he traditional view of a translator's work environment has been that of a person who works alone, fenced behind a stack of dictionaries, paper, and all sorts of documentation material. Not that this has ever been completely true, but it is certainly a picture surviving in the minds of some outsiders to the profession. As a service, translation has customarily been postponed until the last phases of production, when all other aspects of product development have been completed. Translation has inevitably been associated with the end of the process, either as an update of a manual or the manual itself, as a letter to a client, a best-seller to be introduced to a new cultural community, a contract establishing legal bounds between two parties, or the marketing material of an international corporation.
For some time now, we have been witnessing a gradual momentum which is remodelling the translation profession and transforming it from service into industry, that is to say, an economic activity concerned with the output of a specific product and which manages complex processes, develops synchronized goods, and which is based on team work, leverage of existing material (corpus), and information technology. In this sense, the translation industry is pointing to a new set of skills translators have to deploy if they are to act in a context of larger translation volumes, faster delivery times, stricter customization demands, and global production teams. The translator needs to develop the expertise of a project manager, a computer scientist, a documentalist, a DTP specialist, a terminologist, a language engineer, an evaluator, a localizer, and a technical writer.
Project management is about coordination, teamwork, planning, and control techniques
In this complex paradigm and with so many factors interacting in the translation process, project management emerges as the key element marrying crafts, needs, and expertise. Translation teams share data, clients, targets, projects, people, and resources, and it is the task of project managers to plan, instruct, monitor, and control large amounts of data quickly and accurately, while facilitating the problem-solving and decision-making process (Burke, 1).
The present article is an attempt at introducing an analysis of how translation can benefit from, and is actually taking advantage of, project management techniques and quality control processes. I will first advance some key concepts in the discipline and will then show current implementations of the procedures in the translation industry.
Project management: the basics
Anything we do has a life cycle: a beginning, a middle, and an end. Some life cycles are more complex than others because some tasks are more complex than others, too, and therefore require careful planning and attention. This is common sense and, at first, it might look as if project management were no more than applying a certain amount of wit and discernment to our daily work. Especially in translation, this might seem to be the case since the practice of project management has just started to emerge in recent years and does not yet have the long tradition it has in other fields.
Project management is about coordination, teamwork, planning, and control techniques. Any activity mankind has embarked on in the course of times has definitely required somebody managing it, either in the construction of the pyramids in Egypt or the Great Wall of China (Burke, 11). In the field of translation, I won't go as far back in time to stress the point and would rather concentrate on the recent changes the profession is experiencing which definitely affect modes of work and operational environments. On the one hand, there is no doubt that the translation market is facing a dramatic increase as the figures demonstrate: the worldwide market for translation and localization is about to reach US$20 billion in 2004; Microsoft executes over 1,000 localization projects a year with a revenue in fiscal 1998 of US 5$ billion; there are about 100,000 people translating professionally in Western Europe, and 317,000 full-time and part-time translators worldwide (Sprung, ix). On the other hand, the working environment of the translator is evolving towards decentralized global teams where technology is at the core of the process. In any case, it is certainly true that project management has lately gained a name in the translation profession due, mainly, to market growth and virtual teams. When translation is subcontracted to teams communicating through the Internet, productivity becomes the focus and it is in the area of planning, tracking, and measuring for volume and quality where project management offers essential tools for translation providers (Devaux, 12).
The Project Management Institute, in its Guide to the Project Management Body of Knowledge (PMBOK) defines a project as "... a temporary endeavour undertaken to create a unique product or service. Temporary means that every project has a definite end. Unique means that the product or service is different in some distinguishing way from all similar products or services" (1996: 4). Accordingly, this definition can be easily adapted to translation projects since the key elements, temporariness and uniqueness, are essential aspects in any translation assignment. You are assigned a translation, you agree on a deadline, the material to be translated is unique in the sense that it meets a precise series of criteria which make the translation product distinct from any other and which constitute the basis of any decision taken at each step of the translation process.
At the same time, there is a critical feature that needs to be added to this definition of a translation project, and that is leveraging. While each and every translation assignment is unique and has a definite end, it is nevertheless associated, to a lesser or greater extent, to other translation projects we handled in the past or that we might do in the future, either because we are working for the same client on different updates of the same product or because we are specializing in a particular field. In this sense, projects for the professional translator have a cumulative effect, and there is no doubt that a pool of previously translated materials provides an invaluable source of knowledge and expertise from where suitable translations can be retrieved, thus contributing to the project's efficiency in terms of speed and quality.
