PEOPLE | May 10, 2018

Free software: is there a market?

Interview with Roberto di Cosmo, Director of the Innovation and Research Initiative for Free Software (IRILL)

In France the free software market is worth 4,5 billion euro, equivalent to 9,9% of the ICT market.
What is the secret of this success?

The answer comes from Roberto Di Cosmo, an Italian “lent” to France, computer scientist, graduate of the Scuola Normale Superiore di Pisa, and now Director of the Innovation and Research Initiative for Free Software (known by its French acronym, IIRILL).

“The story of the constant and rapid market growth of free software in France began about twenty years ago, when free software made its first transition towards real industrial implementation”.

It is thus an old story that will be explained in detail during two meetings, on Monday 14 and Tuesday 15 May, addressed to public administrations and businesses, and organized by the Nexa Center of Politecnico di Torino along with the University of Turin, Torino Wireless Foundation, Istituto Superiore Mario Boella, CSI Piemonte and the ICT Group of the Turin Industrial Union.

“Although it appears to be a French case, we can say that we found ourselves faced with a combination of two key factors: a political voluntarism which provided a top-down thrust, and a fertile ground of individual skills and implications that produced a real bottom-up thrust. The result was double-digit growth that has taken the free software market and associated services from practically zero in 2000 to almost 5 billion euro in 2018. In Italy, unfortunately, this favorable conjunction has not been seen in the past, despite more or less sporadic individual efforts and ideas. Beyond analysis of the reasons for this absence, it is more interesting to ask what Italy could do to catch up. One of the rare advantages of being behind is the opportunity of observing what has been done by those who arrived before. This is the purpose of the cycle of seminars scheduled for next week at the Politecnico di Torino”.

Today there is increasing talk of choosing cloud for both PAs and businesses. What is your opinion on this trend?

Is it really outdated, as some say, to talk about choice of free applications?

“One of the important advantages of free software is giving the user back a high level of control over technology and its evolution, thanks to access to the source code of solutions and its process of development.

Use of the cloud can call into question these advantages, especially if you choose PaaS or SaaS solutions over which you have no control, but it is not a question of inevitability: you can use cloud services with free applications at various levels of the software stack, and then keep freedom of choice and control of technology”.

In Italy, the Digital Administration Code gives indications about the preference to be reserved for solutions that are free or in reuse, but this has not been sufficient for dismantling old monopolies.  So why is it needed?

What are the best migration practices you know of that one could draw inspiration from?

“As the story of free software in the French public administration teaches us, laws can be important for removing brakes and excuses on the adoption of solutions based on free software, but they are not sufficient.

To go further, it is essential to rely on a pragmatic approach that favors two important axes: the quality of the information system built at the service of the Public Administration and management of the change necessary for accompanying the transition”.

What is the value of data today? Does it still make sense to talk about Open Data?

Data are very important, but with various facets. On the one hand, there is a clear will of numerous governments and public administrations to provide the information of public interest they hold as Open Data: for example information on toponymy, cartography, public spending, but also many other data, which can then be used to provide value-added services to citizens. In Italy I would mention the dati.gov.it portal, in France the work of Etalab or in Canada the working group on Open Government. In this context, the major work that remains is to make Open Data useful, providing the data in open formats, and if possible updated in real time (we are far from having reached that point).

On the other hand, there is a strong desire by many businesses to build a set of data (anything but open, in this case) extracted from the mass of personal use, with the aim of extracting information with high added value, and the use is not at all reassuring (see the classic case of Cambridge Analytica). In this case, it is important that the legislator protects personal data, and the GDPR is an important step in this direction.

In between, there are all the data produced by research activities, which should, in principle, be made open: here we are talking about public research in the great majority of cases, and it is a question of making the result of research activities available to everyone.

Here the issue is to make sure that these data are not privatized, but also that any sensitive data are not made public“.


Sonia Montegiove