| Pages in topic: < [1 2 3] > | Selling Glossaries? Thread poster: Camille Abou Jamra
|
|---|
Kevin Fulton 
United States
Local time: 12:18
German to English
| Terminology research is slow-going work | Mar 4, 2009 |
RobinB wrote:
We do quite a lot of contract terminology work for customers, billed by the hour. Although our customers rarely want more than "bare bones" terminology entries (e.g. term pair with context and examples, no gramcats), the *average* productivity is probably no more than 4 to 5 terms per hour.
It's always astonishing to me how few translators (and clients!) appreciate that terminology, as intellectual property, is a vital, valuable asset for translators and clients alike.
This confirms my experience
In 2007-2008 I worked as a terminologist on a mega-project. My product was a "bare-bones" glossary based upon a customer-supplied terminology list that was rife with errors. Although I felt a sense of accomplishment if I managed to find/validate 10 terms in an hour, there were just as many instances when developing and validating a translation for a term took an hour or more. Four to five terms an hour is certainly an accurate assessment of average hourly production over the life of a project.
From time to time I'm approached by agencies to develop terminology, paid on a per-word basis, just like a translation of standard text. Even assuming an output of 20 terms per hour (unlikely!!) at twenty cents per term (also unlikely), there's no way to make even a minimum wage.
| | | | | Entirely agree! | Mar 4, 2009 |
RobinB wrote:
It's always astonishing to me how few translators (and clients!) appreciate that terminology, as intellectual property, is a vital, valuable asset for translators and clients alike.
Absolutely vital! I think all translators should take a course on any termbase tool and start recording terminology with information as complete as possible. The benefits for the translator are clear: A) Using the right term in the first place, B) Not having to repeat research work/dig in old files along time, C) having some traceability of the terms used when and for what customer.
The only challenge is, as you said, reserving enough time to research terms for the future.
PS: Robin, we finally agree on something! I am terrified.
| | | | RobinB
United States
Local time: 11:18
German to English
| Termbase tools are fine, but.... | Mar 4, 2009 |
Tomás Cano Binder, CT wrote: Absolutely vital! I think all translators should take a course on any termbase tool
... in practice they're an optional extra. Unless you really have a lot of spare time on your hands, it's going to be much faster entering terms into, say, Excel sheets, or even Word tables, and using an indexing tool to retrieve them. I'm afraid that applications like Multiterm may be great for students doing their terminology dissertation (often the case here in Germany), but they're simply too unwieldy to use sensibly in day-to-day practice.
PS: Robin, we finally agree on something! I am terrified.
I'm not aware of ever having particular disagreements with very much you've written, Tomás. Indeed, I agree with a lot of what you write in threads to which I do not contribute, either because I don't have the time or because I refuse to get sucked into some futile argument with, for example, a bunch of nationalist socialists.
But I'm getting old and doddery (how many other people can remember terminology work back in the days before Windows, I wonder?) and my mind's probably going...
Robin
- who brought the world "MARTIF Lite" - check it out - as well as "Simplified Universal Terminology Interchange" (SUTI, pron. "sooty") and the "Simplified Universal Interchange Protocol" (SUIP, pron. "sweep").
| | | | hazmatgerman (X)
Local time: 18:18
English to German
| RobinB / DOS world | Mar 5, 2009 |
Thanks for your reminder that term work existed pre-WIN. A lot can be done with a DOS-based dBase application despite the restricted field length. On productivity your 5 or 6 terms/h may sound low but in the long run it matches my calculation.
On a different note: IMO selling validated glossaries at cost-based prices is rarely possible except to large institutions who already have similar glossaries and know how much it costs. They have to be treated as sunk cost instead.
Regards.... See more Thanks for your reminder that term work existed pre-WIN. A lot can be done with a DOS-based dBase application despite the restricted field length. On productivity your 5 or 6 terms/h may sound low but in the long run it matches my calculation.
On a different note: IMO selling validated glossaries at cost-based prices is rarely possible except to large institutions who already have similar glossaries and know how much it costs. They have to be treated as sunk cost instead.
Regards.
[Edited at 2009-03-05 10:11 GMT] ▲ Collapse
| | |
|
|
|
| That's the kind of terminology that's worth paying for | Mar 5, 2009 |
RobinB wrote:
fully validated, structured bilingual terminology records in machine-readable format (containing at least keywords, subject area codes, explanations, examples and/or context, and some gramcat information)
That's the kind of quality of validated terminology that would be worth purchasing. But very few translators (if any) would be able to afford glossaries at a rate of 5 Euro per record, on their own.
The terminology team of the business software company for which I worked productivity similar to what Robin mentions.
The glossaries produced were made available to the in-house translators and also to the outside translators provided by a translation company (to work at our offices). That is, translators get this kind of terminological support only when a big company is behind it.
| | | | | Plug for WordFast as a terminology management tool | Mar 5, 2009 |
RobinB wrote: Tomás Cano Binder, CT wrote: Absolutely vital! I think all translators should take a course on any termbase tool ... in practice they're an optional extra. Unless you really have a lot of spare time on your hands, it's going to be much faster entering terms into, say, Excel sheets, or even Word tables, and using an indexing tool to retrieve them. I'm afraid that applications like Multiterm may be great for students doing their terminology dissertation (often the case here in Germany), but they're simply too unwieldy to use sensibly in day-to-day practice.
