Andrzej Zydroń, XTM International
CTO @ XTM International, Andrzej Zydroń is one of the leading IT experts on Localization and related Open Standards. Zydroń sits/has sat on, the following Open Standard Technical Committees: LISA OSCAR GMX, LISA OSCAR xml:tm, LISA OSCAR TBX, W3C ITS, OASIS XLIFF, OASIS Translation Web Services, OASIS DITA Translation, OASIS OAXAL, ETSI LIS, DITA Localization, Interoperability Now!, Linport.
Zydroń has been responsible for the architecture of the essential word and character count GMX-V (Global Information Management Metrics eXchange) standard, as well as the revolutionary xml:tm (XML based text memory) standard which will change the way in which we view and use translation memory. Zydroń is also chair of the OASIS OAXAL (Open Architecture for XML Authoring and Localization) reference architecture technical committee which provides an automated environment for authoring and localization based on Open Standards.
Zydroń has worked in IT since 1976 and has been responsible for major successful projects at Xerox, SDL, Oxford University Press, Ford of Europe, DocZone and Lingo24 in the fields of document imaging, dictionary systems and localization. Zydroń is currently working on new advances in localization technology based on XML and linguistic methodology.
Highlights of his career include:
The design and architecture of the European Patent Office patent data capture system for Xerox Business Services.
Writing a system for the automated optimal typographical formatting of generically encoded tables (1989).
The design and architecture of the Xerox Language Services XTM translation memory system.
Writing the XML and SGML filters for SDL International’s SDLX Translation Suite.
Assisting the Oxford University Press, the British Council and Oxford University in work on the New Dictionary of the National Biography.
Design and architecture of Ford’s revolutionary CMS Localization system and workflow.
Technical Architect of XTM International’s revolutionary Cloud based CAT and translation workflow system: XTM.
Specific areas of specialization:
Advanced automated localization workflow
Advanced Translation memory systems
Translation Related Web Services
XML based systems
Web 2.0 Translation related technology
Abstract of the speech of Andrzej Zydroń:
How to link productivity and quality
Over the past decade we have seen many advances in the field of translation technology: Statistical and now Neural Net based Machine Translation, improvements to translator tools from automated alignment and terminology extraction, sub-segment matching, fuzzy match completion, predictive typing, voice recognition through to collaborative Internet based real-time sharing of resources. In addition Machine Translation post-editing has become more and more commonplace in translation workflows. The emphasis lately has been on consistency and quality. There have been many excellent improvements regarding quality assessment as well as efforts to standardize measurement and metrics. This presentation will concentrate on providing clear information and guidance regarding best practices linking productivity and quality and the latest developments as well as looking forward to what the future may bring to our industry.
1. John Moran, Transpiral
John Moran is a German to English technical translator, software engineer and publishing researcher specialised in measuring how linguistic technology impacts on translators. In the 1990’s he worked as a lecturer in technical translation in Trinity College Dublin and founded Transpiral, a technical translation agency based in Dublin, Ireland. He is a regular speaker on the topic of translator productivity at translation industry and translator conferences and the driving force behind WordFace Analytics, a software application used by large translation buyers to measure translation speed in Trados and OmegaT.
Abstract of the speech of John Moran (Panel – How to link productivity and quality):
Translators care about how many words per hour they translate as they are normally paid by the word, but most translator agencies and other requestors care only about word price and translation quality. As technologies like full-sentence machine translation, example-based interactive machine translation and speech recognition begin to have an impact for some language pairs this view is slowly beginning to change.
In my presentation I will provide a short overview of how to gather words per hour productivity data in a range of desktop-based and web-based CAT tools and a description of some of the statistics that can be gathered alongside time data to get a better picture of the translation process. I will discuss the difference between Segment Level A/B testing and job level A/B testing for machine translation and outline a data format used in WordFace Analytics to capture User Activity Data generated by translators as they work in OmegaT or Trados Studio. I will end the discussion of speed data by showing how it could be used to encourage translators to try automatic speech recognition and how that might impact on their health.
Finally, I will touch on an objective and cost-efficient quality metric we use in Transpiral called a Blind Provenance Test which can be used as a ‘smell test’ to see if MT has impacted on quality.
