It’s not the first time that Dick Costolo, CEO of Twitter, has made this promise. During a conference in September he stated that before the end of the year we would be able to download our entire tweet history on the social network in just one file. Now, he has mentioned it again, assuring us that we will have access to said service in the coming weeks.
A priori it would seem that such a thing wouldn’t lead to a lot of problems for the engineers, but upon learning a bit more about the internal workings of the system, it seems that the delay has all the excuses in the world. Keep in mind that today Twitter is used by more than 500 million users sending 250 million tweets per day. Without diving deeper, a few days ago we mentioned how during the U.S. presidential elections 300,000 tweets related to the elections were sent per minute, reaching a total of 30 million tweets.
Given this volume of information, it is normal for Twitter to be designed as a social network centered on offering real-time content, which is why it is so hard to find messages that you published months ago if they didn’t have a big impact. The system is designed as a service for real-time distribution and search, not as a data base of historical content, which is why its internal processes give preference to more recently updated tweets than previous ones to ensure that all those millions of messages per second are offered and managed correctly.
If the promise finally is fulfilled, we will be able to store all our tweets locally, without having to worry that they’ll be lost among the boundless amount of content that is published each day on the successful network. Even so, similar services to that of Twitter already existed that allowed you to download your tweets, although it wasn’t directly or in their entirety. Without diving too deep, Twournal is a website that can send you both a PDF and a physical book with your history on the network from 2009, and TwimeMachine can show you directly the latest 3200 messages on your wall.
