While we may not think of it very often, the European Union is home to 24 official languages: Slovenian and Swedish, French and Finnish, Polish and Portuguese. And that is only a quarter of them.
It certainly makes for a rich and rewarding socio-cultural environment in which to live and work. But it also brings difficulties for people and businesses to understand each other – especially online - and to operate across borders.
Many people have voiced their concerns to me about where, and how, multilingualism fits into our plans to build a Digital Single Market (DSM) in Europe.
These concerns stem partly from a perception that while the European Commission often refers to many barriers that are holding back the creation and progress of the DSM, the language barrier hardly gets a mention.
We should recognise that language differences can also act as a barrier, and increase the fragmentation of markets – which is what the DSM is trying to prevent.
Take cross-border e-commerce. Here, there are recognised obstacles: for example, restrictions based on a customer's residence, nationality or location; a practice known as geo-blocking.
But then we consider language differences in this market.
Online shoppers often have limited language skills: perhaps reflected best in the phrase "don't understand, won't buy" - which is, in fact, an unfortunate reality. At the same time, e-retailers can be reluctant to service customers in many different languages.
While this matters less for global online shopping platforms, it is a more serious issue for smaller e-retailers and web-based traders.
This is especially true when it comes to maintaining 24 language versions of your website and also providing after-sales services in the same multitude of languages. It means that multilingualism can come with a price tag that turns it into a business hurdle.
Or take data analytics, a fast-growing and important sector for the DSM that we are building. Different data sets expressed in different languages cannot always be easily synthesised or processed.
There are also issues of semantics and equivalence of meaning to take into account. That might slow down or complicate the full use of data analytics.
Fortunately, there have been significant achievements and advances in automated multilingual technologies such as machine translation. In many areas, automation has become the main source of translations. Not only does it provide translations quickly, it can also significantly cut the cost of multilingualism.
The European Commission has already invested more than €200 million over the last seven years on research and innovation in language technologies that have the potential to break through language barriers.
We are well supported in this task by EU countries and by the language technology industry.
The Commission also helps by supporting the infrastructure that paves the way to smooth deployment of language technologies.
This is now being done with funding from the Connecting Europe Facility (CEF) for machine translation and other mature technologies to provide multilingual public e-services. The CEF also promotes the collection and sharing of language resources from all EU countries – plus Norway and Iceland – to make European public services multilingual.
This week, Slovenia has been hosting a major multilingualism event: the Language Resources and Evaluation Conference (LREC 2016), held in Portorož. I am confident that the expert brainpower of the producers, researchers and practitioners of language resources gathered there will assist us further in our work to turn Europe's linguistic diversity from a barrier into an asset.
Overcoming language barriers is vital for building the DSM, which is by definition multilingual. It is now time to reduce and remove the language barriers that are holding back its advance, and turn them into competitive advantages.
Another blog soon.