Twitter is Dangerous: revista

I'm probably miles behind some tech complaint curve everyone else knows about, but having spent four hours trying to gain control of my Twitter account I'm at the point of pulling what's left of my hair out. Here's what's happened.
Written by Dennis Howlett, Contributor

I'm probably miles behind some tech complaint curve everyone else knows about, but having spent four hours trying to gain control of my Twitter account I'm at the point of pulling what's left of my hair out. Here's what's happened.

The last few days I noticed the number of people I am allegedly following rise dramatically. The reason I noticed was because I started to get a bunch of direct messages thanking me for following.  (see pic below)

Within a couple of days the number of these spurious follows had rocketed from some 1,300 to over 1,900 with around 300 added last night. I reckon I'm a reasonably sociable if curmudgeonly person but there is no way that I can realistically follow random people to that extent. Scoble's got the lock on that ability. That's why I regularly cull follows and only follow those where I can see a likely connection to my interests.

Without wishing to appear a total dope I asked whether anyone else has experienced the same problem. Lo and behold I am not alone - so not a total dope. (see pic below but there are other examples) The irony is that @paulthomas_tx is someone I've never heard of but I am following.

OK - so what to do?

My first plan was to lock down my account by going into 'protected Tweet' mode. That didn't help much because the 'follows' continued to pour in. I am presuming there's a backlog of messages coming through. (As I write this, the direct messages of thanks are still coming in.)

I then tried deleting the follows I don't want. That seemed to be working but then I find I am getting follow requests from the same people - or at least some of them. Delete those as well.

You'd think that would help but no. Next I find the follow/follower list/request continuing to grow. At this stage I am getting mighty annoyed but someone suggested changing passwords. Having done that three times - in part because I was being locked out of my Twitter account - I then find I can't login via the AIR clients I use: Seesmic Desktop or TweetDeck, nor can I get in via Nambu but I CAN get in via the native web client.

Someone else suggested blocking but I am loathe to do that to people who might otherwise be innocent and simply the victim of some 'stuff' of which they are unaware. Some candidates however were pretty obvious so they got the block treatment.

Next step - get help from Twitter. See below:

Apologies for the small image size but when I sent the original lockout query along with some details about the other problems, Twitter's autoresponder assumed I was talking about login problems and replied: "This problem is deemed solved." Puhleeeease. The lockout problem cleared so that's one issue out the way. It would have helped if Twitter provided a time estimate for lockout instead of it's 'hang on' message.

I have been successfully using Topify to filter Twitter messages coming in via email but it doesn't work as well in protected mode. The result is that my email inbox is being swamped with follow requests. That's easily solved with more filtering but waking up to 154 follower requests was not my idea of fun.

What's happened? I really don't know. People in my network made good suggestions for at least temporarily fixing the problem for which I am immensely grateful. But... as an enterprisey person with plenty of things to do other than clean out Twitter crap, this is unacceptable. It reminds me of the days when I used to clear up Windows infections. Twitter support has to improve.

Autoresponders are fine when they work, but when they don't they're a guaranteed way to elevate the grump levels. Locking out the account because of failed logins is fine but discriminating between services is not. It's confusing.

Operating a service that is drop dead easy is fine, but having issues that require tech support solutions that are undocumented or not immediately obvious is far from acceptable for the average person.

It has been suggested I might have picked up the 'Mikeyy' or 'StalkDaily' Twitter worms or some such. I have no idea but couldn't Twitter scan accounts for this, notify individually and then do a clean up? Robin Wauters at TechCrunch documented a fourth round of the worm infection and said that Twitter is advising. Unfortunately, the link he provides goes to a dead Twitter page.

Back in December 2007, Mike Krigsman declared Twitter is Dangerous. At the time he was thinking of other issues. I'm thinking that with Salesforce.com and others jumping on the Twitter train, the issues I have experienced for reasons I still cannot fathom renders Twitter even MORE dangerous. Questions:

  • Do I really want my business email junked up with spurious follows/follower requests? If not then what is Twitter prepared to do to provide users with protection against this kind of problem yet still allow the service to grow?
  • Does Twitter have coherent processes in place to help its users? This seems to have been an ongoing issue that as yet remains unresolved.
  • Is the notion of an open API that has no apparent requirement for destruction or malware testing acceptable? In the enterprise the answer is a solid no but with consumery apps creeping into the enterprise is it any surprise when IT stamps them out?
  • What are established services like Salesforce.com doing to ensure that Twitter users get what they want from the service but without this type of headache?
  • Has the time come for more attention to be paid to services like Yammer or ESME where use issues inside the firewall are being addressed but which will have to extend outwards at some point?
  • As Dick Hirsch asked: Am I ready to declare Twitter bankruptcy? Answer: I'm within a gnat's whisker.

Editorial standards