually take a look at the content of the message?

Twitter couldn't possibly be determining this solely the quantity of updates, because I am sure that the team wants people to use their service as much as possible because it is new and still growing. On the other hand, how would the team be able to check the content of every message that goes through the service? Or maybe they just use the unofficial Twitter Blacklist.

The official word from Twitter, according to the Twitter Development Group, is that the team has been "planning on simply removing the accounts of users who have violated our Terms of Service, as opposed to freezing their account as we've done in the past." The only problem here is that Twitter's ToS only states that users "must not create or submit unwanted email to any Twitter members." Since users must first 'follow' other users before being sent their updates, how can unwanted messages be sent? The receiving user has already confirmed that they were willing to accept any and all messages from the 'spamming' user if they decided to follow them. And if spam really is being sent, wouldn't the offending user have to unsubscribe from the spammer's updates for them to be considered 'unwanted.'

I am not trying to make Twitter's decision to remove spammers seem bad (it's actually a good move), but I think that this service is going to have to define 'spam' much more clearly and make those updates to the terms. Just as a definition for 'email spam' had to be determined and filters developed as everyone began getting email, the same will have to be done for 'twitter spam,' as these two communication tools have different uses and capabilities. What works for one will not necessarily hold true for the other.