| { | |
| "paper_id": "1997", | |
| "header": { | |
| "generated_with": "S2ORC 1.0.0", | |
| "date_generated": "2023-01-19T10:25:48.088249Z" | |
| }, | |
| "title": "Why don't they use translation tools?", | |
| "authors": [ | |
| { | |
| "first": "Hanne", | |
| "middle": [], | |
| "last": "Fers\u00f8e", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "", | |
| "institution": "Hanne Fers\u00f8e", | |
| "location": { | |
| "postCode": "DK-2300", | |
| "settlement": "Copenhagen S", | |
| "country": "Denmark" | |
| } | |
| }, | |
| "email": "" | |
| } | |
| ], | |
| "year": "", | |
| "venue": null, | |
| "identifiers": {}, | |
| "abstract": "Anybody who has ever had to translate a handbook, on no matter what subject, has experienced the challenge of establishing and maintaining a good and consistent terminology throughout the translation. A terminology data base is an obvious tool for meeting that challenge. So why doesn't everybody use one? The same question can be asked for other translation tools as well. As a language technology consultant, I have heard many different explanations over the years. It will be my pleasure to contribute a number of these as input to the following discussion.", | |
| "pdf_parse": { | |
| "paper_id": "1997", | |
| "_pdf_hash": "", | |
| "abstract": [ | |
| { | |
| "text": "Anybody who has ever had to translate a handbook, on no matter what subject, has experienced the challenge of establishing and maintaining a good and consistent terminology throughout the translation. A terminology data base is an obvious tool for meeting that challenge. So why doesn't everybody use one? The same question can be asked for other translation tools as well. As a language technology consultant, I have heard many different explanations over the years. It will be my pleasure to contribute a number of these as input to the following discussion.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Abstract", | |
| "sec_num": null | |
| } | |
| ], | |
| "body_text": [ | |
| { | |
| "text": "A continuous source of wonder for the language technology consultant is the relatively low market penetration today's translation tools have in spite of the fact that although not perfect, the tools can be extremely helpful, they offer a good return on the investment, and their integration into the documentation process gives the company a competitive advantage.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Introduction", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "I am therefore grateful that the organising committee of the 1997 EAMT Workshop has offered me this opportunity to share and discuss with its audience a number of reasons for not using language technology that I have encountered over the years as a language consultant. I have examined the reasons a bit closer and interpreted them with the purpose of discovering the motives that may lie behind them. I offer these findings and my recommendations to the companies for discussion in this forum.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Introduction", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Just by participating in the workshop this audience is showing an interest and a level of awareness which confirms my general impression of an increasing, positive, and enthusiastic interest in translation tools. Some five years ago it was not unusual to be met with a suspicious and somewhat hostile attitude; that is not normal today.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Introduction", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Some of the audience may not be using tools in their organisation today, and knowing that they have good reasons for this choice may find my observations provocative. I do intend to be a bit provocative in order to stimulate the discussion, but no offense is meant and none should be taken. EAMT Workshop, Copenhagen, May 1997 ", | |
| "cite_spans": [ | |
| { | |
| "start": 291, | |
| "end": 326, | |
| "text": "EAMT Workshop, Copenhagen, May 1997", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Introduction", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "I ----------------------------------------------------------------------------------------------------------------------------------------------", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Introduction", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "This presentation builds on experience in language technology consultancy since 1990, in several different countries, including countries outside of Europe. But since most of the consultancy work has taken place in Denmark, which therefore constitutes the major source of inspiration, the examples build on my findings in Danish companies only.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Experience Basis", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "The problems and challenges for companies in handling one or more foreign languages are at the same time unique for each company and basically the same all over the world. If we look at broader perspectives such as size of economy, language area, and telematics infrastructure, however, there are certainly differences between the countries.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Experience Basis", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Characteristic features for Denmark are for instance:", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Experience Basis", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "\u2022 An economy characterized by small and medium sized companies with less than 10 employees \u2022 A small language area (5,5 mio. people) \u2022 A long tradition for professionally educated translators \u2022 A broad working knowledge of English \u2022 A good knowledge of foreign languages in general \u2022 Very few language technology tools exist for Danish \u2022 A well developed telematics infrastructure", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Experience Basis", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "The company types represent a broad cross section of the information intensive part of the Danish business life. There are Danish subsidiaries of multinational companies, Danish companies, export as well as import oriented, and of course translation service providers. The business areas are manufacturing, trade and media. There may be a number of specific reasons why this or that company should use one tool or the other and how the tool should be used in the most appropriate, company specific way. However, over the years, these reasons have all boiled down to four simple and well known arguments for using translation tools.