In this sense, and as we will see in the following sections, the appropriate leverage of translation data is of key importance to project management to the extent that in some contexts we can even see translation as part of an InfoCycle (Lockwood, 1998). This is particularly true in the sectors of software publishing, computer hardware, telecommunications, and automotive industry, where information is conceived within a coherent environment, a cycle which moves from development, delivery, consumption, and update, and back again to development. Under these circumstances, translation is part of the information workflow, an essential part, and no longer isolated from the rest of the process.
Translation workflow and Operational Environment
Current trends in the language industry indicate that the new methods implemented in information management are pervading the translation world. In this respect, the field of information engineering is already directing its procedures towards managing interdependencies, since the complexity of producing documentation impels us "to abandon the silo perspective of product development, marketing communication, technical writing, translation, and product support" (Hofmann and Mehnert, 60). A translation project's life-cycle must comply, then, with the demands of a wider production environment which considers multilingual information management in the form of information objects (IO). According to Hofmann and Mehnert, the IO is a collection of information identified as a unit, and defined by its communicative purpose, the specific user it is addressed to, the business entity it represents (a product line or a corporate function), the information it provides (in a specific format and for a target audience), and some publishing restrictions (61).
Assuming then that translation is an essential part in the information cycle that deals with information objects, I will concentrate now on the segment of the cycle concerning translation, as described in figure 1.
A translation project's life-cycle begins with a Request for Quotation and commissioning. At this point, the project manager receives a bid for translating a particular item (some advertising material, a user's handbook, an instruction leaflet, a web site, etc.), and the immediate action is to define client's needs, evaluate internal and external constraints that may affect the project's life cycle, and evaluate alternatives and options. This is usually known as the feasibility study and its main purpose is to gather information on all parties interested in the project--the stakeholders--so that the project manager has a clear idea of the environment where the project will evolve and how those affect the latter. Stakeholders are your client's requirements1, your own company's organization structure, market requirements, competitors, new technology, rules, and regulations or the economic cycle, to name but a few. These have to be adequately handled if the translation is to be satisfactorily completed and delivered.
In the process of preparing the feasibility study, it is the project manager's task to inquire about the following:
- translation's function
- translation's environment
- translation's working life
- standards and format specifications
- reliability requirements
- future expansion
- technology to be used
- company's resources
- quality assurance requirement (is ISO 9000 a requirement?)
- team availability
- office space and equipment available
- deadlines and possibility of meeting them
- terms and conditions, as specified by the client, and subsequent assessment of their practicality
- training required and feasibility in terms of money, people and expectations
- outsourcing needs and logistic constraints
- viability of other translation projects running in parallel
- quotation by the word or by the hour. And if you charge by word, make sure you really count all words in the project!2
With this information in hand, it is advisable to evaluate possible alternatives and options in terms of time (can the project be completed quicker?), cost (can the budget be reduced?), quality (can the quality be reduced and still be acceptable to the client, but more cost effective and quicker to produce?) and resources (what parts of the project can be automated?) (Burke, 40-46).
Once project and stakeholder's requirements are formalised, the next stage in a translation project life-cycle is planning. An essential aspect that needs to be kept in mind at all times is that the project is a whole, from conception to delivery, and that all interrelated activities leading to completing the project must be perfectly identified and carefully planned, to the extent that later, during project development, everyone participating considers the previous and subsequent steps as part of the overall workflow. Since the project has a specific start and a finish, a budget, and a set of tasks to be performed by a number of people in a certain time period using different resources, the project manager has to devise a coherent plan where all these items are adequately coordinated.
One of the first steps in planning is to identify key objectives and decide how to achieve them. The best way to proceed is by breaking up the path leading to the objectives into work packages (WP) and from these, decide who is responsible for performing each, assign lines of responsibility, level of authority, and lines of communication inside and outside the WP. Each WP has other elements, aside from people, associated to it: work calendars, resources, key dates, budget, quality control, and reporting frequency. Finally, each WP produces a deliverable (a subject specific glossary, the alignment of two files, a draft version of the translation) which is either the input to the next WP or contributes directly to one of the key objectives. A carefully devised plan will allow for the efficient completion of the project and will also facilitate the manager's job if any readjustments are needed in terms of time, people or budget.
