Transit PE Instructions Help

there is no need to zip the files, as .txf and .pxf are actually zipped files. Zipping them further will not make them smaller, and may possibly lead to complications.

I received your contact from Andrey and was wondering if you'd be into helping me out with translation memory. I have one girl in Australia who seems quite an expert, but she is pretty busy and doesn't always answer me. Right now I am using Transit, know the program fairly well, and would like to use it for all our larger translations, if applicable. When I get the budget, I'd even like to develop my own software and create a sort of wikitranslations, but charge a very modest amount. No burning issue at the moment, but just wanted to establish contact, as this subject interests me. The only problem I have at the moment is I am bidding on work for Panasonic and they like to use InDesign and FrameMaker, which require rather expensive filters for both Transit and Trados. I prefer to use Transit, since it does not force the translator to buy the expensive program, and it is supposed to be better at importing and exporting from these two programs. Also, thought I might be able to work around this by importing through PDF files, but was surprised to find out that Transit does not offer this import option. Would enjoy hearing from you.

I've worked for customers in the past who worked on Mac and it was a bit of a pain, but possible. Usually we'd convert things through Word. Anyway, I don't know what programs Panasonic uses but we'll figure this all out if I win the tender. Right now I am submitting samples to them, and if they like our quality and price, we can move towards resolving the rest. But it does not hurt to discuss these things in the meantime, as I realise it can take a long time to resolve such complicated matters. Do you have Transit? Are you able to get a demo filter so that we can at least experiment with the import before considering such an expensive purchase? I'd prefer that than paying someone every time to piddle around with each page and manual dtp work. Not economical or competitive. But needs testing for sure.

• Karel, you need to get more specific information about USED PROGRAMS, VERSIONS, OPERATING SYSTEM from Panasonic, and if they use MAC, then we need to make a test, if all be OK, then we could work) 2. You can get filter and use Transit, but you would have the same problems, which we had: - You will pay to translator for unneeded symbols (numbers, commas, points and other), sometime it is very big money. - Filters, if you have not them, will be cost too highly - There is no warranty that filter will work properly (with all versions of document) - If your TRANSLATORS are not experienced, then they can to broke some TAGS - Special symbols (DIAMETER, TRADEMARK and etc.) will be replaced at IMPORT - After TRANSIT the files need to be make-uped anyway (because not all text will have its previous formatting) 3. You no need in buying filters. 4. I suggest you to recognize originals to simple format (Word, RTF or HTML), and then translate it in TRANSIT or TRADOS or using any other Translation Memory to any number of languages. After that, it need to restore text into files, it is not difficult. You can send me a test page of any document to try... Best regards, Eugene FreeTeam

sorry, was flooded with a few projects and the person I asked to do the Chinese sample did not answer. Can you do it now? By when can you get it done? The OCRed Word files are messy indeed. Can you ignore it and just translate directly into Word from the pdf file? I have to buy some expensive filters to go straight from InDesign etc. into Transit, but I'll worry about these technical issues if/once I win the tender. Panasonic understands this and just wants to see the sample for now.

The standard is in Word, through the Properties Statistics section as the RW instructions should explain. But if a company wants to use Transit/TM, it must be calculated based on fuzzy matches and what Transit/Trados says. For example, for the Dutch project, the customer has Trados and we agreed on a rate based on their fuzzy/repetitions analysis. Since I used Transit and the statistics are slightly different, I paid the translator based on that. An ideal situation, which is a new idea I am rather proud about, is the customer has no clue what TM is, I charge them based on the final Word export, and pay the translator based on fuzzies. But to be a nice guy, I would give the translator some percentage share of the extra bonus. And I might charge the customer a little bit less just to be more competitive. I will have to play this by ear. But I want the translator to be happy using Transit, and create more juicy profits for myself (which would be shared among the project managers as well). This way I could try to use Transit all the time, and build up a massive translation memory, such that it might be possible one day that 80% of the translation is repeated from previous translations, meaning I only have to hire a translator to translate 20% of it, plus pay them something for double checking etc. So even more juicy profits and/or competitive prices offered to the customer. And perhaps I already mentioned this, but once the TM gets large enough and I got some spare cash to invest (I always seem to have many ideas I want to develop), I get a programmer to develop wikitranslation, so that customers or translators can log in and pay for our previous translation memory. Some modest rate would be charged, like perhaps 3 cents a word. But it would be all automated and I wouldn't have to do piddly squat. Of course, if the customer wants the remaining 20% of their text translated, we can help them with that. Or just manage the entire project for them. Just some more long term thinking.