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Experience Basis", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "The tools can contribute to a better product quality, not only in the final product but in the entire production process. Competition is very hard for e.g. consumer products such as kitchen appliances, video machines, and even computers, and distinguishing features and functionality are minor. Certified quality management processes are competitive factors, and here language technology tools have an important role to play in the production. Text consistency is a traditional problem encountered by any company producing for the world market who has to deliver documentation in many languages and perhaps maintain documentation for several versions or releases of the product at the same time. The consistency issue is perhaps most critical for terminology, but it is relevant for style decisions throughout the entire text. Consistency in terminology and style is a central quality issue for the end user of the product, and the producer can achieve it much easier with the use of e.g. translation tools.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Quality", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Reduction in time to market is a critical factor in most industries. Translation is the last step in the production process, and very often too little time is left for the translation due to delays in earlier steps of the process, which have eaten up the time set aside for translation. By applying translation tools it is possible to reduce the translation time by reusing previous translations and thus avoid having to translate, proof read, and validate text which has already been translated, proof read, and validated before, and it is possible to distribute the translation of one document or one project between several persons and thus gain speed while maintaining consistency.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Time to market", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "In the early days of language technology, much focus was on the possible cost reduction in a labour intensive process. Cost reduction is important for the companies but not primarily through reduction of the number of translators. Really good technical translators are a scarce resource, and most companies are willing to hire them and sorry to loose them. The tools can help make the good translator more productive by handling routine work and by doing trivial things right every time an thus avoid rework.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Reduced cost", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Trained technical writers and translators with a good knowledge of the company terminology and style and of the products represent at the same time an asset and a liability for the company. The company is left very vulnerable if it should loose one of these key persons, because the training of new people requires not only money, but also time, which is the most critical factor. Consequently, the companies will want to document as much as possible of their company specific core knowledge in such a way that the knowledge can be reused by new translators and reduce their training time. A good way to document the knowledge is to use translation tools such as terminology databases and translation memories. ", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Documentation of knowledge", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Documentation of knowledge", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "The User's view: Why we do not use tools Just like there is a number of arguments describing why companies should use tools, there is an equal number of reasons argued by users explaining why they do not use tools. And similarly, the collection of reasons, explanations or even bad excuses can be boiled down to five basic reasons:", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Documentation of knowledge", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "\u2022 The human translator is better \u2022 The tools are too expensive \u2022 We don't have the time to look into this \u2022 The tools don't fit into our work flow \u2022 The tools don't fit into our technical architecture It is unarguable that these reasons, individually or collectively, clearly stated or hidden in some disguise, have stopped the purchase and implementation of translation technology in companies in Denmark. It is therefore not the intention of this presentation to attempt to prove them wrong and hope they will go away. These are all valid reasons and they represent real problems companies have had with translation tools. They should be taken seriously and worked with in order to understand if and how they can be overcome.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Documentation of knowledge", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "In the remaining part of this presentation, the intention is to examine each one of them in more detail. Usually things are more complex below the surface than upon first appearance, and frequently much is left unsaid and open to interpretation. In parallel with the examination, I will offer my interpretation of the motives that may lie behind them, and my recommendation on how they can be overcome.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Documentation of knowledge", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "The human translator is better ", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Documentation of knowledge", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "------------------------------------------------------------------------------------------------------------------------------------", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Documentation of knowledge", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "The human translator is better", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Documentation of knowledge", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "The users will argue that a human is a better translator than a computer programme, because the human is creative, can solve ambiguities, and can make the correct style decisions, plus a number of similar arguments. All these arguments focus on the language understanding capabilities in the human brain, which are so unique and which no computer programme is capable of matching. Anybody would agree with these arguments, but they are not as such sufficient reason for not using translation tools.