How did Scilab get the data for Twitter? (Because GDPR makes data crawling difficult nowadays as well)