I'm looking for folks who've first hand experience having documents translated into foreign languages. Said documents to include diagrams (think: callouts, legends), #included text, etc. I've a fair bit of experience with I18N/L10N for software but the extent of the effort, there, is usually fairly limited. And, there's less of a need for a cohesive approach as the interactions are "punctuated" (no pun intended). Recommendations for firms to do this? (no, finding multilingual "friends" to do same is far too unprofessional -- though they may have value in proofing the results) I suspect there is some value in having a single firm handle all of the translations (in the hope that they will create a consistent SET of translations, even if different individuals are involved for each) Relative effort? (i.e., closer to reading speed or writing speed?) Time frame? (is this effort-bound or business-bound) Cost? (and, "unit of measure"?) Finally, how to check the translation for accuracy and "feel" (i.e., ensuring it is true to the original intent)? With translations in hand, do you (thereafter) maintain individual documents? Or, merge them into a conditional document? Horror stories of attempts gone horribly wrong (i.e., what to avoid)? Thx!
Translation services/strategies/costs
Started by ●December 22, 2021
Reply by ●December 22, 20212021-12-22
On 22/12/2021 08:02, Don Y wrote:> I'm looking for folks who've first hand experience having > documents translated into foreign languages. Said documents > to include diagrams (think: callouts, legends), #included > text, etc.I have a friend who is a patent translator. They actually do into English from several European languages that they are fluent in. Advice is generally that the translator must be fluent in both (written rather than spoken) and *very* strong preference that they are a native speaker of the destination language in each case. I was involved in porting software into Japanese a long time ago and it was quite entertaining. The jump from 1 byte to 2 byte DBCS character representation made things even more fun back then. But the worst part was finding a technical translator that could understand domain specific terms and give them the right nuance. eg. "Dwell time" caused trouble. First pass it came out as "time in home" - there were lots of others. In Japan it wasn't uncommon to meet perfectly good English to Japanese translators who were just too shy to speak English even though they could understand it perfectly well and write it almost fluently. In some ways their knowledge of more complex English grammar was better than many native English speakers today. Correct usage of "I would be obliged if you could" vs "would" - there is a hidden insult in the first phrase which has now sort of been lost in common English usage.> I've a fair bit of experience with I18N/L10N for software > but the extent of the effort, there, is usually fairly limited. > And, there's less of a need for a cohesive approach as the > interactions are "punctuated" (no pun intended). > > Recommendations for firms to do this? (no, finding multilingual > "friends" to do same is far too unprofessional -- though they may > have value in proofing the results) I suspect there is some > value in having a single firm handle all of the translations > (in the hope that they will create a consistent SET of > translations, even if different individuals are involved for > each) > > Relative effort? (i.e., closer to reading speed or writing speed?)The effort involved is quite high and gets higher and more expensive the more important the documents are and the exact precision of technical translation required. Patent translation is about the most extreme. Many now use a hybrid machine translation human correction approach at least in the UK where person hours are expensive. It works well if you translate a lot of similar things with their own set of phrases (as happens quite a lot in patents).> > Time frame? (is this effort-bound or business-bound) > > Cost? (and, "unit of measure"?) > > Finally, how to check the translation for accuracy and "feel" > (i.e., ensuring it is true to the original intent)?These days machine translate back with eg Google and see if you can recognise anything of what you had originally written. You really need to have it all proof read by another native speaker of the language who is a domain expert in whatever the content is. That way you find most of the peculiar ambiguities that translation and punctuation can insert. "Eats shoots and leaves" for example.> > With translations in hand, do you (thereafter) maintain > individual documents? Or, merge them into a conditional > document?The way we did it was have a separate set of documents and resource files for the text components in each language and a language code. I doubt if that part has changed. They get bigger with time. Then you can add new phrases to the end as and when needed. That way additional languages can be added easily if someone else is willing to do the work.> Horror stories of attempts gone horribly wrong (i.e., what to > avoid)?Translations by some willing amateur who is not a native speaker of the destination language and which you are unable to check for veracity. Think most Chinglish instruction manuals for cheap hitech gear. -- Regards, Martin Brown
Reply by ●December 22, 20212021-12-22
On 12/22/2021 2:50 AM, Jeroen Belleman wrote:> On 2021-12-22 10:17, Martin Brown wrote: >> [...] In some ways their knowledge of more complex English >> grammar was better than many native English speakers today. Correct >> usage of "I would be obliged if you could" vs "would" - there is a >> hidden insult in the first phrase which has now sort of been lost in >> common English usage. > [...] > > Would you elaborate on that a little? The nuance escapes me. > Is it that 'could' throws doubt on the other's ability?Yes. It's the "Can" vs. "May" issue. There are innumerable other "technical screwups" that have crept into the language that folks either are oblivious to or unconcerned with fixing. (My personal pet peeve is the use of "what" in place of "that": "He's the guy what sold me that lemon of a car!")