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Human skill", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "A computer programme cannot replace a human in all his or her functions as a translator, but it can perform better in some of the functions, so why not let it handle EAMT Workshop, Copenhagen, May 1997 those and let the translator concentrate on functions which cannot be taken over by the computer?", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Computer strength", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Much technical documentation is characterized by repetitions. It can hardly be intellectually challenging to the creative mind of a skilled translator to repeatedly have to translate identical sentences and expressions like e.g. 'Press Fl for help'. But the machine is very patient with these repetitions, it can go on for a long time, and it delivers a correct result every time. The same phenomenon can be seen for terminology. It takes good subject specific knowledge and linguistic creativity to decide on the best translation for some term, but once the translation of a term is researched, documented and stored in the terminology database, the computer programme can take over and deliver consistently correct translations of the term throughout large documents and whole projects.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Computer strength", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "As stated earlier, translation time is critical and introduces the need for translation projects or even documents to be shared among several translators. This requires a very well developed coordination effort in order to maintain terminological and stylistic consistency. Coordination of large translation projects is made easier with translation tools.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Computer strength", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Fear and insecurity It is very likely that some of the prejudice against translation tools is motivated by the translators' fear of being replaced by a computer programme, and by insecurity towards new work processes based on a technology which to many traditionally educated translators is still new and something they do not feel comfortable with because they do not master it.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Computer strength", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "For their own sake, the translators must overcome their fear and take control over the tools and the related processes rather than let themselves be controlled by them. The tools will become more and more integrated into the work processes and there will be a demand for skilled translators who are able to combine their linguistic skills with technical skills with regard to the tools.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Computer strength", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "The tools are too expensive", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Computer strength", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "\u2022 They cost more than a word processor \u2022 We can't even get decent dictionaries Product prices are a frequently used argument against the purchase of translation tools. Many people seem to think that translation tools should cost more or less the same as word processors, and seeing the difficulty they have in being allowed to buy proper professional dictionaries, consider it impossible to ask for a translation tool, which is much more expensive.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Computer strength", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Another cost issue is the manpower it takes to implement the tools. It is argued that the translation function has too few resources to even meet their deadlines, and therefore they do not have the time to spend on implementing new tools and processes.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Computer strength", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "The consultant would argue that not product prices but lack of management attention is the real issue here. Language issues traditionally have a low priority with management, and since translators are usually not familiar with strategic IT planning and purchase issues they view the subject of price from an almost private point of view. This turns \u00a32,000 or DM 5,000 or DKK 20,000 into a high price.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Lack of management attention", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Language technology tools are professional tools and professional companies must invest in core technology for their business to stay in business. In many business areas professional tools are considered an indispensable investment, and it is fairly easy to list examples of such tools that all cost more than a word processor and all of which are sold much more than language technology tools: The purchase of translation tools is a professional business decision and it should be part of the company's IT strategy, based on relevant return on investment goals and on relevant cost benefit analysis.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Lack of management attention", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "The professionals who see the advantages of the tools must learn to deliver business arguments rather than technical or linguistic arguments to their management in order to raise the level of awareness towards the purchase of translation tools.