Twitter spam again

When I added a simple spam-fighting addition to Dr. Twoot a couple of months ago, a couple of people wondered how effective it would be. So did I. It has just a couple of rules:

  1. A tweet is identified as spam if it’s from someone I don’t follow and it contains just my handle and a link.
  2. The spammer is blocked and reported.

So far it’s been quite effective. All but a couple of the Twitter spams I’ve received since then have been caught and I’ve had only one false positive, whom I unblocked almost immediately. It’s not like I get huge amounts of Twitter spam—just a few a week, usually—but it’s nice to know that the only spam-fighting tools I have available are being deployed automatically and immediately.

I agree with Marco Arment that Twitter isn’t doing even the most elementary and obvious tests to identify and block spam. For example, here’s what Dr. Twoot caught this afternoon:

Spam caught by Dr. Twoot

And here’s what I saw when I went to the spammer’s Twitter page:

Spammer's Twitter home page

No followers, not following anyone, and 180 tweets with nothing but a username and a link. How can this character be anything but a spammer?

In fairness to Twitter, I tried to get to this same home page just a few minutes ago and the account had been suspended. What I don’t know, though, is whether the account would have been suspended if I (and perhaps some others) hadn’t reported it for spam.

Marco said in that post that he’s stopped reporting spammers because it isn’t worth his time to do so until Twitter starts being more aggressive on its own. My first instinct was to disagree, but then I realized that I’d added the antispam code to Dr. Twoot because I didn’t want to spend any more time blocking and reporting, either.

And as long as the great bulk of Twitter spam keeps to the current format, I won’t have to. Dr. Twoot will do it for me.