Reply by ●December 22, 20212021-12-22
On 12/22/2021 2:17 AM, Martin Brown wrote:> On 22/12/2021 08:02, Don Y wrote: >> I'm looking for folks who've first hand experience having >> documents translated into foreign languages. Said documents >> to include diagrams (think: callouts, legends), #included >> text, etc. > > I have a friend who is a patent translator. They actually do into English from > several European languages that they are fluent in. > > Advice is generally that the translator must be fluent in both (written rather > than spoken) and *very* strong preference that they are a native speaker of the > destination language in each case.This suggests N translators for N languages.> I was involved in porting software into Japanese a long time ago and it was > quite entertaining. The jump from 1 byte to 2 byte DBCS character > representation made things even more fun back then. But the worst part was > finding a technical translator that could understand domain specific terms and > give them the right nuance. eg. "Dwell time" caused trouble. > First pass it came out as "time in home" - there were lots of others. > > In Japan it wasn't uncommon to meet perfectly good English to Japanese > translators who were just too shy to speak English even though they could > understand it perfectly well and write it almost fluently. In some ways their > knowledge of more complex English grammar was better than many native English > speakers today. Correct usage of "I would be obliged if you could" vs "would" - > there is a hidden insult in the first phrase which has now sort of been lost in > common English usage.Yes, I had a neighbor who was native Japanese and her husband spoke it fluently (having lived there for many years "teaching english"). He tried explaining the different "styles" of Japanese speaking (e.g., from what we'd consider "colloquial" to "audience with the Emperor")>> I've a fair bit of experience with I18N/L10N for software >> but the extent of the effort, there, is usually fairly limited. >> And, there's less of a need for a cohesive approach as the >> interactions are "punctuated" (no pun intended). >> >> Recommendations for firms to do this? (no, finding multilingual >> "friends" to do same is far too unprofessional -- though they may >> have value in proofing the results) I suspect there is some >> value in having a single firm handle all of the translations >> (in the hope that they will create a consistent SET of >> translations, even if different individuals are involved for >> each) >> >> Relative effort? (i.e., closer to reading speed or writing speed?) > > The effort involved is quite high and gets higher and more expensive the more > important the documents are and the exact precision of technical translation > required. Patent translation is about the most extreme.But I would imagine highly technical is easier to verify correctness. Less formal prose would rely on the translator understanding colloquialisms, analogies, etc. I.e., the *intent* of the prose. There also seems to be a fair bit of "cultural" risk involved. [I recall making a version of an arcade game for export and finding that the skull-and-crossbones used to signify the death of a player was highly discouraged in some markets whereas the US market wouldn't even shrug at the iconography] I'm also concerned that graphs/illustrations may require alteration to accommodate different string lengths, etc.> Many now use a hybrid machine translation human correction approach at least in > the UK where person hours are expensive. It works well if you translate a lot > of similar things with their own set of phrases (as happens quite a lot in > patents). >> >> Time frame? (is this effort-bound or business-bound) >> >> Cost? (and, "unit of measure"?) >> >> Finally, how to check the translation for accuracy and "feel" >> (i.e., ensuring it is true to the original intent)? > > These days machine translate back with eg Google and see if you can recognise > anything of what you had originally written. You really need to have it all > proof read by another native speaker of the language who is a domain expert in > whatever the content is. That way you find most of the peculiar ambiguities > that translation and punctuation can insert.Exactly. So, it's like finding *two* service bureaus instead of just one. Any value to having a second group translate back into the original language?> "Eats shoots and leaves" for example. >> >> With translations in hand, do you (thereafter) maintain >> individual documents? Or, merge them into a conditional >> document? > > The way we did it was have a separate set of documents and resource files for > the text components in each language and a language code. I doubt if that part > has changed. They get bigger with time. Then you can add new phrases to the end > as and when needed. > > That way additional languages can be added easily if someone else is willing to > do the work.I don't understand. Are you building a translation *dictionary* that you apply to create the document(s)?>> Horror stories of attempts gone horribly wrong (i.e., what to >> avoid)? > > Translations by some willing amateur who is not a native speaker of the > destination language and which you are unable to check for veracity. Think most > Chinglish instruction manuals for cheap hitech gear.Yes, exactly. But, that would be easy to avoid. The bigger fear is hiring a service bureau and later discovering they were little better than mechanical translators (again, because "you" likely can't review the result, directly).
