Work on HTTP/2 by the Internet Engineering Task Force HTTP Working group has been completed, according to Mark Nottingham. HTTP/2 is set to go through the final editing process before it is published and becomes and official Web standard.
The announcement came after little more than a week after Google announced that it was discontinuing SPDY in favor of HTTP/2 inside Chrome. SDPY won’t fully removed from Chrome until early 2016 but HTTP/2 will be implemented in Google’s browser in the coming weeks.
Why does it matter?
Since HTTP is part of the foundation of the Web, any changes to the protocol is a very big dead. HTTP/2 is said to make response times faster for Web browser and reduce the loads on servers. It will take some time for the new standard to roll out across the web and all of its issues to get sorted out.
The biggest difference between these two standards, is a new feature called multiplexing. When you pair it with header compression, it allows multiple server requests to be sent at the same time. HTTP/2 uses fewer connections between the server and the client which allows the server to push content straight to a browser.
The load times will be dramatically improved because your browser might even receive the date before requesting it.
One item that won’t be implemented in HTTP/2, is mandatory SSL/TLS (HTTPS) encryption. It was supposed to be the original plan back in 2013 but since then, it was scrapped. HTTP/2 will make TLS encryption easier to implement according to Nottingham, because the new protocol is designed to reduce the speed hits that site usually take that are using HTTPS at the moment.
Accoring to Nottingham, developers for Google Chrome and Mozilla Firefox have said that the two browsers will only use HTTP/2 over TLS. The drawback is that if site developers do not add TLS to and HTTP/2 enabled site, won’t be able to use the new standard with Chrome and Firefox.
The bulk of the work on HTTP/2 is done but the IEFT HTTP WG isn’t making any progress. In fact, it’s already looking ahead to the possible development of an HTTP/3 as well as improving the current HTTP specs with other features such as HTTP message signing for improved server-to-browser authentication