Open-source desktops: where are we at?

CNLL’s latest study published in early 2020 showed that “the open-source software industry in France is predicted to grow by 9% consistently every year until 2023, which is above the average the overall IT market which was close to 4% in 2019-20. In almost 80% of companies, the use of open source will increase in the next two years”.

According to Red Hat’s “The State of Enterprise Open Source” report, open-source software is increasingly present in organisations. It is even expected to “overtake proprietary software in two years”. Red Hat contends that proprietary software is losing favour because of expensive and inflexible licences and vendor lock-in.   

Open source has been making its way up through infrastructure layers, all the way to business applications and end users, but is the thought of an entirely open-source desktop realistic today? Where are we at? What are the drivers and stoppers at play?

Usability as a barrier to entry

In early January 2020, French Defence Minister Florence Parly announced in the Official Gazette of the Senate (09/01/20) that the government was looking into a “completely open source desktop”. This announcement came in the wake of the renewal of a contract between… Microsoft and the French Ministry of Defence. Said renewal was made without a call for tenders and in spite of security and sovereignty experts’ advise against it (Zdnet). 

The Ministry is therefore contemplating the acquisition of fully open-source desktops (operating system and office software) while continuing to sign contracts with major proprietary US vendors.

But besides these giants’ commercial – and lobbying – clout, why does open source still struggle to impose itself on user desktops?

Because the thorn in the side of open source is users. Or rather, what we refer to as “usability”: a software’s user-friendliness and learning curve. People are weary of open source because it is perceived a developers’ domain. Users don’t relate to it and they believe that it is an IT people’s thing.

This is particularly true in enterprise email, in which user habits rule. Business email is the communication channel most used on a daily basis. It handles organisations’ critical information (invoices, orders, appointments, documents, key data, etc.). No CIO can afford to alienate its users on this subject and risk hurting productivity.

Building a viable alternative to major US vendors’ offering is not just an issue of sovereignty or technical feat. Europe does have an exceptional skill pool. But the key driver of change is human rather than technical. Transitioning to healthier technologies will not succeed if it is made to the detriment of users – it can only be considered with them, with user comfort at the centre of the game. This premise – as far as end-user applications are concerned – is forcing a re-thinking of the open-source economic model to serve better user experience.

Where do users fit into the open-source service model?

Traditionally, open source has been based on a “service model”, i.e. providing and selling services for open-source software (which you may be developing or may be available as-is online), e.g. installation, integration, maintenance, support, training, consulting…

This model has proven effective in building the foundations of digital technology: infrastructure, technical components, networks and middleware. Shared development and shared expertise among engineers has led to the development – partly through standardising – of some areas such as the internet or, more recently, DevOps.  

Yet, in its blog, the Open Source initiative concedes that the historical model of open source does not work well for user applications as traditional open-source projects failed to take sufficient account of end users, essentially focusing on software quality and power to resolve a given technical task instead: “Open Source, by its nature, is built by developers for developers. This means that although the primary consumers of open source are end users, they get pretty much no say in how the project evolves. […] users often don’t know what they want and even if they do they can’t put it in technical terms, meaning that all user-driven product development involves extensive and expensive product research which is far beyond any open source project.”

So open source’s approach to user desktops has to change: it must shift from focusing on technical features to improving user experience.

Open source software publishing: a new dimension

With the old cliché of the bearded, garage-bound geek clandestinely coding clinging to it, open source often finds it hard to get its economic rationale through. Its development philosophy is based on sharing, mutual help and transparency. It is not a pre-packaged, off-the-shelf product.

“If a piece of software requires configuring 20 separate options in order to be usable, it’s accumulated far too much UX debt to be usable by most people, as it requires specialized knowledge that only a limited number of people have.” explains Máirín Duffy, Senior Principal Interaction Designer at Red Hat.

It is by levelling out this debt that open-source software can rise up from infra-layers to the end user. How? “We must embrace our business model” says Pierre Baudracco, CEO of BlueMind and co-Chairman of  CNLL (National Free Software Council). “We’re an open-source software publisher. We don’t sell source code, we sell solutions. There’s a huge difference.”

Unlike in the traditional open source model, we focus on usage. At BlueMind, we offer an open-source collaborative email solution. As a service, email used to be addressed and discussed among technicians (IT managers, system and network administrators, etc.) looking at performance, network bandwidth, architecture, etc. As long as messages came in and out properly, that was enough as far as usage was concerned. This service has since become ubiquitous (it is by far the most used in the business world and it is continuously growing). It has reached everyone across all organisations and has gained more features. Email has expanded outside the technical sphere and, as a result, discussions about email services now concern functional features and usage (will I be able to use Outlook? Will VIPs be able to access their calendar and shared features from their mobile phones? Will the browser interface be user-friendly? Will users be able to delegate their calendar privileges? Etc.)