Reply by ●December 22, 20212021-12-22
On 22/12/2021 10:47, Don Y wrote:> On 12/22/2021 2:17 AM, Martin Brown wrote: >> On 22/12/2021 08:02, Don Y wrote: >>> I'm looking for folks who've first hand experience having >>> documents translated into foreign languages. Said documents >>> to include diagrams (think: callouts, legends), #included >>> text, etc. >> >> I have a friend who is a patent translator. They actually do into >> English from several European languages that they are fluent in. >> >> Advice is generally that the translator must be fluent in both >> (written rather than spoken) and *very* strong preference that they >> are a native speaker of the destination language in each case. > > This suggests N translators for N languages.If you have an idea what you want they tend to form clubs that allow them to do a combination of different languages and load share. My friend used to run a translation bureau for the same sort of work before she went freelance as part of a group. There is a UK translators club with verified exam standard pieces. I expect it is similar in the USA.> >> I was involved in porting software into Japanese a long time ago and >> it was quite entertaining. The jump from 1 byte to 2 byte DBCS >> character representation made things even more fun back then. But the >> worst part was finding a technical translator that could understand >> domain specific terms and give them the right nuance. eg. "Dwell time" >> caused trouble. >> First pass it came out as "time in home" - there were lots of others. >> >> In Japan it wasn't uncommon to meet perfectly good English to Japanese >> translators who were just too shy to speak English even though they >> could understand it perfectly well and write it almost fluently. In >> some ways their knowledge of more complex English grammar was better >> than many native English speakers today. Correct usage of "I would be >> obliged if you could" vs "would" - there is a hidden insult in the >> first phrase which has now sort of been lost in common English usage. > > Yes, I had a neighbor who was native Japanese and her husband spoke it > fluently (having lived there for many years "teaching english"). He > tried explaining the different "styles" of Japanese speaking (e.g., > from what we'd consider "colloquial" to "audience with the Emperor")Yes. Using the wrong highly formal method of address can come across as incredibly sarcastic or demeaning if you get it wrong. There is a more neutral form which is safe enough for foreigners to get away with.>>> Relative effort? (i.e., closer to reading speed or writing speed?) >> >> The effort involved is quite high and gets higher and more expensive >> the more important the documents are and the exact precision of >> technical translation required. Patent translation is about the most >> extreme. > > But I would imagine highly technical is easier to verify correctness. > Less formal prose would rely on the translator understanding > colloquialisms, > analogies, etc. I.e., the *intent* of the prose.Its the ones that you are not aware of that catch you out. Where a phrase in English has a literal translation in the destination language that has another unintended or unwanted meaning. eg. Nova cars in Spanish.> There also seems to be a fair bit of "cultural" risk involved.Best one I recall was a before and after photo advert for washing powder designed by a Western ad agency but used in a country where they read right to left. Why would anyone buy a powder that turns clean clothes dirty!> I'm also concerned that graphs/illustrations may require > alteration to accommodate different string lengths, etc.We had some real difficulties with feet & inches on engineering diagrams when a US buyout meant that they had to be supported in a previously entirely metric system with decimals engineering design software.>> Many now use a hybrid machine translation human correction approach at >> least in the UK where person hours are expensive. It works well if you >> translate a lot of similar things with their own set of phrases (as >> happens quite a lot in patents). >>> >>> Time frame? (is this effort-bound or business-bound) >>> >>> Cost? (and, "unit of measure"?) >>> >>> Finally, how to check the translation for accuracy and "feel" >>> (i.e., ensuring it is true to the original intent)? >> >> These days machine translate back with eg Google and see if you can >> recognise anything of what you had originally written. You really need >> to have it all proof read by another native speaker of the language >> who is a domain expert in whatever the content is. That way you find >> most of the peculiar ambiguities that translation and punctuation can >> insert. > > Exactly. So, it's like finding *two* service bureaus instead of just one. > > Any value to having a second group translate back into the original > language?Not really. Having a native speaker proof read it for unwanted ambiguities checked against the English original is worthwhile though. Google translate might be good enough these days to give you enough hints of where things might well have gone awry. Many technical terms look similar in most languages (except French where the language police insist that software is logiciel hence Logitech).