We can learn from the failure of the French-style sovereign cloud by looking at how some of our European neighbours, particularly Finland, have succeeded in creating an economic dynamic for the cloud at national level. Their recipe? The opening up of a collaborative ecosystem between all the stakeholders, without making the administrative state apparatus 'sovereign' over decisions. So what role can the French government play in the cloud "à la française"?
Act 1: The shipwreck
The 'Andromède' sovereign cloud project, initiated by the French government, has not been rescued from the sea. In 2012, Andromède gave birth to CloudWatt and Numergy, triggering numerous disputes in the cloud ecosystem, and today Cloudwatt is shipwrecked, even though it was rescued by Orange, which took over all the shares in March 2015. Numergy is not doing much better, or at least not as well as expected. It's clear that, after promising to achieve combined sales of nearly €1 billion in 2017 with these two companies, the sales figures announced for 2014 were enough to disappoint one of the main shareholders, the French state, which invested €150 million through the Caisse des Dépôts et Consignation, with €75 million in the capital of each of the companies. For the record, CloudWatt announced sales of €2m for 2014 (with a net loss of €29m in 2013) and Numergy €6m, but this came mainly from its main shareholder... SFR.
With the French government officially withdrawing from the French sovereign cloud project in March, Bercy guarantees that less than half of the €150m had been spent by early 2015. This means that, at the very least, more than €65m of public money has already been invested, with no hope of a return... The state has acted like any other shareholder, taking risks, the market, especially the public cloud market, was not guaranteed and was undoubtedly overvalued, and so on. The case should therefore be written off.
But the State is not a shareholder like any other, and it is accountable to its citizens for its decisions.
Act 2: The roots of evil
The real question concerns the exact role that the State should play. Should the State have the right to regulate private players in the French Cloud market? Define the needs of all businesses? To judge who will be able to provide the best solution, even if that means creating new players? To what extent should it, and more importantly, can it?
How can an administrative machine that is not well known for having flexible budgets in the sense that capital expenditure and operating expenditure are compartmentalised, for example, be the voice of all customers who use the cloud and its various uses?
It is understandable that the State should be involved, particularly in guaranteeing French companies a "sovereign cloud" label that guarantees compliance with data security and confidentiality rules on French soil, in compliance with current legislation. If there are at least some functions where the regalian powers of the State are the subject of little controversy, it is certainly those of security and law.
By contributing to the creation of CloudWatt and Numergy, the State acted like a pyramid structure that decided on its organisation from the top down, according to its policy and its traditional channels - CloudWatt and Numergy are nothing more than offshoots of large historical groups born with subsidies to "do cloud" and not companies born from the energy of a few entrepreneurs, natively oriented towards cloud service offerings. To have a chance of existing and surviving, these companies need to create a business model that rapidly delivers value to end-user customers.
At the time when Cloudwatt and Numergy were being launched, there were a large number of competing technologies and French cloud SMEs, and not just in the public cloud. Intrinsec is well placed to know this, having launched its offering in 2007. As a result, the government's approach, by ignoring the voices and dynamics of the players who are already up and running, has led to a great deal of confusion and a wait-and-see attitude in the deployment of the cloud in France, automatically diminishing customer confidence. This partly explains why France is lagging behind other European countries in the adoption of the cloud (see below).
Rather than creating "new" players in 2012, it would have been wiser to involve all the stakeholders in the existing ecosystem, and for the State to lead the discussion as a stakeholder too. With the specificity of being a stakeholder in a position to voice the particular needs of citizens with regard to public services, local authorities and administrations with regard to digital services, but also the fabric of SMEs and large French companies, particularly in terms of privacy, security and competitiveness.
Setting real objectives for the sovereign cloud
The State can set objectives for the French sovereign cloud, in terms of security, functionality and the level of services expected by government departments and local authorities, as part of a vision of what digital transformation should look like on a national scale, in order to develop it responsibly. But to develop the economic fabric and competitiveness, it must also listen to the various emerging and native French players in cloud services, be they in IAAS, PAAS or SAAS mode, their own needs and constraints, and leave them the autonomy of their strategies. It's not a question of "letting the market do its thing", but of helping to bring together national players, suppliers and customers, by understanding all aspects of the cloud, and giving new meaning to the "sovereign cloud" through explicit, shared objectives for the services to be provided.
In what we might call the 'Andromeda affair', the roots of the evil were to encourage the creation of new private structures that would be torn apart by rivalry - it's typical of Andromeda to have Atrides as its descendants - with public funds, without even attempting to define clear objectives for what a French-style sovereign cloud could be, which didn't need that, and without even deigning to consult the existing ecosystem. Instead, the focus should have been on creating links of cooperation and emulation within the ecosystem, so that we could work together to exploit all the potential value and experience of each individual. It's not too late to do this, because it's the only way to restore our competitiveness and achieve a successful 'digital transformation' at regional level!
Act 3: Instead of French-style planning, why don't we get busy?
Admittedly, the new government plan released in 2014 to accelerate the development of the French cloud industry contains virtuous measures, including:
- Creation of a Secure Cloud label for Cloud Computing services.
- Exemplary public action and the creation of a digital marketplace for local authorities
- The adoption of the cloud by businesses (with the notion of aid for digital transformation and marketplaces for digital solutions for SMEs)
- Support for innovation in the cloud, with a focus on usage (development of cloud services for education, health, senior citizens, etc.).
If we can see that the first points are progressing somewhat, we can wonder about this 'French' way of planning. We're on the right track, but there's still a long way to go, and it's certainly not just a question of investment. In this respect, we can take the example of Finland, which has the highest rate of digitalisation in Europe thanks to massive government investment and a favourable environment.