I was going to write a few more of my annoyances with Twitter today, but a few pleasant announcements have come out of Twitter lately that I thought I’d share.
First, just announced today, Twitter will be shutting down accounts entirely that they have marked as “spam” on their blacklist. What??? Twitter has a blacklist? Yes, you heard it right.
Before today, Twitter would mark accounts as “spam”, but not tell the owners of the accounts they marked them as spam. Those owners of the accounts could follow others, but no one was able to follow them, and there was no way for the owners of those accounts to know they had been blacklisted. It was the same with the API – it was actually a huge headache for me as a Twitter API developer because despite me following the user, Twitter would respond with nothing as though the follow went through, but the next day the user would show up again as not being followed. I checked with Twitter and they confirmed the accounts I was having trouble with were indeed accounts that were on their “blacklist”.
Today, Alex Payne confirmed on the Twitter Dev mailing list that from now on users marked as Spam on Twitter will have their accounts suspended entirely for violating the Terms of Service. He also confirmed that they would avoid the headaches Facebook has had (aka Scoble’s account being suspended) and contact the owners beforehand to let them state their case.
I strongly welcome this new practice – Twitter has all the stats on their end to flag accounts as possible spammers, and by removing their accounts promptly this will make Twitter a much nicer place to use. The only other thing I would suggest they add now that this is in place is a way for developers to notify them of possible Spam accounts. You can do this as a user by “blocking” the account in question – I have no way to do this as a developer that I’m aware of.
The second, very welcomed feature is the addition of a “since” variable when requesting the friends or followers of an individual. In the past on SocialToo.com I had to do a request on all of an individual’s friends and all of an individual’s followers, and do a compare to see which of those following you, you weren’t following. For several A-list bloggers with near tens of thousands of followers, this was taking a few hours to complete the entire script, not to mention increasing the number of HTTP requests back to Twitter’s servers, I’m sure further burdening their already overburdened bandwidth.
Now, with the “since” variable, I can keep a tally of when I last checked your following/follower ratio and only do a request on those friends that have followed you since the last check. I can now go from checking those you follow on a daily basis, to checking almost every minute!
Last, Twitter has added the ability to determine, with one API call if a friendship exists between two individuals. Now rather than having to get a list of those you’re following, and then those that have followed you, I can just get a list of those following you and check to see that a friendship exists.
It is very refreshing to see such an active effort on Twitter’s part to help out the development community. I hope they continue to maintain the relationship they have with us, and continue to listen – many of us really want to see them succeed.