2. Paul Filkin, Client Services Director, SDL Language Technologies
Paul is a Client Services Director for SDL Language Technologies and has worked with SDL since the end of 2006. His main focus is helping users of SDL technology to get the most from their investment and can be regularly seen on Twitter, the SDL Community and in many of the public forums providing advice to anyone who needs it. Paul started the SDL Community for translators and has continued to build on this by creating communities of like-minded users so they can benefit more from sharing their experiences directly in an environment where SDL can most effectively get involved. More recently Paul has been working with a team specifically focused on supporting developers who wish to work with the APIs for the SDL Language Platform; this was an industry first and is still unique to SDL today. He also regularly maintains a blog addressing many of the practical issues faced by translators and translation companies in using technology for their work (http://multifarious.filkin.com).
Abstract of the speech of Paul Filkin (Panel – How to link productivity and quality):
The title of this panel is “how to link productivity and quality” and this immediately suggests the two are at odds. In fact in practice they have probably been at odds with each other for several hundred years and in many industries have been diametrically opposed. This presentation will take a look at how we can use technology within our industry to avoid a strict dichotomy between productivity and quality and produce valuable metrics that can support continued improvement and education as we do it.
3. Patrick Cadwell, ADAPT Research Centre, Dublin City University, Ireland
Patrick Cadwell is a lecturer in translation studies in the School of Applied Language and Intercultural Studies, Dublin City University. He is also affiliated with the Centre for Translation and Textual Studies (www.ctts.dcu.ie) and the cross-institutional research centre Adapt (adaptcentre.ie). He teaches translation theory, terminology, and specialised translation at undergraduate and postgraduate levels, and his research interests include the sociology of translation, the human experience of translation technology, and workplace-based studies of translation activities. He previously worked as a translator in the JA>EN language pair.
Abstract of the speech of Patrick Cadwell:
Translators’ adoption and non-adoption of MT; a comparison of two case studies
This paper arises from two projects being carried out by the ADAPT Research Centre at Dublin City University to examine the human factors which influence the adoption or non-adoption of machine translation (MT) by professional translators during their translation tasks. In the first project, we ran 17 focus groups in September 2015 with 70 translators at the European Commission’s Directorate-General for Translation. In the second project, we ran four focus groups in February 2016 with 20 professional translators employed by a language service provider in the UK, Alpha CRC Ltd. The two focus-group based studies were designed to answer three main research questions:
- Do participants in the cohort of translators studied use MT in their work?
- What reasons do they advance for this adoption or non-adoption of MT?
- What could account for the reasons put forward by these translators?
The paper will point out areas of similarity and difference in the reasons advanced by translators in both settings for deciding to use MT or not in their translation tasks, and it will focus on issues of quality, productivity, and text type, as well as on broader contextual factors, such as translator autonomy, agency, and authority.
4. Peter Reynolds
Peter Reynolds is a consultant at TM-Global specialising in translation management and standards. He is also a part-time Executive Director at Kilgray Translation Technologies. Prior to Kilgray he worked at Idiom Technologies Inc. (now SDL PLC), Berlitz GlobalNet, Bowne Global Solution and Lionbridge. Peter has been actively involved in the development and promotion of standards (notably XLIFF) for over a decade – he is an Irish expert to ISO/CEN, and Deputy Chair of the Irish ISO TC 37 Mirror Committee. He holds and BSc and an MBA degree from Open University, and is a localization and translation industry veteran.
Abstract of the speech of Peter Reynolds (Panel – How to link productivity and quality):
Peter Reynolds intends to speak about the current trends within the fields of translation and localization. In particular, he will discuss issues connected with interoperability, integration, standards and automation.
