I use DVX and can freely combine segments across line breaks and then At one time at least, it was difficult to do. Splitting segments in OmegaT, but I do not think it has been Version on this score.) There has been a loud demand for joining and Older versions of Trados allowed you to expand segments to a certainĮxtent, but there were limitations. Highlight and open a segment of any length or expand to what you want. To join and split segments (DVX, MemoQ and others. Jon Johanning // best solution to this problem is to use a CAT tool that allows you Require Trados and I can get out of these jobs that way), but there isĪn agency I work for who will start giving me jobs using another CAT appĪt some point, and I will see what happens with their stuff when it arrives. So far I haven't done any CAT jobs sent from agencies (usually they Obviously one has to use some judgment about what jobs are worth Up the text file from the OCR to some extent before I feed it to theĬAT. What I do with the OCR'ed pdfs I often deal with is to fix Looks very similar to the procedure I have arrived at (I use // as a > the linguistic elements within the source and target segments will not > effect creeps in and you will lose much of the benefit of the TM since > If you don't do any reshaping of the source text, then the GIGO > their proper location in the final English version. > (allowing the claim sentence to flow freely), and reintroduced in These can be moved to the end leaving a marker like 化1 > The same goes for structures or figures that appear within a > hard returns where they make sense to mimic the structure of the > re-stitching together during my final proof, and then reintroduce the > and reformulate the text to make sense in English. > bookmark at positions where I know that I will later need to come back > symbol (I use red double asterisks, for no special reason) as a Another is to put in some easily recognized > many/most of the hard returns within a single claim, which improves > I use Trados, and get by with a few tricks. > last moji string in the Japanese version. > the first word/phrase in the English version will correspond to the It is particularly troublesome in the claims. > In my experience, this is a fact of life when doing JA>EN patent On 6/22/10 1:09 PM, Matthew Schlecht wrote: If you don't do any reshaping of the source text, then the GIGO effect creeps in and you will lose much of the benefit of the TM since the linguistic elements within the source and target segments will not really be equivalent. These can be moved to the end leaving a marker like 化1 (allowing the claim sentence to flow freely), and reintroduced in their proper location in the final English version. The same goes for structures or figures that appear within a claim. I do this re-stitching together during my final proof, and then reintroduce the hard returns where they make sense to mimic the structure of the source claim. Another is to put in some easily recognized symbol (I use red double asterisks, for no special reason) as a bookmark at positions where I know that I will later need to come back and reformulate the text to make sense in English. One is to remove many/most of the hard returns within a single claim, which improves the logical flow a bit. I use Trados, and get by with a few tricks. Usually, the first word/phrase in the English version will correspond to the last moji string in the Japanese version. In my experience, this is a fact of life when doing JA>EN patent translations.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |