The Football Association has sent a video to Fifa in a bid to prove that Dele Alli's middle-finger gesture was aimed at England team-mate Kyle Walker and not an official or opponent. Television pictures showed the gesture during England's 2-1 World Cup qualifying win over Slovakia last week. World football's governing body opened proceedings against Alli, who said it was a "joke with good friend" Walker. An FA spokesperson said it had responded to Fifa on Wednesday. The spokesperson added: "We understand the case will be reviewed by the disciplinary committee before the end of the month."

England midfielder Dele Alli says his middle-finger gesture during the World Cup qualifying win over Slovakia was aimed at team-mate Kyle Walker. Television pictures showed Tottenham's Alli gesturing in the 77th minute of Monday's 2-1 victory at Wembley. The incident occurred after he was body-checked by Martin Skrtel, and referee Clement Turpin waved play on. "Just to clarify, the gesture was a joke between me and my good friend Kyle Walker," Alli, 21, wrote on Twitter. "Apologies for any offence caused."

A simple Google image search highlighted on Twitter has been said to highlight the pervasiveness of racial bias and media profiling. "Three black teenagers" was a trending search on Google on Thursday after a US high school student pointed out the stark difference in results for "three black teenagers" and "three white teenagers". Kabir Alli of Virginia posted a clip to Twitter of himself carrying out a straightforward search of "three black teenagers", which overwhelmingly turns up prisoners' mugshots. He and others erupt in laughter when the result for "three white teenagers" show stock photos of smiling, wholesome-looking young people. The tweet has been retweeted by more than 60,100 users and favourited nearly 55,500 times since it was posted on Tuesday – but Alli's video was later reposted by World Star Hip Hop, an entertainment website with an enormous following on social media.

Distributed computing offers a high degree of flexibility to accommodate modern learning constraints and the ever increasing size of datasets involved in massive data issues. Drawing inspiration from the theory of distributed computation models developed in the context of gradient-type optimization algorithms, we present a consensus-based asynchronous distributed approach for nonparametric online regression and analyze some of its asymptotic properties. Substantial numerical evidence involving up to 28 parallel processors is provided on synthetic datasets to assess the excellent performance of our method, both in terms of computation time and prediction accuracy.