Pages in topic: < [1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16] | Complimentary TM-Town membership for ProZ.com members Thread poster: Jared Tabor
| I see your point... | May 10, 2016 |
Maxi Schwarz wrote: I waited before posting until I knew whether you had authorized that announcement. Frankly, I was shocked. My own transactions with clients are private. I would never imagine that the management of any site where my name is listed, and who has access to my private information, would see fit to publish that information. And not just in the body of a post, even in the subject line. I don't know if this falls under confidentiality, but it does seem to fit into privacy, trust, and judgment. ... and I must admit I found it a little unusual myself. I wouldn't have done that, but in the end, it was an isolated case, and the only information Kevin disclosed is that I've been contacted by a client for a particular project, in support of the argument he was trying to make. Of course, if something like that became a habit, then yes, I could definitely see a problem with that, for all of the reasons you mentioned. | | | Sheila Wilson Spain Local time: 21:06 Member (2007) English + ... This seems an appropriate point to repeat the question I put to staff | May 10, 2016 |
Mirko Mainardi wrote: Tom in London wrote: Jared has already made it clear in a previous post that those who pay the extra and get the TmTown thing will have no advantage over those who don't. it seems to me Jared only said that your "It does seem that unless you sign up for this TMTown thing, and pay extra, you will actually end up being disadvantaged in your Proz ranking" was incorrect... And indeed, as things stand, those are two separate rankings, and I don't think TM-Town membership is (currently) affecting directory rankings on ProZ... However, since this also seems to be a situation which will inevitably evolve (somehow) in the near future, due to the fact that ProZ did acquire TM-Town, "the point is" exactly what I previously wrote, that is to say, the hope that " someone else will read and take into consideration the doubts and concerns expressed here by me and many other ProZ members when deciding how to move forward with this new ProZ+TM-Town combine", hopefully in order to avoid scenarios similar to the one you were referring to with your words I quoted above, and, possibly, to make TM-Town a useful, reliable, equitable and reputable tool for ProZ members (as opposed to a "competing" platform that could actually damage some of them...). Back on 15 April, on Page 1, I asked staff: How is TM-Town being marketed to outsourcers? Kevin later addressed the points I raised from a TM-Town viewpoint but there has been no staff response. I quote from that first post (with typos corrected ): I really think it's crucial for us to be informed of the way this [TM-Town] is being sold to agencies and direct clients. I'm totally against being shortlisted by such an arbitrary method and I would really like to be able to evaluate whether it's worth continuing paying for membership, if that's the future here. At the moment expertise, writing skills, reputation, etc all help channel potential clients in our direction. Through TM-Town, it's more of a big-data, number-crunching lottery. To take a random person (far from my own case), how would someone with a lifetime's experience of translating in the sector of religion, a lay preacher and teacher, etc fare against someone who had aligned and uploaded the Bible and other such texts in two languages? Is volume going to be master now? And how on earth can clients know whether there's any actual skill behind that volume? So, are clients going to be encouraged first and foremost to keep doing things the way they have been doing them, i.e. contact us through job postings, [traditional, not TM-Town] directory lists and personal profile messages? Or are they now going to be encouraged to simply put in some details of the job and see who gets chosen by the machine? | | | Tom in London United Kingdom Local time: 21:06 Member (2008) Italian to English
Mirko Mainardi wrote: So... what was your point again? If I could understand your post I'd tell you. | | |
Tom in London wrote: Mirko Mainardi wrote: So... what was your point again? If I could understand your post I'd tell you. For my unintelligible English, then, but that shouldn't prevent you from clarifying what your original point was. Ciao ![](https://cfcdn.proz.com/images/bb/smiles/icon_smile.gif) | |
|
|
Andrea Muller (X) United Kingdom Local time: 21:06 English to German + ... Who will benefit in the long run - translators or translation buyers? | May 10, 2016 |
Slightly off topic, but I wonder who will turn out to be real beneficiary of this 'upload your TMs' trend - and I suspect it won't be translators. Translators are encouraged to upload their TMs. They hope that a translator buyer will select them as the most suitable translator, because their uploaded TMs show that they have translated many similar texts. But will translation buyers be satisfied with finding someone who has demonstrated that they are good at translating... See more Slightly off topic, but I wonder who will turn out to be real beneficiary of this 'upload your TMs' trend - and I suspect it won't be translators. Translators are encouraged to upload their TMs. They hope that a translator buyer will select them as the most suitable translator, because their uploaded TMs show that they have translated many similar texts. But will translation buyers be satisfied with finding someone who has demonstrated that they are good at translating a particular type of document (assuming that TM matching is indeed an indicator of translation quality)? Or are they more interested in saving money and expect that translators will offer a discount, because they have already translated many similar documents and might get some fuzzy matches against their own TMs? I have seen adverts on ProZ from another company called TM-something, which other contributors to this thread mentioned already. That outfit offers "Delivery of TM segments for EUR 0.04per TTSL (Standard Line in the Target Text) of the TM segments delivered". And they describe the point of their services as follows: "The situation: Many standard texts are completely retranslated time and time again. The goal: A comparison of each text with the TMs of all the translators in the world taking part." ▲ Collapse | | | Tom in London United Kingdom Local time: 21:06 Member (2008) Italian to English