it seems strange that none of these program can do this. After all, the text is accessible in the pdf files if it is not only an image file. Anywho, I had a question concerning another project. I have a customer who is doing some work for Kodak or something, and their task is only to proofread translations. Like they are a second vendor, where the first vendor is responsible for translations, and Kodak wants to divide it up for accountability purposes, so if someone wants to take them to court because their medical translation are faulty, they can say, "Hey man, I did my best and followed all the rules, and used two different sources, one for the translation, another for the proofread, so you can't sue me man. I did my best." Well, the one proofread we did found faulty translations into Czech, but whatever. If some Czech gets sick on some drug, it´s not on my back. But there is supposed to be a lot of repeats. So obviously it would be better to do this all through one vender using Transit, but this is not the case. What I am wondering is if it would be possible to somehow create a proofreading project, where the TM would remember when something has already been proofread, so that the translator would skip over it in the usual ALT INS manner. Unfortunately, the files they are sending me are like the one I attached. I tried to create a project Turkish to Turkish, hoping it would somehow recognise and filter out the English, but it did not let me choose the same language like this. My first thoughts was to choose two languages which use the same fonts/characters, and proceed as normal, except the translator wouldn't need to refer to the source file, because the source text is already in the target file. I'd rather avoid getting all complicated and time consuming by trying to create alignment projects from these tables. The customer is paying a very small amount. Alternatively, I can ask the customer to ask its customer for a TMX export and work somehow with those. The customer says that there are a lot of repetitions and is asking for a "light" proofread. Like the translator is supposed to assume that much of the files are repeats, have already been proofread, so are "probably good quality" requiring only a "light" proofread. Well, as you can imagine, I feel this is a rather retarded approach and am trying to figure out a more intelligent strategy.

it seems I have already erased the originals so we will not be able to compare the two, but I fiddled around with Trados in the past and it seems it was not very good at creating the Repetitions file. Do you have experience with it? For analysis only it seemed fine but not for the repetitions file. Perhaps you could try yourself. If you cannot learn it, then please learn it for Wordfast so that you are ready when a job comes. What version of Trados do you have? Do you have a paid or a cracked version?

you must have thought I disappeared. Anyway, still working on your samples. Everything is translated and some already proofread, and vetted, so it is near finalisation. I've been swamped with various projects so I've been juggling, and since you said your projects will not start end of January, I focused on active projects. But I feel pretty confident with the quality and I think you will be happy with the quality and prices. I have several good people in each language. I've asked a few more Arab translators concerning the above. On another note, I was wondering if you wanted to start toying around with the technical end of things. I've been talking with several people, and my Transit expert, and she says that Trados requires the purchase of one filter for either InDesign or FrameMaker, where Transit requires two expensive filters: one for each. But apparently Transit is better because Trados has some quirks with the importing and exporting, and requires extra work at the end with DTP manipulation - especially with certain languages. Another possibility is a TMX export from Trados, which I can import into Transit and prepare it for the translators in a free Transit program. Some translators already have Trados, while for others I may be able to provide them with a version I found. All this requires some experimentation, so I was wondering if you wanted to started playing with that. You should have your samples soon.

I'd like to do this through Transit, which means first the .pxf file, and then use that to translate the rest. Can we do it that way? 39% of the document is self repeating, so I'd like to use a translation memory tool. You should make more for your time than if translating from scratch.

okay, then please learn it. It is not complicated. I would like all my projects to be run through Transit, although it can import/export tmx files with Trados, so I would also like to learn about the possibility of using both programs together, as I know many translators with Trados do not like learning a new tool. And I wish you'd turn off your confirmation. It is really annoying to always get another email from you informing me that you will get to that email as soon as possible. It is really redundant and annoying, and should only serve a purpose if you are going away on a vacation or something. We are all bombarded by emails, and if we will work together in some future projects, I will demand that you turn that off. I get too many emails as it is.

I've been told it is similar to Trados. Anyway, I'd eventually like to translate all languages and certain projects through Transit. I can give you a percentage savings of the repetitions on top of the usual fuzzy matches to make it more interesting for you. I am offering this to all translators. I don't think it will be difficult for you to learn.

I first need you to do the Repetitions file, and then I can send you the rest of the project. The Repetitions file is rather small, so it would be good if you could complete it by Monday earlyish. Then you could do the rest of the 1700 words at ease. Try to make sure the Repetitions file is good and without problems, to avoid making repeated changes later. About 40% of this last project has repetitions.

I've researched this extensively. The free version of Transit you would be using does not get upgraded, but only the paid version I have, which is far superior to Trados. I have been discussing this with many people for ages.

I've researched this extensively. The free version of Transit you would be using does not get upgraded, but only the paid version I have, which is far superior to Trados. I have been discussing this with many people for ages. It would be good if you could do the Repetitions file first, in Transit, so that I can prepare you the rest of the Transit files while you are translating the Word files.

there was nothing attached to this email. My other email should explain what I need from you now: two Word file translation, one Transit repetitions file. I will send you two more Transit files once they are translated into English.

just received your translation and the first thing I noticed was that you added entire sentences to the dictionary. This is not the way the program works. A dictionary has individual words, sometimes collections of words, but never more than that. The program itself remembers entire sentences and phrases.