Ideally, we would like our clients to understand that translation is a critical aspect in their product development, in the sense that when products are conceived for a global market and translation is placed at the core of the production line, many problems that may arise in later stages of the product life (production, distribution, update, marketing) can be easily foreseen so that solutions are planned in advance. Take, for instance, the case of web sites. The complexity of keeping web sites conveniently updated in different languages is currently imposing a challenge to many companies that did not plan how this task was to be achieved. A single translator cannot usually keep track of the different changes the site might experience at a particular time since different actors might be involved in the translation process (the project manager, the translator, the editor, the engineer, the quality assurance manager). In any case, it is important to stress the fact that only with careful planning can the project manager envisage any needs the client might have in the future, even if the client does not know s/he might have them. After all, is it so rare that one of our clients comes back to us with a new version of an old job demanding a quick translation?
With good project management, scalability should be considered so that the translation does not have to be rebuilt every time our client calls. The elements to be considered are, for example, new formats for old translations, new communication means with the client, new standards to be applied, new technology to be used in the translation process, new team distribution (some of the translators may choose to work freelance or in-house). Only with an efficient plan incorporating a flexible structure and a suitable operational environment can the translation project be successfully carried out.
The operational environment
Much has been written on the tools and applications translators use on their daily work. In fact, the notion of the translator's workstation was first conceived by M. Kay in 1980 (Kay, 1997). We have witnessed many changes since then, and many more still remain to be seen, but the basic idea is still the same: technology at the hands of the translator and at each step of the translation process.
The link between technology and project management is of key importance to translation project management, since only with a set of good tools can the translation industry respond to market requirements. In any case, it is important to remember how the human factor plays a critical role in the whole process since "the problem in translation occurs when technology is overvalued and too much reliance is placed on the machine-assisted part to the detriment of the translator" (Pellet, 55-56).
At the moment, the industry is using tools for term extraction, text segmentation, text alignment, indexing, translation storage and reuse, machine translation, consistency check, missing detection, and grammar check.3 And it uses them at different stages of the translation process: during groundwork, translation or tender. The relationship among all of them is shown in Figure 1. Obviously, different project characteristics would require different use of tools but, normally, any translation job will start with preparing materials and documentation so that the team can share resources and maintain linguistic and stylistic consistency thorough the translated version. Accordingly, one of the first tasks is to specify glossaries to be used, identify new terms to be included in the termbank and check translations for old terms (those already used in previous projects). Ideally, this should be an automatic process and it will imply comparing previous translations of the product, client-specific terminology, subject-specific terms, or product-related names. At the moment, there are a number of software packages for terminology extraction available in the market with different degrees of sophistication.
Together with term management, and in order to provide the text with its translational context, groundwork may involve processing previous source and target information, consulting any related documentation, checking any visual aids explaining how things described in the text work, the running application of the software to be translated, detailed product description. This information is key to the translation team since it directly affects the successful completion of the project and, therefore, needs to be available at any time in the easiest and most effective possible form. Whenever the information takes the form of previous source and target versions of the current text, it is advisable to feed it into the translational databases, so that a translation memory program can later take full advantage of it. Most translation memory packages include a tool for text segmentation and alignment. The basic idea behind alignment is to match a source language segment with its corresponding target language so that the matching pairs can be fed into the translational database. This is sometimes seen as a time-consuming task which keeps translators from doing actual translation. In my opinion, alignment is directly related to a key element in translation management, namely, reusing materials, and the more you invest in an effective leveraging process, the more efficient your project will be.
Other aspects of this phase may sometimes involve preparing texts for translation. In a localization project, for instance, this involves identifying any potential problems such as which strings in the code are translatable or which files are shared . In other projects this might involve deciding whether machine translation can or should be used for some part of the text as is, for instance, the case of the European Commission, where in 1996 alone 220,000 pages were run through Systran, "making its Translation Service (SdT) the most prolific user of MT in the world in terms of sheer volume" (Brace, 220).
The translation phase of the project is, obviously, the core part of the project and, at the same time, is only successful when the rest of the project phases have been adequately planned and satisfactorily completed. It is also a delicate phase from the project manager's point of view, since this is when resources are shared, information distributed among the team members, and communication workflow decided upon. Translation memories, or translational databases as they are now called, are currently the key tools. The availability of this tool is vast at the moment and the choice of one or the other depends mostly on your project's requirements. Quite often, this decision is not taken on objective grounds but is left to chance or subjective criteria (Rico, 2001).
TMs have come at last to fill this gap where, for the purposes of future projects, any given translation work used to be wasted. After hours of laborious and careful work, once a translation was finished and handed over to our client, we were able to save, at the most, a copy of the source and target text in different files, but with no efficient means of keeping a record of the actual links between both texts. While we could certainly create, if any, our own jerry-built databases for keeping track of our precious daily work, these were easily damaged or lacked the standard flexibility and dependability needed for language management.