“Everyone uses email at all levels of a company. For an open-source solution to have a chance of competing with leading US vendors, it must take into account users, most of whom use Outlook. Respecting user habits means being capable of supporting Outlook as transparently as with Microsoft Exchange. It also means allowing all usage and migration scenarios while offering the best Thunderbird, Mac, mobile and web-based support. That’s what we’ve done with BlueMind v4” Thomas Cataldo, our Technical Director, explains.

BlueMind is designed as a complete solution rather than mere technically-powerful source code. This vision as a “solution publisher” (as opposed to “code producer”) will help us improve usability and by extension, expand the adoption of open-source solutions among workers.

Being a publisher means that we spend 20-30% of development time on producing code, the rest being spent on auxiliary tasks – testing, documentation, version support on platforms, ensuring continuity, client-scenario support, auxiliary tools, ecosystem development, etc.

Learning to change: a new role for CIOs?

The benefits of moving to open-source solutions in terms of technology, independence and economics are clear, as a 2018 Cigref study pointed out. The spanner in the works is change, both at the end-user and CIO levels.

This is precisely the issue France is struggling with, as evidenced by Defence Minister Florence Parly’s announcement. Beyond the technological migration it obviously entails, an open-source desktop implies a change in mindsets – Users’, on the one hand, whose productivity IT tools affect directly and who are therefore afraid of having their habits upset. CIOs’, on the other hand, who must take on a new role, that of agents of change.

Gartner predicts that by 2021, CIOs will be as important to their companies’ culture change as Chief HR Officers. “A lot of CIOs have realised that culture can be an accelerator of digital transformation and that they have the means to reinforce a desired culture through their technology choices,” said Elise Olding, research vice president at Gartner.

Obviously, CIOs can and will play a key role in digital transformation, and this won’t happen if they restrict their attention to infrastructure and servers as opposed to users, (as in some extreme cases in which organisations build a cloud and then ask what they might put in it to “justify” it). But if the organisation — and/or its management — isn’t itself convinced and driving the change, it will be difficult — and perhaps impossible — for it to succeed!

Plenty of incentives actually come from the top!

Everyone must embrace their responsibilities and their choices!

In September 2019, Horst Seehofer, German Minister of Interior, commissioned a “Strategic market analysis about how to reduce dependency on single software vendors” to the management consulting firm PricewaterhouseCoopers. It concluded that Germany was strongly dependent on few software providers, in particular on Microsoft. This results in “pressure points in the federal government, that work in opposition to the government’s [stated] strategic IT goals,” the report notes. Concerns about information security at Microsoft could “endanger the country’s digital sovereignty”.

Horst Seehofer, German Minister of Interior

The report also outlines several options for improvement, including setting a framework and rules for the use of other software in the future, in particular open source.

This report is particularly significant considering that Germany had made the news back in 2006, when the city of Munich had switched to open source desktops before switching back to Microsoft in 2017, triggering massive controversy.

The grim situation we’re currently confronted with only goes to highlight what we already knew. Border closures, some countries’ inward-looking attitudes, the US’s unashamed egocentrism (yet another example of which recently transpired) are just a few reminders of why preserving our sovereignty is so important. Digital sovereignty is part of it. It implies choosing non-hegemonic, sovereign solutions. But to turn this into reality, this choice must be asserted and the efforts it entails must be made. It is up to decision makers, States and companies’ management.


User desktops is one of the last remaining strongholds that continues to resist the constant progression of open source in businesses. This state of affairs is the result of several decades of open source being restricted to technical layers but also the image it is struggling to shed – and rightly so – as an obscure, exclusive field.

Things are already changing with many initiatives by players developing and packaging complete open-source solutions, such as digital workplaces or BlueMind’s v4 — compatible with all usage scenarios and offering a comprehensive product-centred ecosystem.

The skills and solutions are coming or already exist, promoted by open-source publishers through an ad-hoc economic model.

CIOs and company executives have a role to play and it’s up to them to get things moving. The current context shows that change, even with some temporary discomfort, can be accepted and the result is worth it. 

What remains to be seen is what kind of digital technology and independence we want for our companies and our society! #TheChoiceIsHere

Picture of Pierre Baudracco

Pierre Baudracco

Share this article

Leave a Reply

Your email address will not be published. Required fields are marked *

11 + = 17

Subscription to the newsletter

One e-mail per month to keep up to date with all BlueMind news