>> "Eats shoots and leaves" for example. >>> >>> With translations in hand, do you (thereafter) maintain >>> individual documents? Or, merge them into a conditional >>> document? >> >> The way we did it was have a separate set of documents and resource >> files for the text components in each language and a language code. I >> doubt if that part has changed. They get bigger with time. Then you >> can add new phrases to the end as and when needed. >> >> That way additional languages can be added easily if someone else is >> willing to do the work. > > I don't understand. Are you building a translation *dictionary* that you > apply to create the document(s)?Not a dictionary so much as a set of key phrases that will appear in dialogue boxes in the software - each having a unique numeric token. It is also a part of the way the computer assisted translation engines work for commonly used constructs.>>> Horror stories of attempts gone horribly wrong (i.e., what to >>> avoid)? >> >> Translations by some willing amateur who is not a native speaker of >> the destination language and which you are unable to check for >> veracity. Think most Chinglish instruction manuals for cheap hitech gear. > > Yes, exactly. But, that would be easy to avoid. > > The bigger fear is hiring a service bureau and later discovering they > were little better than mechanical translators (again, because "you" > likely can't review the result, directly).I think you have to ask around for recommendations in the territory or where you want to have the work done. Some of our translations were done by the national distributors (Korean for instance) and seemed to go OK. The big jumps were doing the first non-English one that included accents and top bit set characters and then the Japanese one with full DBCS. It got a bit easier after that. There is a lot more support for internationalisation DBCS these days. -- Regards, Martin Brown
Reply by ●December 22, 20212021-12-22
Don Y wrote:> On 12/22/2021 2:50 AM, Jeroen Belleman wrote: >> On 2021-12-22 10:17, Martin Brown wrote: >>> [...] In some ways their knowledge of more complex English >>> grammar was better than many native English speakers today. Correct >>> usage of "I would be obliged if you could" vs "would" - there is a >>> hidden insult in the first phrase which has now sort of been lost in >>> common English usage. >> [...] >> >> Would you elaborate on that a little? The nuance escapes me. >> Is it that 'could' throws doubt on the other's ability? > > Yes. It's the "Can" vs. "May" issue. > > There are innumerable other "technical screwups" that have crept into > the language that folks either are oblivious to or unconcerned with > fixing. (My personal pet peeve is the use of "what" in place of "that": > "He's the guy what sold me that lemon of a car!")Interesting. OTSOTP saying "could you do X" is polite by itself, but "would you do X" needs a 'please'. The implication of 'could' includes "of course you would if it were possible", which it might not be, e.g. by reasons of time, authority, law, and so forth. In contrast, over here 'would' sounds like 'of course you could if you actually wanted to help me, but maybe you don't care enough.' Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC / Hobbs ElectroOptics Optics, Electro-optics, Photonics, Analog Electronics Briarcliff Manor NY 10510 http://electrooptical.net http://hobbs-eo.com
Reply by ●December 22, 20212021-12-22
On a sunny day (Wed, 22 Dec 2021 07:51:29 -0500) it happened Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote in <b96443b3-03a8-8b25-024e-138cde7329b0@electrooptical.net>:>Don Y wrote: >> On 12/22/2021 2:50 AM, Jeroen Belleman wrote: >>> On 2021-12-22 10:17, Martin Brown wrote: >>>> [...] In some ways their knowledge of more complex English >>>> grammar was better than many native English speakers today. Correct >>>> usage of "I would be obliged if you could" vs "would" - there is a >>>> hidden insult in the first phrase which has now sort of been lost in >>>> common English usage. >>> [...] >>> >>> Would you elaborate on that a little? The nuance escapes me. >>> Is it that 'could' throws doubt on the other's ability? >> >> Yes. It's the "Can" vs. "May" issue. >> >> There are innumerable other "technical screwups" that have crept into >> the language that folks either are oblivious to or unconcerned with >> fixing. (My personal pet peeve is the use of "what" in place of "that": >> "He's the guy what sold me that lemon of a car!") > >Interesting. OTSOTP saying "could you do X" is polite by itself, but >"would you do X" needs a 'please'. > >The implication of 'could' includes "of course you would if it were >possible", which it might not be, e.g. by reasons of time, authority, >law, and so forth. In contrast, over here 'would' sounds like 'of >course you could if you actually wanted to help me, but maybe you don't >care enough.' > >Cheers > >Phil Hobbshttps://songteksten.net/lyric/400/67817/simon-garfunkel/el-condor-pasa.html and he sailed away faster than the wind...