I'm learning with this as well. Usually I would not be translating through another language like this, meaning I could have sent you the originals so that you could have studied them and understood the context during your translation of the Repetitions file. But the customer is a bit stressed for time, so I wanted to move forward with all this. To save you time with repetitive text, perhaps you could go back to the Repetitions file and make the appropriate changes, now that you understand the context better. Then I can recreate the new .pxf files for you. I will leave it up to you, and take a share out of the savings from the repetitions to make it worth your while (the extra work etc.). After you finish all the .pxf files, they will be exported to Word and THEN proofread, meaning everything will be in context during the proofread (and spellchecker used).

40% of the file is repeats, so there is no point stopping at 100% matches. If that were the case, we might as well have translated everything in Word. I'd like to take advantage of the repetitions, for both of us to earn more, in which case I think it would be better to upgrade the Repetitions file. Or if it won't be too many changes, the proofreader can deal with it.

pretty well. Now that you understand the context better of the Repetitions segments, since 40% of the files are repeats, you might want to make changes to the Repetitions file so that I can recreate the .pxf files for the remainder of the two documents. I will leave this up to you. If there are not many changes, the proofreader can take care of the repeating updates, but it would be good to let her know what to look out for. Usually I do not translate through another language like this, so this case is exceptionally complicated, also because the customer is in a rush. But once you are done I think you will understand the process much better.

thanks for the file. Note that if you make changes to the Repetitions file, you must send that to me first, THEN I recreate the .pxf files of the rest of the documents, and ONLY THEN can you start working on the new ones I send you. The Repetitions segment are impregnated within the two .pxf files, and the Repetitions file by itself is nothing. So if you want to make changes to the Repetitions files, first send it to me and wait until I recreate the other files for you. Otherwise, if there are not too many changes required, you can always do a search and replace in Transit, or give instructions to the proofreader to do so in Word. But it would be better to finalise your translation.

I know this translation memory stuff can get a bit complicated, but you should understand it clearly once you've worked in it a bit. The Repetitions file is for you only. The proofreader will be working on the exported Word files, exported from Transit. 40% of the text of the two files to be translated are repeating, which is why I want to translat the repeating text first (Repetitions file). Once that is finalised, THEN I create the two files. When you press ALT INS and it goes to the end, it is because you have already pressed ALT INS on all those segments, marking them as Translated, and why it skips over them. So you have to just scroll through manually, or use Search and Replace, to update the Repetitions file Therefore, I am waiting for you for your instructions. There are two possibilities here: 1) you update the Repetitions file by going through it quickly, possibly using Search and Replace, send that finalised Repetitions file to me, I pump it into Transit, and send you back new .pxf files, which are the ones that actually need to be translated. Within those files, 40% would have already been translated and you will be able to ignore, because you correctly updated the Repetitions file. I am at the internet for the next hour, so I can quickly process the Repetitions file for you, but not for a long time after that. 2) You go ahead with the two translation files based on wrong repetitions. When ALT INSing, you will skip over these wrong repetitions. You can go back at the end with a big search and replace and try to fix this, or pass this onto the proofreader, who will be working in the Word exports, but this is generally not a great idea because the translator should be responsible for delivering close to a final product. But because of this special situation of translating through English and not having them translated yet for your context reference, it is creating these little glitches. So I will leave it up to you how you would like to approach this, but I will be here only about another hour, during which I can quickly process the Repetitions file. Which, by itself, will not be sent to the customer. It is only used for creating the two files to be translated (Auftrag AG.doc and Auftrag GmbH.doc - translated in Transit).

okay, leaving the internet now, so please just proceed and try to fix things with a global search and replace later. We'll figure this out somehow.

the paid version of Transit does this, and possibly the free version of Transit within the same file, but not across different files. But perhaps within the same file but different Language Pair files. Did you try it? I am slowly learning all the limitations myself. If it doesn't work you can send me the file you translated and I'll create the last file for you, with your previous translations in it. At some point I definitely want to start experimenting with import/export between Transit and Trados.

really? One translator using the free version told me it worked within the same file but not across different files. Anyway, next time we can try to combine it with Trados. I was just warned against it because the final export can get all screwed up, but I guess if we are dealing with simple Word files it should be okay.

> Here is the first of the two Transit files...

I don't know what you did but I cannot import your file. Simply Send the translation to produce a .txf file. Before you send it, you can temporarily change the .txf ending to .zip and unzip it to take a look at its contents to make sure your translation is in there. Usually this does not happen, so I do not know what you did.