27/01/2023

Our translation and localisation system oneSuite integrates all logically related components for professional use. In this article, we explain the benefits and special features of the system – and how these are related to the open-source development approach and cloud-based software architecture.

oneSuite, our proprietary browser-based translation and localisation system, is a next-generation CAT (multi-)tool. In addition to the “traditional” functions, such as translation memory and terminology management for computer-aided translation, it integrates numerous other application programs and features for review, post-editing, multimedia localisation, and evaluation and terminology processes.

The system is available as a complete solution with components that logically and effectively complement one another, which can of course also be used as stand-alone applications.

As an agile software solution, oneSuite is based on the “collective intelligence” of a large number of professional users, so that development steps can be derived very quickly from user feedback, preventing undesirable developments that do not meet the requirements of users. Since the first day of the roll-out, the system has provided an impressive, fully operational range of functions, offering more than many comparable systems, even at that early stage.

Unlike proprietary CAT systems, oneSuite’s functions can be expanded and customised as required. Thanks to the open-source approach, the system is much more flexible and offers almost endless possibilities for integration and expandability, as the basic technology based on an open-source approach generates innovations faster than proprietary systems.

Thanks to its flexible integration options and open-source concept, specific customer requirements for processes can be implemented quickly and flexibly. External systems and services can be integrated and automated at any time. During the ongoing agile development of the system, additional functionalities are added depending on requirements and user feedback.

The latest addition, for example, is quality estimation, a function that can provide an automatic quality estimate for machine translation without having to rely on human reference translations. Quality estimation therefore helps find out how good or bad the translations produced by an MT system are without the need for human intervention at this stage.

oneSuite is a development from practice and for practice, a noticeable feature every time each of the services is used. The system was created in close cooperation with Marc Mittag and MittagQI, who developed the basic technology translate5 in coordination with the developer consortium, which includes oneword as a founding member and co-developer.

In contrast to proprietary CAT systems, it is not a single company that develops the basic software. Behind translate5, and also oneSuite as a particular instance, is a broad-based, skilled team of translation and language management experts – all of them user specialists with a thorough grasp of the requirements of both translators and industry customers, who keep an eye on changing requirements.

Another special feature is that, unlike comparable systems, the system is based on new, modern software architecture – the container principle.
To explain: oneSuite is based on a horizontally and vertically scalable cloud infrastructure using Kubernetes and Docker. In the development process, the Docker platform enables fundamental software components, also called microservices, to be ‘containerised’, while Kubernetes serves as a platform for running and managing containers from many container microservices. (You can find more in-depth information here.) This completes the story, because this modern cloud set-up is part of the reason oneSuite is so unique and flexible to use.

Do you have extensive translation and localisation work that requires multiple applications to be connected in a meaningful way? Then take a look at oneSuite. Contact us. Our experts will be happy to advise you on solutions and options.