In recent years and with the advent of TMs, the translation professional has finally found a tool which allows for these links between source and target text to be established, not only in the personal brain of each individual translator--which was, in the end, the only reliable place the whole translation process could be recorded--but in a robust and dynamic way which can accordingly be reused in later work. Memory-based translation has then come as a revolution in the translation market affecting the way translation is approached. Source and target texts are automatically coupled and matched as pairs of corresponding sentences and this material is easily retrieved in subsequent jobs. Nothing has to be translated more than once, terminology is better maintained, increasing, then, translation consistency and quality; reformatting of translated documents is almost no longer necessary (Hoffman and Mehnert, 68).
As for Machine translation, although for fifty years research teams have been attempting to mechanize this purely human activity of translation, the ultimate system capable of offering an instant and perfect translation is still far from becoming a reality. Computers' capabilities to understand language are the main barrier that has to be overcome, but, in the meantime, a practical and realistic attitude towards MT shows that automatic translation is already producing spectacular results in controlled environments where language is somewhat formalized so that computers can easily process it (Rico, 2001).
It is important to remember that MT applications should be selected after careful consideration of your translation needs. Toy systems that can be purchased off the shelf are just that, toys, and the tool you might want to use for automating your translations might need some customizing before you can actually implement it in your operational environment.
Finally, the wind-up stage of the project should be considered as important as the rest. No translation project is finished before sufficient time has been allocated for revision (consistency and grammar check), detection of any fragments missing and, in the case of localization projects, testing of the applications.
Is Quality Important?
Quality is about stakeholders' satisfaction, about work efficiency, about team cohesion, about control and communication techniques. Quality involves all aspects of project management and, in translation, affects both the process and the product. In this respect, L. Brunette (2000) in her article "A Comparison of Translation Quality Assessment Practices" provides a thorough definition of the different methods used for quality testing. She identifies five different procedures: (a) didactic revision, or formative revision, intended as a careful comparison of source and target texts with the aim of improving translator's skills; (b) translation quality assessment (TQA), related to management techniques and performed over a portion of the translated text with the aim of measuring productivity or quality according to a predefined checklist; (c) quality control, which is also an instrument for management purposes and ensures the compliance of the final translation, i.e translation as a product, with a set of requirements, norms, and criteria established in advance; (d) pragmatic revision, usually performed by an individual reviser who does not have contact with the translation and whose aim is to improve the final version; and (e) fresh look, which considers the translation as an independent text that has to conform to target readers' expectations.
These five assessment practices summarise the complex and broad field of translation quality management and offer a useful introduction to a much discussed area of quality definition for language services. In practice, quality assurance in the translation industry is usually identified with certifications such as DIN 2345, ISO 9002, UNI 10574 or SAE J2450 in the automotive industry. But still, general agreement on quality standards has yet to be reached since different clients expect different services from translation suppliers: "In one case, high quality is achieved by creating a translation which closely follows the source text and in another case, by completely rewriting the source" (Mehner, 13). In the end, quality definition for language services depends on client needs, which are always "multidimensional and often highly dynamic."
In this respect, the Localization Industry Standards Association (LISA) has developed a quality assurance model applicable to localization projects which originated from real-life quality metrics provided by Microsoft, Digital Equipment, Rank Xerox, IDOC Europe, DLS, and IBM (Koo and Kinds, 147). The LISA Model distinguishes between quality assurance, a series of checks on translation samples, and quality control, which implies a full review. According to the model, translations are given a Pass or Fail depending on the maximum number of error points allowed, as agreed by the client or service provider. Errors are classified on seven categories (mistranslation, accuracy, terminology, language, style, country, consistency) and they are graded as critical, major or minor.4
From the standpoint of quality as a deontological issue, the European Translation Platform has prepared the European Code of Practice for Translators, Interpreters and Translation Companies, a document summarizing the elements which make up best practice in these professions. Among the aspects covered, the Code mentions "qualification of translation providers," "contractual relationships," "responsibilities of the customer and the service provider," "delivery," "fees," and "copyright." This document is intended as a guide to clients on what to expect from translators, and to translators as a reference on how to meet those expectations. Quality as a specific issue in project management is not addressed, since the Code offers a broad and general perspective of translation but, in my opinion, it serves the purpose of an excellent starting point.
