News

Changing how a country types

France adopts new keyboard standard created with state-of-the-art algorithm
New AZERTY keyboard
The new keyboard standard, created with a state-of-art-algorithm, makes typing in French easier and more comfortable. Photo credit: Mikko Raskinen/Aalto University

Update on 27 January 2021: Link to academic publication added.

Keyboards touch our everyday lives yet, despite the well-known drawbacks of current layouts used across the globe, the position of characters has largely remained the same. Researchers at Aalto University, as part of an international collaboration, have now used computational methods to place keyboard characters for easier, more comfortable typing. The result is a new keyboard standard created with an advanced algorithm, introduced by France on 2 April 2019. The research was subsequently published in the February 2021 issue of Communications of the ACM.

‘Algorithms, like the ones we have developed for the French keyboard, can help us make better decisions. They can quickly evaluate the problems and benefits of different designs and achieve fair compromises. But they also need the guidance from humans who know about the problem,’ explains Dr Anna Maria Feit, the lead researcher of the project.

With concern voiced by the French government in 2015 on the existing keyboard—and its inability to support the proper use of French—priority was on creating a new standard that allows easy and quick use of required symbols. The algorithm created by the Aalto University-led team automatically arranged the characters in an optimal way.

The new AZERTY standard includes commonly used characters in the French language, such as œ, « », or É, as well as 60 other new characters, which are arranged in groups predicted by the algorithm, making the layout more intuitive to use. Characters like @ and / have been moved to more accessible locations, as have ligatures and accents.

‘When rearranging the symbols on the keyboard, there are conflicting things to consider,’ says Feit, who completed her doctoral studies at Aalto and now works at ETH Zurich.

‘Characters that get used the most should be moved to a position that is easy to reach. But if you move it a long way from where it was originally, people will take a long time to learn it and be less likely to use your new layout. You might also want to keep symbols that look similar and have similar functions together to make them easier to find and use, like the colon and the semi-colon, even though one gets used more than the other,’ she explains.

To inform the design, researchers built statistical models of character use in modern French, drawing on newspaper articles, French Wikipedia, legal texts, as well as emails, social media, and programming code. In contrast to previous work that assumes people use their fingers in certain ways, they gathered the key presses of over 900 people in a large-scale crowdsourcing study to see what counted as an ‘easy’ key press. In addition, they included state-of-the-art findings from ergonomics literature.

‘The trick to making the collaboration effective was using our data to build a tool that the French experts in the standardisation committee could put different conditions into, and see what the optimal keyboard that resulted from the data looked like,’ says Aalto University Professor Antti Oulasvirta.

Dr Mathieu Nancel, a former researcher at Aalto now based at Inria Lille – Nord Europe in France, brought the algorithm to the French committee and helped them to work with it. ‘Before we started working together, they tried to place over 100 characters by hand. Our tool allowed them to focus on higher-level goals, such as making typing special characters fast or keeping it similar to the previous layout,’ he says.

‘Together with the committee, we tried different parameters and discussed the layouts suggested by the computer algorithm. We could also change the layout by hand and the tool would tell us how this impacted typing speed or ergonomics. We then adapted the underlying computer model to also take into account, for example, cultural aspects and comments from the French public,’ Nancel adds.

The algorithm that Dr Feit and team produced for the French committee can easily be adapted for any language; it simply requires data for modelling. Most countries use the standard QWERTY keyboard—originally designed for the English language—despite frequently used accented characters or different styles of punctuation. Dr Feit hopes that the model produced for France could be used by other groups in the future.

‘Our goal is that in the future people and algorithms design user interfaces together,’ she says.

Frequently Asked Questions

Related resources

Further information

Prof Dr Antti Oulasvirta (speaks English & Finnish)
Aalto University
[email protected]
+358 50 3841561

Dr Anna Maria Feit (speaks English & German)
ETH Zurich
[email protected]
+41 44 632 35 01

Dr Mathieu Nancel (speaks English & French)
Inria Lille – Nord Europe
[email protected]
+33 3 59 35 87 96

Maximilian John (speaks English & German) 
MPI for Informatics, Saarbrücken 
[email protected] 

Dr Andreas Karrenbauer (speaks English & German) 
MPI for Informatics, Saarbrücken 
[email protected]

  • Published:
  • Updated:
Share
URL copied!

Read more news

Koeputkia, joissa on eri värisiä nesteitä.
Press releases Published:

A greener route to blue – a new method drastically reduces the amount of solvent needed to produce widely used organic dyes

Phthalocyanines dyes can be produced with solid-state synthesis instead of high-boiling organic solvents.
Havainnekuva rintasyöpäsoluista mikroskoopissa
Press releases Published:

Breast cancer cells use forces to open up channels through tissue

A new method reveals that cancer cells can spread by rapidly applying forces to tissue material.
Tuoleja ravintolatilassa, taustalla asiakaspalvelutilanne
Press releases Published:

New technologies can help people make sustainable dietary decisions

Blockchain-backed app provides information about food impacts and combined customer choice
An illustration with a graph on the left and a molecular structure inset in a cube on the right. Each curve on the graph is a different colour, and each is connected by a line to an inset circle with a specific molecular feature corresponding to that curve. Above the cube with the molecular structure is a squigly arrow coming in, labelled "hv", and a straight arrow going out, labelled "e-". The entire figure (graph and inset cube) is labelled "XPS".
Press releases Published:

Machine learning gives material science researchers a peek at the answer key

A model trained to predict spectroscopic profiles helps to decipher the structure of materials