I agree with the opinions expressed here about the impracticability of selling glossaries. I would be very dubious about the quality and source of something like that offered for sale since the pricing issues mentioned would make it unlikely for the really great translators to be able to put the time into creating something like this for other people to use.
On a side note regarding Robin's comments above, I think WordFast provides a pretty easy-to-use way to organize and access terminology. Essentially, it is the Word table Robin mentions - a WordFast glossary is simply a 2- or 3-column text file. But once combined with WF, it's pretty powerful.
In my case, the first column in the file is German, the second English, and the third source (of the English term). I don't need a source column for the German term because my low volume allows me to simply keep different WordFast glossaries for different customers/projects so as not to mix things up. For my practical purposes, I don't need grammatical categories either. If I do need to note something like that, I include it in the third column.
Apart from anything else, using this system helps keep me "in line" while translating because once the glossary is imported, words in the glossary that pop up in the source text are highlighted so I have a visual reminder to check the glossary for the preferred term. This is the only way I can possible handle stuff like the client who gave me a 3,000-entry Excel glossary that needed to be followed to the letter. Looking up every possible term manually would have driven me crazy, not to mention blown any idea of translation speed out of the water.
Another useful feature is the ability to add terms to the glossary on the fly by simply highlighting the source and target terms while translating and clicking a button.
Anyway, that whole explanation was off topic, but I love this terminology function and plug it when I can because questions asking about good terminology tools seem to get posted regularly.
| | | | | What's the point of having glossaries, if you don't understand the text? | Mar 5, 2009 |
Tomás Cano Binder, CT wrote: Camille Abou Jamra wrote:
Do you consider your cleaned and specialized glossaries as "drafts of terms" of no value as stated by a colleague? Glossaries in the form of bilingual lists may serve the purpose of putting some translated word there, but they don't help you understand what the matter is all about (which of the twelve translations of "screen" is correct in my case????). They don't provide the intelligence to understand what word is best in a context.
I agree: many translators seem to think that if only they can find the "right" translation for some term, it doesn't really matter if their understanding of the source text is hazy (or sometimes non existent). They don't realize that if they really don't understand a text, they will never be able to translate it properly.
I've written about this recently, in The Magical School of Terminology Management.
| | | | krizck
Mexico
Local time: 10:18
English to Spanish
| It's up to you! | Mar 6, 2009 |
Some people may think that selling your glossary is like giving away one of the most important tools. The problem is not the money per se, but "competition". If you think that the person who wants to buy the glossary may represent a problem, you might not want to sell it. However, you will receive some money in return. So if you need some money, why not?
| | |
|
|
|
| terminology tools | Mar 6, 2009 |
RobinB wrote:
... in practice they're an optional extra. Unless you really have a lot of spare time on your hands, it's going to be much faster entering terms into, say, Excel sheets, or even Word tables, and using an indexing tool to retrieve them. I'm afraid that applications like Multiterm may be great for students doing their terminology dissertation (often the case here in Germany), but they're simply too unwieldy to use sensibly in day-to-day practice.
To me, glancing at the results of the automatic terminology search and pressing ALT+right to insert a term into my text isn't too cumbersome (that's all it takes if you're translating with Trados and use Multiterm).
Perhaps you prefer copying the source term, switching to a different window, pasting it, hitting search checking if anything came up, copying the result, switching back to your translation window and pasting... all I can say is: to each their own.
| | | | RobinB
United States
Local time: 11:18
German to English
| Automatic term entry? | Mar 6, 2009 |
FarkasAndras wrote: To me, glancing at the results of the automatic terminology search and pressing ALT+right to insert a term into my text isn't too cumbersome (that's all it takes if you're translating with Trados and use Multiterm).
Perhaps you prefer copying the source term, switching to a different window, pasting it, hitting search checking if anything came up, copying the result, switching back to your translation window and pasting... all I can say is: to each their own.
I evidently didn't make myself sufficiently clear: the problem isn't with copying over terms from a termbase. The problem is getting terms into that termbase in the first place. Terminology research is slow and very manual: in the fields I work in (financial and legal), the source documents for high-grade terminology tend to be either paper-based or (often locked) PDFs, so entering terms into a termbase is something you do when you've got time on your hands. Which happens maybe once a year. It's not as if you can vacuum up thousands of terms from the web. And you can't use the terminology extraction programs available today for any form of sensible ex ante terminology generation. So if you can get parallel corpora in a machine-readable (and copyable) format, the most effective way to generate terminology quickly is by copy/paste into something simple and reliable like an Excel worksheet or a Word table.
