Splitting your Twitter personality

About a month ago, somebody suggested to me in a Buzz conversation that I set up a different Twitter accounts for my activity in Romanian, my native language. I must admit I flirted with that idea before, but I thought it would be complicated and always postponed it. This is a rather thorny issue with all Twitter users twitting in multiple languages; despite some attempts by Google to tackle the problem, the language barrier is still dividing Internet users worldwide. Even though some Twitter clients like Brizzly integrated translation of tweets and Twitter may do so in the web interface sometimes soon, I can’t rely on machine-based translation to do a decent job, especially with a ‘niche’ language like Romanian.

So I went ahead and created a new Twitter account (gmoga), in addition to my old one (EXDE601E) that will be (mostly) for English tweets. I also tweeted about it to notify my followers of the change, in case someone was interested to follow the new account (apparently they weren’t). The bigger problem however was migrating some of the people I followed, namely those that also tweet mostly in Romanian. I wanted to separate the two timelines as much as possible. Twitter is not very helpful when it comes to managing connections. You can’t sort or see if people are following you back, unless you check if you can DM them. Clearly, I needed an external application for that.

The solution I found was FriendOrFollow. Based on your user name and without requiring access, this service generates three lists: people you follow, but are not following back, “fans” and “friends”. You can sort these lists in a number of ways on the web or, better yet, you can export the information in CSV format and process it in a spreadsheet software like Excel or Google Spreadsheets! That’s what I did; then I added a column to the table to distinguish between people I would continue to follow from the first account, people I will move to the new one and people I will unfollow for good. The user names are already in the table and with them I generated the URL for each profile page. Finally I opened up two Google Chrome windows, one in normal, one in incognito more, to log into both accounts simultaneously. From that point on it was only manual labor: copy URL from Excel, paste into both Chrome windows, unfollow left, follow right, start over. I’m sure there are other, more automated methods, to do this kind of migration, but for a small number of followed people this one is reasonably quick and accurate.

HootSuite is another client that can make migration subscriptions from one account to another less painful. After you connect HootSuite to your Twitter accounts, in the ‘People’ tab you have the option to follow people with one account and unfollow with another all from the same context menu. Unfortunately, like Twitter web, there is no way to separate people who follow you back from the rest or to sort the list alphabetically. So it’s easy to make mistakes, like skipping people or removing them from the wrong account.

Until now, I haven’t touched people who followed me back, but I think I will start removing some of them soon and follow them back from the new account. Either way, I’m sure FriendOrFollow will come in handy again. The exported data from Twitter includes the date when the account was created and when the status was last updated, so another possible use is to remove inactive accounts or potential spammers based on personal rules and with full control.