5. Marek Pawelec, Freelance Translator
Marek Pawelec graduated in Molecular Biology at the Jagiellonian University in 1992 and worked as researcher at faculties of Medicine and Chemistry of the Jagiellonian University, Kraków, Poland. In 2001 he began to work as a full time English to Polish freelance translator: started with literature, later added technical translations of medical, biochemical and chemical texts. Translated 32 novels and millions words of medical texts, mainly medical equipment manuals, but also a lot of pharmacology, IVD and clinical trials-related texts. Experienced computer aided translation (CAT) software user and trainer. Certified memoQ trainer. Presented lectures on several translation conferences and teaches on practical aspects of freelance translation, CAT tools and translation related software. He is interested in technical communication, passionate about terminology management, and sometimes lends a hand to Translator Without Borders. Website: http://wasaty.pl Twitter: http://twitter.com/wasaty LinkedIn: http://linkedin.com/in/wasaty
Abstract of the speech of Marek Pawelec:
The topic of this workshop is relation between productivity and quality. And while it’s easy to define productivity, the quality is a bit more elusive. Everybody wants their translation to be top quality, and subconciously we all „know it when we see it”, but what does that actually mean? How do we define quality of translation and how can we measure it? In this presentation I will talk about definitions of translation quality with practical applications and present the most widely used and emerging quality metrics, including tools for linguistic quality assurance.
6. Ewa Legumina, Quality Project Manager, MAart Agency Ltd.
Quality Project Manager at MAart Agency, responsible for quality assurance and translation resource management in case of complex translation projects for public institutions and big organisations for nearly a decade. Expert at the Polish Committee for Standardization (PKN) Technical Committee 256 on Terminology, Other Language Resources and Content Management (ISO TC 37 Mirror Committee) since 2010. Graduate of the Institute of Applied Linguistics at the University of Warsaw.
Abstract of the speech of Ewa Legumina (Panel – How to link productivity and quality):
In my presentation I shall focus on industry best practice and challenges faced by translation agencies specialising in premium quality translation services while responding to the constantly changing needs of the market, tight deadlines and big translation volumes.
I shall discuss the issues of designing the most efficient project workflows and setting project management milestones in compliance with the prevailing industry standards aimed at ensuring consistently high quality of translation services (i.e. the ISO 17100 standard in particular). ISO 17100, published on 1 May 2015, provides requirements for the core processes, resources, and other aspects necessary for the delivery of a quality translation service that meets applicable specifications. In fact, some client specifications are nowadays aimed mainly at significantly boosting productivity and cost optimisation. If this expectation is approached and managed wisely, it may encourage the Translation Service Provider to come up with efficient solutions aimed at increasing productivity while, at the same time, maintaining an effective quality assurance process.
However, one should bear in mind that translation projects should always be managed according to well-established terminology, technology and process standards and not all translation projects can be approached with the same productivity objectives. So, it is important to recognise what are the priorities in case of each project, if the project specification reflects these priorities, and if increased productivity is one of the aims then it is not boosted at the expense of the client’s quality requirements. Otherwise the effects, including handling of the customer’s feedback and corrective action, may turn out to be far more costly and convoluted than any planned productivity gains or cost optimisation might justify.
7. Szymon Klocek, project manager, Directorate General for Translation, European Commission
Szymon Klocek is a computational linguist and an IT project manager. He has worked for the Directorate General for Translation (DGT) of the European Commission for almost 10 years, initially as translator and later as project manager in the DGT’s IT Unit. Since 2014 he has been responsible for the translation quality of the European Commission’s Machine Translation system, MT@EC (Machine Translation at the European Commission). As Machine Translation Data Administrator, he is also responsible for managing, for machine translation purposes, of DGT’s internal data collection, which is one of the biggest sets of aligned corpora of high quality human translations in the world.
Szymon Klocek is a graduate of the Institute of Applied Linguistics and of the Institute of International Relations of the University of Warsaw. He is particularly interested in practical aspects of deploying CAT tools in large organizations, computer-assisted terminology management and limitations of statistical Machine Translations systems when dealing with translating into morphologically rich languages.
Abstract from the speech of Szymon Klocek (Panel on trends in technology):
Machine Translation at the European Commission – Opportunities and Threats
Since 2010, the Directorate General for Translation of the European Commission has been developing a machine translation system called MT@EC (Machine Translation at the European Commission). In my presentation, I will explain the infrastructure which allows MT@EC to be fully integrated into the European Commission’s working environment of computer assisted translation tools and the adaptations which were necessary to make MT@EC’s output exceed the typical quality of a plain statistical MT system. I will also discuss practical aspects of MT@EC, trying to show how it is perceived by translators who use it on a daily basis, including both opportunities and threats inherent to using a MT system in a large translation organization.