Andrea Muller wrote: ... their uploaded TMs show that they have translated many similar texts If there's one thing translating agencies are keen to get their hands on, it's TMs. if a schoolboy in his bedroom is able to hack into the CIA's databases I imagine it will not be difficult to get hold of these TMs. | | |
It seems to me that this would be recycling material already delivered to previous clients. Taking a subject where CATS are most useful, let us say I specialized in translating technical instruction manuals, and I translated a manual for a toaster and uploaded it to TM Town. Then since it was my specialty, I was asked to translate another instruction manual for a toaster of a competitor company. The material to translate would likely be very similar, and thanks to using the same TM,... See more It seems to me that this would be recycling material already delivered to previous clients. Taking a subject where CATS are most useful, let us say I specialized in translating technical instruction manuals, and I translated a manual for a toaster and uploaded it to TM Town. Then since it was my specialty, I was asked to translate another instruction manual for a toaster of a competitor company. The material to translate would likely be very similar, and thanks to using the same TM, the client would receive a very similar text to his competitor, even though there are still different ways to say the same thing, even in this technical subject. This would be worse with any subject more creative than an instruction manual. What about self plagiarism? I think every client expects and deserves to receive new content. I do not use the same TMs for different clients because of this. I use them for consistency of terms and with previous works for the same client, but I think my clients expect me to provide a fresh translation and do not want exactly the same content as all my other clients in my area of specialty, even if this content is fragmented and slightly different. As to confidentiality, I would not go ahead and upload my clients content to any server without their explicit permission, but I am not prepared to solicit my clients' permission for this. I do not like to bother my clients often, I also usually do not provide references in order not to bother them. But even if the concept appealed to me, I could not upload my clients content with out explicitly asking them first. I am convinced that the TM Town company takes all possible measures to ensure data security, but nothing is secure 100%. Remember when a number of celebrities' iCloud accounts were recently hacked, data leaks and hacks do happen. ▲ Collapse | | | Michael Beijer United Kingdom Local time: 21:06 Member (2009) Dutch to English + ... Deshi, Deshi, Deshi | May 16, 2016 |
Juliaq wrote: As to confidentiality, I would not go ahead and upload my clients content to any server without their explicit permission, but I am not prepared to solicit my clients' permission for this. I do not like to bother my clients often, I also usually do not provide references in order not to bother them. But even if the concept appealed to me, I could not upload my clients content with out explicitly asking them first. I am convinced that the TM Town company takes all possible measures to ensure data security, but nothing is secure 100%. Remember when a number of celebrities' iCloud accounts were recently hacked, data leaks and hacks do happen. I'm not sure I understand the first part of your post about recycling of TMs, but the confidentiality issues you raised can be solved entirely by Deshi. See: https://www.tm-town.com/desktop-application MJWB | |
|
|
Out of curiosity, where are TM-Town's servers located? | | |
Hello, I don't know if this question has been arisen before: will Deshi support Word files too? Thanks. | | | ..... (X) Local time: 05:06
Hi Susana, Susana E. Cano Méndez wrote: will Deshi support Word files too? No, currently Deshi supports translation memory files in the following formats: • .tmx • .xlf • .sdlxliff • .sdltm • .txt (Wordfast) • .txml | | |
Kevin Dias wrote: Hi Susana, No, currently Deshi supports translation memory files in the following formats: • .tmx • .xlf • .sdlxliff • .sdltm • .txt (Wordfast) • .txml Thank you, Kevin. | |
|
|
Maren Postel Germany Local time: 22:06 Italian to German + ... Not off topic at all! | Jul 24, 2017 |
Andrea Muller wrote: Slightly off topic, but I wonder who will turn out to be real beneficiary of this 'upload your TMs' trend - and I suspect it won't be translators. Translators are encouraged to upload their TMs. They hope that a translator buyer will select them as the most suitable translator, because their uploaded TMs show that they have translated many similar texts. But will translation buyers be satisfied with finding someone who has demonstrated that they are good at translating a particular type of document (assuming that TM matching is indeed an indicator of translation quality)? Or are they more interested in saving money and expect that translators will offer a discount, because they have already translated many similar documents and might get some fuzzy matches against their own TMs? I have seen adverts on ProZ from another company called TM-something, which other contributors to this thread mentioned already. That outfit offers "Delivery of TM segments for EUR 0.04per TTSL (Standard Line in the Target Text) of the TM segments delivered". And they describe the point of their services as follows: "The situation: Many standard texts are completely retranslated time and time again. The goal: A comparison of each text with the TMs of all the translators in the world taking part." | | | Maren Postel Germany Local time: 22:06 Italian to German + ... Not off topic at all! | Jul 24, 2017 |