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Lack of management attention", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "We don't have the time to look into this", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Lack of management attention", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Being busy all the time and maybe not always be able to meet deadlines anywaywhat better argument could you want in order to justify the introduction of tools to help you?", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Too busy", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "To be too busy is a very common argument used to avoid changes, but in reality companies are constantly seeking ways to optimize the production process and reduce the costs, and resources are allocated to these activities because it is necessary in order for the company to stay competitive. The real issue is more likely discomfort with new tools, processes, workflow, and responsibilities.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Too busy", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Many people argue that the tools are not good enough. This is a problematic argument because it seems to be based on the false assumption that one day the tools will be so perfect that they can replace the translator and that only then should the company start taking an interest in them. But the point is that although not perfect, the tools are excellent support tools that can make the employees more productive by taking over some of the more tedious tasks and they only become better by being used and by users requiring improvements and new functionalities.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "The perfect tool", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "The simplest and most obvious tool to this end is a terminology database, and it is most surprising that there are not lots of terminology projects going on in Danish companies.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "The perfect tool", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Again, the issue here is lack of management attention. Many companies invest large sums of money these years in order to implement quality management systems in their production line and obtain certification. Since language issues traditionally have a low priority with management, there is a risk that the documentation process is left out of the quality manual. Development of strategic IT plans and quality manuals represent a good opportunity for the language workers in companies to ", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "The cure", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Finally we come to the more technical reservations regarding work flow and technical architectures. The arguments listed on the three foils above are all authentic.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Technical reservations", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "It seems to be a fact that a certain percentage of the world's translation volume is not yet ready for electronic tools. Time will remedy that.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Technical reservations", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "The implementation and use of new tools and methods will always require a changed behaviour and revised working methods. Changes are required both from the people who will actually be working with the new tools and from the work flow processes in their surroundings. Translators have been and often still are conservative or hesitant towards new methods and technologies, but if an organisation does not want to implement changes, then most likely it is not ready or mature for those changes, either because the business itself is declining or because the cost benefit ratio is not sufficiently attractive.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Technical reservations", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Summing up the above findings and recommendations, translation is a management issue as much as it is a language issue. Until that is acknowledged, the penetration of translation tools into the market will continue to be relatively low.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Conclusion", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "EAMT Workshop, Copenhagen, May 1997", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "EAMT Workshop, Copenhagen, 21. May 1997 __________", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "", | |
| "sec_num": null | |
| } | |
| ], | |
| "back_matter": [ | |
| { | |
| "text": "We are very busy We are behind schedule We don't have the resources to look into this We want to wait until the tools are better Lack of management attention and awareness Need for IT policy Fear of new technology Need for education", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "We don't have the time to look into this", | |
| "sec_num": null | |
| } | |
| ], | |
| "bib_entries": {}, | |
| "ref_entries": { | |
| "FIGREF1": { | |
| "type_str": "figure", | |
| "uris": null, | |
| "num": null, | |
| "text": "The Consultant: Why use tools?\u2022 Better quality through better consistency\u2022 Reduced time to market \u2022 Reduced cost due to less rework \u2022 Accumulation of knowledge in the company 1997 EAMT Workshop, Copenhagen, 21. May 1997 ________________ The Consultant's view: Why should companies use tools?" | |
| }, | |
| "FIGREF2": { | |
| "type_str": "figure", | |
| "uris": null, | |
| "num": null, | |
| "text": "The User: Why we don't use tools\u2022 The human translator is better \u2022 The tools are too expensive \u2022 We don't have the time to look into this \u2022 The tools don't fit into our work flow \u2022 The tools don't fit into our technical architecture 1997 EAMT Workshop, Copenhagen. 21 May 1997" | |
| }, | |
| "FIGREF3": { | |
| "type_str": "figure", | |
| "uris": null, | |
| "num": null, | |
| "text": "Humans are more skilled translators than machines \u2022 Humans are creative \u2022 Humans can disambiguate and make style decisions \u2022 Machines have more endurance than humans \u2022 Patience with repetitions -'Press F1 for help' \u2022 Consistent terminology -all day and night \u2022 Translation in groups -coordination becomes easier \u2022 Prejudice against technology \u2022 Be in control of the tools -not controlled by the tools 1997 EAMT Workshop, Copenhagen, 21. May 1997 s" | |
| }, | |
| "FIGREF4": { | |
| "type_str": "figure", | |
| "uris": null, | |
| "num": null, | |
| "text": "We do not have the resources to implement them \u2022 Lack of management attention and awareness \u2022 Lack of cost -benefit analysis \u2022 Lack of professionalism \u2022 Lack of business arguments from translators 1997 EAMT Workshop, Copenhagen, 21. May 1997 -------------------------" | |
| } | |
| } | |
| } | |
| } |