Reply by ●December 22, 20212021-12-22
On 2021-12-22 11:33, Don Y wrote:> On 12/22/2021 2:50 AM, Jeroen Belleman wrote: >> On 2021-12-22 10:17, Martin Brown wrote: >>> [...] In some ways their knowledge of more complex English >>> grammar was better than many native English speakers today. Correct >>> usage of "I would be obliged if you could" vs "would" - there is a >>> hidden insult in the first phrase which has now sort of been lost in >>> common English usage. >> [...] >> >> Would you elaborate on that a little? The nuance escapes me. >> Is it that 'could' throws doubt on the other's ability? > > Yes. It's the "Can" vs. "May" issue. > > There are innumerable other "technical screwups" that have crept into > the language that folks either are oblivious to or unconcerned with > fixing. (My personal pet peeve is the use of "what" in place of "that": > "He's the guy what sold me that lemon of a car!")Indeed. I still think it's supposed to be "He's the guy /who/ ...". That may be British rather than American English. Languages evolve. Jeroen Belleman
Reply by ●December 22, 20212021-12-22
On 2021/12/22 2:47 a.m., Don Y wrote:> On 12/22/2021 2:17 AM, Martin Brown wrote: >> On 22/12/2021 08:02, Don Y wrote: >>> I'm looking for folks who've first hand experience having >>> documents translated into foreign languages. Said documents >>> to include diagrams (think: callouts, legends), #included >>> text, etc. >> >> I have a friend who is a patent translator. They actually do into >> English from several European languages that they are fluent in. >> >> Advice is generally that the translator must be fluent in both >> (written rather than spoken) and *very* strong preference that they >> are a native speaker of the destination language in each case. > > This suggests N translators for N languages. > >> I was involved in porting software into Japanese a long time ago and >> it was quite entertaining. The jump from 1 byte to 2 byte DBCS >> character representation made things even more fun back then. But the >> worst part was finding a technical translator that could understand >> domain specific terms and give them the right nuance. eg. "Dwell time" >> caused trouble. >> First pass it came out as "time in home" - there were lots of others. >> >> In Japan it wasn't uncommon to meet perfectly good English to Japanese >> translators who were just too shy to speak English even though they >> could understand it perfectly well and write it almost fluently. In >> some ways their knowledge of more complex English grammar was better >> than many native English speakers today. Correct usage of "I would be >> obliged if you could" vs "would" - there is a hidden insult in the >> first phrase which has now sort of been lost in common English usage. > > Yes, I had a neighbor who was native Japanese and her husband spoke it > fluently (having lived there for many years "teaching english"). He > tried explaining the different "styles" of Japanese speaking (e.g., > from what we'd consider "colloquial" to "audience with the Emperor")Not to mention masculine vs feminine Japanese - a person speaks slightly different languages depending on the gender of the speaker. I have no idea if that percolates down into technical matters... My ex is Japanese and in the 90s she was one of two bi-directional accredited translators (Japanese - English) in British Columbia, Canada. John :-#)#
Reply by ●December 22, 20212021-12-22
On 12/22/2021 8:56 AM, Jeroen Belleman wrote:> On 2021-12-22 11:33, Don Y wrote: >> On 12/22/2021 2:50 AM, Jeroen Belleman wrote: >>> On 2021-12-22 10:17, Martin Brown wrote: >>>> [...] In some ways their knowledge of more complex English >>>> grammar was better than many native English speakers today. Correct >>>> usage of "I would be obliged if you could" vs "would" - there is a >>>> hidden insult in the first phrase which has now sort of been lost in >>>> common English usage. >>> [...] >>> >>> Would you elaborate on that a little? The nuance escapes me. >>> Is it that 'could' throws doubt on the other's ability? >> >> Yes. It's the "Can" vs. "May" issue. >> >> There are innumerable other "technical screwups" that have crept into >> the language that folks either are oblivious to or unconcerned with >> fixing. (My personal pet peeve is the use of "what" in place of "that": >> "He's the guy what sold me that lemon of a car!") > > Indeed. I still think it's supposed to be "He's the guy /who/ ...". > That may be British rather than American English. Languages evolve.It would be a tough call to determine if American English had evolved more OR LESS than the original British. I've read that American English is, in many ways, truer to its British roots than modern British English. Pronunciations also evolve, over time. As well as speech patterns. E.g., I was taught "the" should be pronounced as "thee" when preceding a word beginning with a vowel sound: "Thee English", "Thee other guy" but with a schwa ahead of a consonant: "The next one", "the Frenchman". This seems to no longer be the norm. [You're interested in these sorts of things when you design a speech synthesizer; the different "wh" sounds, etc.]
