“Digital Sovereignty” can’t be about an OS “made in France”

Jun 6, 2014
Vote on Hacker News

The phrase “digital sovereignty” has been all the rage in France since Skyrock CEO Pierre Bellanger first introduced it in 2011: “digital sovereignty is the mastery of our present and our fate as expressed in the use of tech and computer networks”. “No national sovereignty is possible without digital sovereignty. The Internet is a global network entirely controlled by the USA. US companies are most often dominant. Dependence and the transfer of wealth caused by this imbalance should lead the government to implement specific internet industrial policies”, he said when the Conseil National du Numérique (National Digital Agency) was created two years ago.

Economics Minister Arnaud Montebourg took the issue at heart when he called for the creation of an “operating system made in France”, which he believes can constitute an alternative to American operating systems and prevent the massive transfer of data from the European to the American continent that “siphons off our jobs, data, private lives, intellectual property, prosperity, taxes and our very sovereignty”. “We don’t want to become a digital colony of global internet giants”, Montebourg said in May. A French OS would be an alternative platform for the development of applications and services that would also ensure protection of privacy. The network effect would give France more power to balance the relatively small size of the French nation. The OS should therefore be regarded as an extension of our state.

The idea may seem attractive, but Bellanger and Montebourg champion the creation of a French network that would be run by… Orange! How could French telco giant Orange, whose culture of customer service and user-friendliness is very limited, possibly compete with Google and Apple? Orange is anything but a software company. Many ‘Livebox’ users have made a habit of cracking jokes about Orange’s poorly designed apps. The very idea of Orange running a platform that targets hundreds of millions of users seems somewhat absurd (or hilarious).

Either the OS project would be started from scratch, in which case it would be ineffective, expensive for taxpayers and it would just die off, or it would use and modify Google’s open source OS, in which case there would be nothing “French” about it, other exterior applications could still be run “backdoors” and it would not be much different from what we already have.

An operating system run by Orange would have little chance of being adopted on a large scale. Without applications it would remain an empty shell. In spite of Microsoft’s firing power, Windows Mobile is finding it exceedingly difficult to break through. Unless we imitate the Chinese and FORCE developers to use the new OS, it seems unrealistic to imagine convincing apps developers to work with an “OrangeStore”.

The development of French-based infrastructural standards is an old pattern that can best be described as France’s “Minitel Syndrome”. French companies had better try and create value by focusing on apps that generate and maintain long-term relationships with end-users. The French government should focus on creating the right ecosystem for the emergence of billion-dollar companies that will invent tomorrow’s new platforms.

French engineers can design beautiful tools… without users. I would love to know what became of Cloudwatt and Numergy, the sovereign cloud solutions launched two years ago, which to my knowledge have yet to offer users something. How long is it supposed to take?

The OS battle has already been fought. There are several independent open-source platforms that are turning the OS into a commodity. As Dave Stutz wrote in 2004: “As a platform succeeds in the market place, its APIs, UI, feature set, file formats and customization interfaces ossify (…). This process of ossification makes successful platforms easy targets for cloners, and cloning is what spells the beginning of the end for platform profit margins”; “platforms exist at many layers of a software system and slowly come and go as their usefulness waxes and wanes over many product cycles”.

Cloning is precisely what makes the very idea of an OS “made in France” somewhat obsolete. France had better focus on future platforms rather than past ones. Because operating systems are a commodity, the only way to create value out of them is to have a billion users, which can’t happen with a new Orange-controlled OS. The development of such a project would probably represent a waste of valuable time and taxpayer money. Our entrepreneurial age does require strong and efficient public intervention but the latter should consist in 3 things: social insurance (unemployment benefits and quality health care) to sustain entrepreneurship, the deployment of open APIs (taxis, for example), and the promotion of venture capital (which should be made more fluid thanks to new regulation and a reformed tax system) to finance innovation. Capital would also be better allocated if there were fewer regulatory barriers.

I’d like to finish with one more Dave Stutz quote which the French elites would do well to ponder: “Software as a service is not service in support of software, but software in support of user-friendly services”.