Happy World Standards Day 2017

Today on October 14 we celebrate the World Standards Day which is a good opportunity to review how standards impact our everyday's life. In fact, many standards help me creating this blog post ranging from web standards (W3C), communication standards (IETF, ITU), and standards defining representation formats (e.g., JPEG, MPEG).

Perhaps you are wondering how such standards are created. In MPEG, for example and in a nutshell, new work items are proposed and discussed within the requirements subgroup, which typically issues a requirements document followed up by a call for proposal. The responses to this call are discussed and evaluated according to predefined criteria and adopted into a working draft. Once the working draft becomes mature, MPEG may decide to issue a Committee Draft (CD), which goes out to national bodies for ballot. If national bodies agree on the CD, which could include comments on how to improve it, the next stage would be Draft International Standard (DIS) followed up Final Draft International Standard (FDIS), each accompanied by a ballot including comments. At FDIS stage mainly yes|no vote is allowed and only pure editorial comments can be integrated before going to International Standard (IS), which is when the standard is finally published. [note: sometimes it's a bit more complicated but this is another story - for the interested reader, I've documented the process when working on MPEG-DASH here]

This may sound like a very boring process but it's also possible to win Engineering Emmy Awards like HEVC did very recently, where also Leonardo Chiariglione received the Charles F. Jenkins Lifetime Achievement Award (see his response to this award here). 

In this context, I often quote the following xkcd comic which shows two sides of the coin. First, the obvious one where one should indeed not create the 15th competing standards, which one may think it's easy but it isn't although there's also a positive aspect about this (see at the end of this blog post). Second, standards should only define the minimum to enable interoperability and leave out enough space for innovation and competition. However, it's not always clear from the beginning, where to draw the line in order to become a successful standard.

https://xkcd.com/927/

In the past couple of years I was heavily involved in the standardization of MPEG-DASH. In the beginning we've been in the situation with multiple competing formats (Adobe HDS, Apple HLS, Microsoft Smooth Streaming, etc.). MPEG-DASH was finally adopted by Adobe and Microsoft, leaving HLS as competing format/standard (i.e., informational RFC 8216) which now utilizes MPEG's Common Media Application Format (CMAF) to allow a common media segment format to be used by both DASH and HLS. Thus, we did not create the 15th competing standard and DASH/HLS/CMAF is an important step towards reducing market fragmentation although it's not yet the end of the path.

I'd like to conclude with two quotes related to standards. One is from one of my professor at university who was saying "if you have sleeping problems, read a standard", which is true - they are boring to read for an outsider - but it's exciting to work on standards as you basically define the path for future products and services. Finally, my favorite quote goes back to Andrew S. Tanenbaum's book on computer networks: "The nice thing about standards is that you have so many to choose from" which I interpret as a positive statement as competition leads to innovation which eventually leads to innovative products and services - that's what we want.

In this spirit: Happy World Standards Day!