How to prepare an index for translation……

url · 发表于 2006-4-7 14:22:43

（转一篇与索引相关的文章，有兴趣的可以看看。） How to prepare an index for translation from English to another language（zt） (Note to the translator: I have also some exhibits - printouts, transparencies, photocopies of indexes - which are not easy to translate or reproduce here. I will simply bring them to the conference, and they will be available for examination.) I am glad to have the opportunity to attend your conference. Since I have studied the Chinese language in school (a long time ago! I hasten to say) I have long wanted to know how indexing is done in your country. I hope I can share experiences with you. I am particularly interested in the reuse of indexes. A subset of this topic is translation of indexes. Let me give you some background on me. I have been a freelance indexer for most of my career - I have also been employed as a computer programmer. I have found a way to combine both fields - to do both indexing and programming. As a matter of fact, the task of indexing is a natural application of computer processing power and software logic. I have worked with various types of publishers, and in particular, textbook publishers. These publishers produce meticulously indexed books, which often come out in new editions. Because text is added, deleted, or moved, the page numbers of the old index will change. This results in a need to produce a new index. Since the work of indexing is so laborious, it is always desirable to try to reuse the index of the previous edition - easier said than done. If it can be done successfully, it is like standing on the shoulders of giants (the giants are the previous indexers, and we pigmies stand on their shoulders). Before beginning the reuse or translation of an old index, it is necessary to evaluate whether it is worth doing. One should perform various tests: • internal consistency checks (such as accuracy of mirror references) • spot checking for accuracy of spelling, locators, description, etc. • spot checking for presence of needed topics • determining that the index is balanced, and does not contain useless entries Realistically speaking, if the index is not at least 95% perfect, it is very difficult to reuse it. However, it may be possible to reuse the terms (the keystrokes) even if the page numbers or descriptions are unreliable. This can be successful if the specialized indexing program has a look-ahead feature (so that if I begin to type "ind" it can bring up "index", "indexer", "indexing" from which the indexer can choose.) There is no saving of money when reusing an index. However, there may be a saving of time, and a preservation of quality of the old index. How does one capture the data of an old index? • Before computers existed, the index would be photocopied and zoomed 400%. Then the poor indexer would go through the index, looking for each page number in turn: 1, 2, 3, etc., or more likely, the 10s, the 20s, the 30s, etc. • Usually the existing index can be scanned, but then it must be carefully proofread. • I have developed methods to ensure quality scanning. • Sometimes it is worthwhile to retype an index directly into a specialized indexing program. • A file may exist already. Usually this is only a word processing file, such as a Word DOC, which the editor may make available to the indexer. • Best of all is when a database file exists. Then the text can be searched, globally changed, sorted, etc. Some types of books use paragraph numbers (such as 1.123) instead of page numbers, and these paragraph numbers do not change when the book is published in a new edition. In this case, the index can simply be translated and rearranged. In other books, page numbers change, because text material has been added, deleted, or moved. In this case, it will be necessary to pull items one by one from the old index and place them in the new index. Here is an example screen shot of a program I have invented (called Deal'm, as in the dealing of playing cards). The indexer reads the new text carefully, determines what he wants to index and what term to use, and then looks in the old index to see if the term exists. This can be useful in ensuring that an old index is thoroughly inspected and reused (assuming it is a good index.) A translated index is simply an extreme case of reusing an index. After the words of the source language are translated to the target language, it is still necessary to analyze the data and make a database that can be searched, edited, sorted, and arranged (alphabetically or otherwise). Languages differ in their philosophies. It is conceivable that a translated index conveys a different impression than that of the original language, because of differences in vocabulary and syntax. This topic is worth exploring in the future. The arrangement of a Chinese language index is a problem, I feel. Usually it is arranged in pinyin order - but what if the reader does not know the pinyin? It can also be arranged in stroke order. Older works might be arranged in order of radicals (zi mu). * * * * Here is a description of a project that I worked on. The book was a Guide for Immigrants in New York City, published in English, Spanish, and Chinese (all in one volume). The publisher was The New York Times newspaper, together with the Immigrant Museum. Here is a step-by-step process: 1. I prepared an English-language index to the English-language book. This was edited and approved in the usual way. 2. I gave the index to a professional translation company in the form of a Word document (.doc). It looked like this: Absence from work, pay deductions for, 26 Abuse, physical and sexual, 69 Accountant, for taxes, 27 Address: change of, 86-87; proof of, 36, 42 etc. 3. I instructed the translator at the company to go through the Word document. In front of each English phrase, he should attach a Spanish translation, separated by punctuation. What the translator returned is this: Ausentismo laboral*Absence from work deducciones del cheque*pay deductions for, 26 Abuso*Abuse fisico y sexual*physical and sexual, 69 Contador*Accountant para los impuestos*for taxes, 27 Domicilio*Address cambio de*change of, 86-87 prueba de*proof of, 36, 42 etc. 4. Some lines don’t require a translation, for example: Banco Popular, 302 (The name of this bank is already in the Spanish language.) 5. When I get back the expanded document, I verify that the translation is correct, by comparing the English and Spanish (this means I need to understand Spanish well enough to read.) 6. Then I remove the English from the document (by an automatic process), and produce this: Domicilio cambio de, 86-87 prueba de, 36, 42 7. Then I analyze ('explode') the data. I arrange the references in page order, and change page numbers to those used in the Spanish text. Again this requires that I understand the Spanish language. I work with a printout that looks like this: 36 [this is a page number] Domicilio: prueba de Tarjetas de credito etc. 37 Tarjetas de debito etc 42 Domicilio: prueba de etc. 86-87 Domicilio:cambio de Etc. After repaginating, I re-alphabetize the entire index data in Spanish order, and edit. 8. I proofread the Spanish index against the Spanish text, in particular checking for correct pagination (since the Spanish translation usually requires more pages, the pagination can be different) 9. I re-alphabetize. I now have a Spanish index to the Spanish text. 10. After making further editorial changes, I prepare a Spanish index, and give it to the publisher as a Word document. This same index that I am discussing was also translated into Chinese. I will bring an example of this translated index to the conference. * * * In conclusion, the reuse and translation of indexes is, I believe, an interesting field that is sure to grow. I would be happy to talk with any of you in more depth. Thank you. * * * Mr. Peter Rooney American Society of Indexers magnetix@ix.netcom.com  <div>此帖由 url 在 2006-04-07 14:26 进行编辑...</div>

		自动登录	找回密码
密码			欢迎注册