This image was created with the OpenAI Dall·e generator, using the prompt "an abstract fortune-telling twitter bot in matrix green"
botnadamus is a twitter bot that predicts your future 🔮
This is how @botnadamus presents himself on Twitter :
👋 Hello. I’m a bot. A bot that knows your future. What is it you wanna know ? Formulate your question precisely in your head. Meditate on your question. Is it the right one ? Once your question is clear enough, ask it. Don’t forget the @botnadamus mention.
You can test it here !
Under the robotic hood is a simple program that performs the following instructions :
- Look for new mentions on his timeline (@botnadamus)
- If new mention : fetches a random gif from Giphy / posts it in reply to the orginal tweet.
- If no new mention : sleeps and starts back.
As you can see, the program is amazingly simple. Yet, the results are powerful. Because humans always seek to generate meaning. Even the simplest random messages are interpreted as a sign of some powerful symbolic divinity.
When I asked botnadamus "what am I doing wrong", on the very day I managed to resolve all issues and finally make him work autonomously on a Heroku Dyno, he gave me this answer :
botnadamus_1.mp4
This is how our brains work : it secretes meaning. The intelligence we perceive in machines and progams is a transfer of our own power of imagination.
Go ahead, ask botnadamus anything. And try to pretend you didn't make sense with the pure randomness of his answer.
The main program uses the following services :
- Twitter API : Twitter API provides you with all the necessary tools and credentials to authenticate your bot.
- Tweepy : The Tweepy library gives you the Python syntax used to acess the Twitter API (note : I mainly used the mentions_timeline method).
- Giphy API : Giphy API provides you with all the necessary tools and credentials to communicate with the Giphy powerful platform.
- Heroku : Heroku is a platform as a service (PaaS) that enables developers to build, run, and operate applications entirely in the cloud. (note : the have a free pricing plan called "Free and Hobby", which I used for this bot).
The main building blocks to create a Twitter bot were taken from this tutorial : https://realpython.com/twitter-bot-python-tweepy/. The step by step guide for deploying a Twitter bot on Heroku comes from here : https://dev.to/emcain/how-to-set-up-a-twitter-bot-with-python-and-heroku-1n39. Note : I skipped the steps involving the creation of a Flask web server, and used the Heroku Redis option instead.
Botnadamus is an art project. It's not optimized for large-scale development. Sometimes, Heroku are doing maintenance on their servers, and it happens that Botnadamus retweets answers he already posted on your solicitations. It's not a big deal I suppose. It will only get you more Twitter activity 😉. But thanks for the understanding. I will find a way to optimize this issue in the future.