Andrea Muller wrote: Slightly off topic, but I wonder who will turn out to be real beneficiary of this 'upload your TMs' trend - and I suspect it won't be translators. Translators are encouraged to upload their TMs. They hope that a translator buyer will select them as the most suitable translator, because their uploaded TMs show that they have translated many similar texts. But will translation buyers be satisfied with finding someone who has demonstrated that they are good at translating a particular type of document (assuming that TM matching is indeed an indicator of translation quality)? Or are they more interested in saving money and expect that translators will offer a discount, because they have already translated many similar documents and might get some fuzzy matches against their own TMs? I have seen adverts on ProZ from another company called TM-something, which other contributors to this thread mentioned already. That outfit offers "Delivery of TM segments for EUR 0.04per TTSL (Standard Line in the Target Text) of the TM segments delivered". And they describe the point of their services as follows: "The situation: Many standard texts are completely retranslated time and time again. The goal: A comparison of each text with the TMs of all the translators in the world taking part." Why should I upload my translations and tms, they are my intellectual property and I'm absolutely capable of administrating them on my own. In this sense, TM-town is obsolete for me. | | | Kaspars Melkis United Kingdom Local time: 21:06 English to Latvian + ... TM-Town biggest problem | Jul 25, 2017 |
is not concerns about confidentiality. A lot of texts translators produce are publicly available anyway. If someone steals your TM, he only saves the effort of doing alignment. I believe that smart algorithms developed by Google are currently trained to do it with very good results. The idea of calculating the score of each translator that shows relation to a given text for translation could be quite useful, in fact. It could save a lot of time of doing translator evaluation and cou... See more is not concerns about confidentiality. A lot of texts translators produce are publicly available anyway. If someone steals your TM, he only saves the effort of doing alignment. I believe that smart algorithms developed by Google are currently trained to do it with very good results. The idea of calculating the score of each translator that shows relation to a given text for translation could be quite useful, in fact. It could save a lot of time of doing translator evaluation and could be beneficial both to clients and translators. Unfortunately, the implementation is very poor and that's why I believe TM-Town will fail unless they change it to make it much simpler. I tried to play with Nakōdo Expert Finder from a client's perspective and got a graph with many data points on 2 axis corresponding to similarity and quantity scores. I have only a vague idea of what they could potentially mean but there is nothing to confirm my assumptions. Then I have no clue in what way I should prefer the one or another and how they correspond to most suitable translator for a given text. There are no explanations given at graph either. If I am looking for the best or even good enough translator for my text, this graph doesn't tell me that. Should I give preference to those who are higher on quantity score, or the one who is further right on the similarity score? Do the creators expect the client to provide the answer they cannot answer themselves? Clients inevitably think about costs as well and they already draw a 2D graph with quality and price parameters. Translation quality is very hard to measure and replacing it with 2 other vague parameters of similarity and quantity scores is not going to make evaluation easier or more precise. Compare it with a credit score. It is one single number that corresponds to a person's statistical ability to pay their debts. we don't know exactly how it is calculated but we know that it takes into account all previous financial history and it is shown to be very precise on group level. It is very simple, easy to understand and quite useful if you are a moneylender. Adding any additional complexity will only lead to disaster as subprime mortgage crisis has taught us. I strongly believe that if TM-Town is to have any success, it needs to be a single number score. Then the client can easily choose the best translator that fits their desired quality/price ratio. I am sorry, for my public critical review. I clearly understand that designing such a score is not a simple matter and any attempts should be encouraged and supported. Any failures are good as we can learn a lot from them. ▲ Collapse | | | Pages in topic: < [1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16] | To report site rules violations or get help, contact a site moderator: You can also contact site staff by submitting a support request » Complimentary TM-Town membership for ProZ.com members CafeTran Espresso | You've never met a CAT tool this clever!
Translate faster & easier, using a sophisticated CAT tool built by a translator / developer.
Accept jobs from clients who use Trados, MemoQ, Wordfast & major CAT tools.
Download and start using CafeTran Espresso -- for free
Buy now! » |
| Wordfast Pro | Translation Memory Software for Any Platform
Exclusive discount for ProZ.com users!
Save over 13% when purchasing Wordfast Pro through ProZ.com. Wordfast is the world's #1 provider of platform-independent Translation Memory software. Consistently ranked the most user-friendly and highest value
Buy now! » |
|
| | | | X Sign in to your ProZ.com account... | | | | | |