For some years now, translation seems to have converged towards the area of project management. The cause is certainly globalization. Some years ago, the idea of translators working in a team scattered around the world seemed science fiction, and language companies could only dream of launching their products simultaneously in different corners of the world. With the advent of new industry methods, new business strategies are called for. Similarly, new client demands call for new management strategies. Therefore, when time to market and global issues become an important factor the translation industry, failure to recognise project management aspects (identifying unnecessary expenditure, challenging assumptions, generating alternative ideas, promoting innovation, optimizing resources, saving time, money, and energy, simplifying methods and procedures, eliminating redundant items, updating standards, criteria, and objectives, among other things) can prove to be expensive.
I have tried to outline here, by way of an introduction, the general framework for translation project management. Those wishing to further their knowledge and skills in this area should bear in mind that project manager's tasks would typically involve coordination, scheduling, and status tracking, client contact, budget control, and resource management. As the single point of responsibility, a project manager should have the ability to anticipate problems, integrate stakeholders' expectations and needs, understand the operational environment, and cope with change (Burke, 9). In a word, if most of your translation team is in Europe, when holidays are traditionally taken in August, and this is exactly the time when your client in the US wants the translation job delivered, you had better plan for this low productivity period or the project would certainly be less than a success. To be fair, you can always count on Thanksgiving to New Year for a more relaxed period with this same client.5
Brace, C. (2000): "Language Automation at the European Commission," in Sprung, R. C. (ed).
Brunette, L. (2000): "A Comparison of Translation Quality Assessment Practices" in The Translator, vol. 6, 2: 169-182.
Burke, R. (1999): Project Management. Planning and Control Techniques. New York: John Wiley.
Devaux, S. A. (2000): "Getting a Grip" in Language International, April 2000: 12-13.
Esselink, B. (1998): A Practical Guide to Software Localization. Amsterdam: John Benjamins.
European Translation Platform. European Code of practice for Translators, Interpreters and Translation Companies, available at:
http://www.jtpunion.org/english/Translation_platform.htm (11 July 2002)
Hofmann, C. and T. Mehnert (2000): "Multilingual Information Management at Schneider Automation," in Sprung, R. C. (ed.): Translating into Success. Cutting-edge strategies for going multilingual in a global age. Amsterdam/Philadelphia: John Benjamins.
Kay, M. (1997): "The Proper Place of Men and Machines in Language Translation," Machine Translation, 12: 3-23.
Koo, S. L. and H. Kinds (2000): "A Quality-Assurance Model for Language Projects" in R. C. Sprung (ed.) (2000).
Lauscher, S. (2000): "Translation Quality Assessment. Where Can Theory and Practice Meet?" in The Translator,vol.6, no.2, 149-168.
Lockwood, R. (1998): "Are You Ahead of the Curve?," Language International, 10.2:24.
Melby, A.K. (1998): Eight Types of Translation Technology. Available at: http://www.ttt.org/technology/8types.pdf (11 July 2002)
Mehnert, T. (1998): "Quality Policy: Comply or Compete?" in Language International, 10.5: 12-15.
Pellet, Mercedes (2001): "Using Technology Wisely in Language Services" in Multilingual Computing and Technology, vol. 12, 4: 55-6.
Pinto Molina, M. (2000): "Implantación de los sistemas de calidad en los servicios de traducción" in Gonzalo García, C. y V. García Yebra (eds.), Documentación, terminología y traducción. Madrid: Síntesis.
Project Management Institute (PMI) (1996): A Guide to the Project Management Body of Knowledge (PMBOK).
Rico Pérez, C. (2001a):"From Novelty to Ubiquity: Computers and Translation at the Close of the Industrial Age" in Translation Journal. Available at: http://www.accurapid.com/journal/15index.html (11 July 2002)
Rico Pérez, C. (2001b):"Reproducible Models for CAT Tools Evaluation: A User-oriented Perspective," in Translating and the Computer 23, London: Aslib.
Freivalds, J. (2001) :"Calculating Return on Localization," Multilingual Computing and Technology, 12, 4.
Sprung, R. C. (2000) (ed.): Translating into Success. Cutting-edge strategies for going multilingual in a global age. Amsterdam/Philadelphia: John Benjamins.
Appendix: Tools for project management in translation
This section is intended as a link to the main tools for project management in translation. It does not aim at exhaustivity but rather at serving as a starting point for those wishing to research further in this area. I have included software tools as well as service vendors. For each of them, a short description is provided.