One of the reasons why terminology is necessarily expensive is that it's a very labour-intensive occupation. Quite apart from the fact that there are so few people with the necessary depth of subject area knowledge combined with terminology management skills. Or the time to use them...
| | | |
RobinB wrote: FarkasAndras wrote: To me, glancing at the results of the automatic terminology search and pressing ALT+right to insert a term into my text isn't too cumbersome (that's all it takes if you're translating with Trados and use Multiterm).
Perhaps you prefer copying the source term, switching to a different window, pasting it, hitting search checking if anything came up, copying the result, switching back to your translation window and pasting... all I can say is: to each their own. I evidently didn't make myself sufficiently clear: the problem isn't with copying over terms from a termbase. The problem is getting terms into that termbase in the first place. Terminology research is slow and very manual: in the fields I work in (financial and legal), the source documents for high-grade terminology tend to be either paper-based or (often locked) PDFs, so entering terms into a termbase is something you do when you've got time on your hands. Which happens maybe once a year. It's not as if you can vacuum up thousands of terms from the web.
Actually, you can vacuum up thousands of terms from the web. Even tens of thousands, quite easily. I know I have done that. Not in highly specialized, narrow fields of course, but there are pretty good general resources for the major areas. Google GEMET, ISI, Eurovoc, CPV or Microsoft glossary for a few examples.
Quite apart from that, any glossary in spreadsheet format can be easily imported into any decent terminology programme. Otherwise, they (terminology programmes) would be a monumental waste of time. Give me a 1000-line two-cloumn excel table and I'll give you a Wordfast or Multiterm termbase in under two minutes.
Back on track: how much for that thousand-line glossary? 
| | | | hazmatgerman (X)
Local time: 18:18
English to German
| vacuum process | Mar 11, 2009 |
Sure the net provides ample ground for any vacuum cleaner to be put to work, and what fills up the bag may be called dirt.
Seriously: term work is a long and slow road, for any term needs ancillary information culled from the (parallel) source texts. Copy'n'paste is not the work but thinking about meanings is. For a specialised subject a rate of say 3.5k words/yr I hold sustainable, not more.
Regards.
| | |
|
|
|
Anders Jensen
Philippines
Local time: 00:18
English to Danish
+ ...
| Of the crowd, by the crowd, for the crowd | Mar 27, 2009 |
Considering the cost and usefulness of termbases and glossaries, it is clear that the only way to do it is to use crowd sourcing.
Of the crowd, by the crowd, for the crowd
| | | | | The crowd does not have a clue! | Mar 27, 2009 |
Anders Jensen wrote:
Considering the cost and usefulness of termbases and glossaries, it is clear that the only way to do it is to use crowd sourcing.
Of the crowd, by the crowd, for the crowd
I completely disagree with this statement. Unfortunately "the crowd" usually does not have a clue about what is the right term translation experts use in their every-day language. Only checking with experts, or checking documentation made by experts, can yield good results in glossaries.
The two only solid approaches I can think of about researching terms (classifying and houskeeping them correctly for actual work is a different issue) is:
a) Checking parallel documents translated by experts, for instance those in the EN standards in their translated counterparts in the different European countries. They are not always perfect, but they are a often good start.
b) Reading the definition of the source term in a monolingual, specific dictionary or reference book, checking a bilingual dictionary for the possible translations of the term, then checking a monolingual, specific dictionary or reference book in the target language to find which of the terms really corresponds to the definition of the source term.
All in all, serious terminology is a lot of work, but it is clearly worth the effort.
Let the crowd do your glossaries, and you will end up to your neck in pure crap. Oops. Sorry for the harsh language!
| | | | | Multiterm in day-to-day practice | Mar 27, 2009 |
RobinB wrote:
Unless you really have a lot of spare time on your hands, it's going to be much faster entering terms into, say, Excel sheets, or even Word tables, and using an indexing tool to retrieve them. I'm afraid that applications like Multiterm may be great for students doing their terminology dissertation (often the case here in Germany), but they're simply too unwieldy to use sensibly in day-to-day practice.
Now we disagree again... We in the office have been using Multiterm for some 4 years now to record and classify the terminology we use. We don't always have the time to record absolutely all terms used in our text, but we do record the critical ones with every job. Each term is defined, depicted with a picture if possible, classified in one or several topics, stored in the 2, 3 or 4 languages depending on how much information we gather along time, and defined with linguistic information (grammar category, gender), along with the source the term was taken from. We also record what customer(s) we have used the target term for, just for some traceability in the future.
It's not really that hard as you can use one term as the template for another (you don't create all "pumps" from scratch; instead you create a full definition of "pump" and copy and edit it to create "gear pump", "screw pump", "piston pump"...).
After 4 years working this way we have gathered over 5,000 entries. It does not sound like a big termbase compared to term lists people may be gathering out there, but for these 5,000 terms we have solid, verified information that is useful forever!
| | | | | Pages in topic: < [1 2 3] > | To report site rules violations or get help, contact a site moderator: You can also contact site staff by submitting a support request » Selling Glossaries? | 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... | | | | | |
Login to reply/comment 
Translation news
