Not so fast, open standards!

My friend Andrew Updegrove wrote a surprising essay in his latest blog post about the irrelevance of open standards. More exactly his point, if I understood correctly, was that open standards were becoming irrelevant as a topic as everyone is using and relying on them, and the software industry can no longer afford to play the game of vendor lock-in towards customers, partners and competitors. If that’s Andy’s opinion I happen to disagree with it, but only partially. Let me explain.

Open Standards fading into oblivion as something that’s not interesting and yet so mundane because everyone would rely on them is somewhat of a good news I think. It is likely that some parts of the industry, such as cloud computing players, cannot afford to “invent” brand new proprietary platforms. Whatever you do, if this particular case is an example, is to design and develop a platform, an infrastructure or a service that is either OpenStack based, or at least fully capable of interfacing itself through Swift, AWS, Azure compatible or otherwise open APIs. While these are not all open standards, it’s a good thing: downstream players want to be compatible, but the upstream, major cloud technologies are open to some large extent as well as it is in their interest to be used and relied upon by the largest part of the market.

There are however some hiccups with vendor lock-in, in cloud computing or elsewhere. It just hasn’t disappeared. The lock-in still exists through proprietary or otherwise unimplementable file formats; through undocumented protocols and weak or non existent reversibility clauses. Vendor lock-in has not gone away, it has become more subtle by moving up the ladder. If your entire business processes are hosted and run by a cloud service provider there may be some good reasons for you to have made that choice; but the day the need for another provider or another platform is felt the real test will be to know if it is possible to back up your data and processes and rebuild them elsewhere and in a different way. That’s an area where open standards could really help and will play an increasing role. Another area where open standards are still contentious is multimedia: remember what happened to Mozilla in 2015 when they chose to embed proprietary, DRM-riddled codecs because of industry pressure.

Now Andrew suggests that the market is turning to FOSS the same way they first turned to open standards. True enough, FOSS has never been as popular as it is today, but I do not believe for a moment that it is because I.T. professionals or their clients understand what Free & Open Source Software is. That’s unfortunate of course, and we do need to keep in mind that open standards and FOSS, while being quite compatible, are two widely different things.

To come back to the original point, I believe something more incidental may explain his perception. From about 2006 to 2010, the world of open standards was full of exciting initiatives, global battles for market domination or liberation. Let’s mention a few of these: html5, microformats, RDF, ebXML and of course ODF, with the OOXML saga. That’s a lot in 4 to 5 years even for the tech industry. In some cases these standards have defined today’s state of the art, in others, they’re found anywhere on the Internet and the enterprise. After these years, open standards continued to grow of course; but the politics cooled down a bit and the bubble deflated.

Open standards are not going away, they still matter and I’m sure they will come back in the spotlight just like with pretty much everything in the I.T. industry. Look, we’re talking again about A.I. I can’t wait for the moment we’ll be bombarded by some paradigm shift in e-commerce or with the fat client as in, fat client and thin server in opposition to the thin client and the fat server where all the logic comes from the server. But I digress. Open standards help everyone who want to have a part to play in the game. Whether that standard ends up being used or not, replaced or opposed by another open standard is not what matters: that’s the life of standards. I’m confident we will see their importance being highlighted again for everyone soon or later.