Description: They describe themselves as "the leading provider of comprehensive globalization solutions and technology." Their list of clients include Bayer Diagnostics, Oracle, IBM, Corell and Disney Interactive, among others. Their list of products and services range from content development and multilingual localization management to a series of software tools to control and keep track of translation projects.
URL: http://www.bowneglobal.com (11 July 2002)
Description: This tool is mainly intended for free-lance translators wishing to keep track of their translation job in a simple and non-expesive manner. Features include Client manager, invoicing, report center, glossary, scheduler, message pad, personal phone book, among others.
URL: http://www.landsw.com/translators.htm (11 July 2002)
Description: Idiom offers software and services for enterprise globalization. Among their customers we find Charles Schwab, General Electric and Travelocity.com. Their services are oriented towards leveraging existing infrastructure so as to "maximize global success." As for products, Idiom's WordlServerTM is Web infrastructure software "that offers the most advanced translation technology," integrating Déjà vu as its translation memory.
URL: http://www.idiomtech.com/ (11 July 2002)
- Knowledge Ability: Working by WireTM (White Papers)
Description: This is a privately owned UK company providing services on consulting and training on virtual work, knowledge management and networked learning. White Papers include "Managing Virtual Teams over the Internet," "Trust in Virtual Teams" and "Team Knowledge Management," amont others.
URL: http://www.knowab.co.uk/wbwpapers.html (11 July 2002)
Description: This is a service company offering translation, localization, electronic publishing and project management in 80 languages.
URL: http://www.linguaforum.com/content/index.html (11 July 2002)
Description: This is a "workflow control software system which supports and facilitates multilingual translation projects." It combines quick access to client and translator databases with a project management feature that monitors progress by recording and archiving every team member's input at each stage of the project.
URL: http://www.langtech.co.uk/eng/organiser/index.asp (11 July 2002)
Description: This is a free online resource for professionals in project management. It includes articles, software reviews, advice and career prospects.
URL: http://www.projectmagazine.com/ (11 July 2002)
- Project Management Institute
Description: Established in 1969 and headquartered outside Philadelphia, Pennsylvania USA, the Project Management Institute (PMI®) is the world's leading not-for-profit project management professional association, with over 86,000 members worldwide.
URL: http://www.pmi.org/ (11 July 2002)
Description: This is a localization services provider focused on Information Technology. They use their Streamliner™ localization methodology and since 1997 they have been using using their StreamNET™ Extranet system to assist in the management of localization projects. This "architecture provides full scalability and flexibility so that Extranets can be highly customized in response to customer needs."
URL: http://www.rubric.com/index.html (11 July 2002)
Description: This is a company currently offering, among other translation software, a tool specifically designed for multilingual content management: SDLWebFlow ("a content management system for maintaining multilingual Websites requiring frequent updates").
URL: http://www.sdlintl.com/enterprise-systems/sdlwebflow.htm (11 July 2002)
Description: Translation Workflow Server provides an "enhanced, scalable and integrated solution which can operate completely automatically or semi-automatically (through interaction with project manager)."
URL: http://www.star-ag.ch/eng/home.html (11 July 2002)
- Trados Corporate Translation Solutions
Description: This well-known company groups a series of tools for translation management: translation memory, language management, front-ends, legacy translation mining, translation production and update management.
URL: http://www.trados.com (11 July 2002)
1 The European Translation Platform (ETP) has issued a working document defining client-translator contractual relations, where the list of client's responsibilities can serve as the starting point to define the client's requirements. By asking the client to provide a set of materials as well as information on the product, the project manager can get a clear idea of what it is that the client needs. The ETP document is available at http://www.jtpunion.org/english/Translation_platform.htm (11 July 2002). On the other hand, the ITI's document "Translation--getting it right," available at http://www.iti.org.uk/pages/news.cgi (11 July 2002), is recommended to anyone wishing to commission a translation since it provides the rules of thumb for getting a translation that works.
2 Before charging your customer, there are a series of aspects you should consider. For example, counting words may turn a difficult, long task, if you are translating a web site. It may contain graphic files, words in tags or in embedded scripts. Similarly, you should consider whether you are charging only translation or translation and verification of the running application. For an account of counting tools, see the section Newcomers/Payment practices at TransRef (http://www.transref.org) (11 July 2002). This is a portal for translators interested in translation technology, professionally set up, maintained, and updated.
3 For a detailed account of CAT tools see TransRef (http://www.transref.org) .
4 For a detailed account of LISA QA Model see http://www.lisa.org/products/lisapub.html (11 July 2002).
5 This is an example adapted from a piece of advice from Devaux (2000, 13).