Mobile menu

Basic CAT-related concepts for a beginner
Thread poster: Pablo Guazzotti

Pablo Guazzotti  Identity Verified
Local time: 15:29
English to Spanish
+ ...
Mar 30, 2007

Hello all,
I just got a copy of Deja Vu 7.02 with which I'll be diving into the CAT world. The first doubt that came to my mind while reading the manual (yes, I wasn't too far into it!) was what exactly are the following itmes, and how they interact with each other:
-translation memory
-terminology database
-lexicon

Naturally common sense dictates me an idea, but I still feel I'm missing the gist of it. Unawareness of what is omitted by the manual for being obvious is not something I'm proud of, but maybe some charitable spirit can point me in the right direction.

Best,
Pablo


Direct link Reply with quote
 

GoodWords  Identity Verified
Mexico
Local time: 12:29
Spanish to English
+ ...
Three memories Mar 30, 2007

The translation memory/memories holds/hold all the segments (usually sentences) you have translated in the past (with Déjà Vu). If the segment you are currently translating matches (or almost matches) one that you have translated before, it will show it (or them) to you.

The terminology database(s) hold(s) all the words and phrases that you have entered into it/them (also from prior translations). If any of these are in the segment you are translating now, it will show it (or them) to you.

The lexicon is a terminology database which is project-specific. Most Déjà Vu users employ it seldom or not at all.

In summary, while you are translating, each time you move to the next segment, the program will check the translation memory(memories) to see if you have translated a similar or identical segment before. If you have, it (they) will be presented to you.
The program will also check to see if any words and phrases in the segment are in your terminology database(s). If there are any, they will be shown to you. If you are using a lexicon, the the lexicon will also be checked.


Direct link Reply with quote
 

ViktoriaG  Identity Verified
Canada
Local time: 14:29
English to French
+ ...
A little more... Mar 30, 2007

I just wanted to add to GoodWords' explanation that the difference between transaltion memories and terminology databases is that the former usually contains sentences and the latter, terms or expressions. You can insert a translation memory match (100% or similar) into the target segment, but you can't really insert the terminology database matches, because often, there are several terminology databse matches, as there may be several terms in the same sentence that are found in the terminology database. You can insert the database matches, but only one at a time, that is, not automatically, as opposed to the memory.

To sum it up, both have their own uses and it is best to use them together. The database is like a regular dictionary and the memory, well, helps you to remember how you translated entire sentences. Having both - is priceless!



[Edited at 2007-03-30 04:51]


Direct link Reply with quote
 

Hepburn  Identity Verified
France
Local time: 19:29
English to French
+ ...
On the subject of terminlology... Mar 30, 2007

It will seem incredible, but I have been using wordfast for five years ad did wonder of and on about why I NEVER got a prompt of a word, - which would be very useful at times. I only got whole sentences prompted from the memory, but somehow sort of accepted it, for lack of time to look into the matter.

I did see the terminology and glossary indications, but had once a look, found nothing and just went on...I thought thet if I had to enter each word manually it might take me more time than looking it up again , as needed.

So, are you telling me that all the words which are in my memory could appear singly in the terminology data bank - without being manually inserted by me???? WHERE are they then? And WHY do theuy not appear singly when needed?
I must be missing something.

I am free this morning and it would be great if I could clear up that problem.

I welcome any information, help, advice.

Claudette


Direct link Reply with quote
 

Pablo Guazzotti  Identity Verified
Local time: 15:29
English to Spanish
+ ...
TOPIC STARTER
Thanks Mar 30, 2007

Thanks for clarifying and confirmed what I had suspected. Now, I'm faced with the task of creating a memory for my translation project. The document I'll translate is of the same nature as another two I have previously completed, and hence the question is inevitable; is there a way I can feed DV with those existing documents (source and target) in order for the software to go some way to filling the data base and start from a couple of rungs up? I was told Trados has a function of that sort, (I don't own trados anyway) so I wonder....
thank you all!


Direct link Reply with quote
 

ViktoriaG  Identity Verified
Canada
Local time: 14:29
English to French
+ ...
You may want to try +Tools Mar 30, 2007

+Tools is a piece of freeware (which does alignment, among other neat and useful things) complimentary to Wordfast, another CAT tool. You can download it here:

http://www.wordfast.org/

I have never tried it so I can't tell you how it works or what the results are like. Try it!

There may be other tools available...


Direct link Reply with quote
 

GoodWords  Identity Verified
Mexico
Local time: 12:29
Spanish to English
+ ...
More information and a question Mar 30, 2007

Pablo Guazzotti wrote:

Thanks for clarifying and confirmed what I had suspected. Now, I'm faced with the task of creating a memory for my translation project. The document I'll translate is of the same nature as another two I have previously completed, and hence the question is inevitable; is there a way I can feed DV with those existing documents (source and target) in order for the software to go some way to filling the data base and start from a couple of rungs up? I was told Trados has a function of that sort, (I don't own trados anyway) so I wonder....
thank you all!


You can import memories from other programs (for example if a client sends you a Trados memory), and you can align existing translations using the aligner tool (see pages 356 and 366 respectively in the Déjà Vu Standard manual at http://www.atril.com/docs/DVX/DVX%20Standard.pdf).

However many Déjà Vu users choose not to bother aligning past translations—I, for one, didn't, and I have seen in discussions on other lists that this is the typical experience. The benefit isn't so much for the first few translations you do with Déjà Vu, but for all the translations you will be doing in the months and years to come. You may see little advantage in the first few projects (unless there are internal repetitions within a project), but you will begin to see it soon, and it will keep increasing always as long as you continue to translate using Déjà Vu.

Claudette Hepburn wrote:
So, are you telling me that all the words which are in my memory could appear singly in the terminology data bank - without being manually inserted by me???? WHERE are they then? And WHY do they not appear singly when needed?
I must be missing something.

I welcome any information, help, advice.


In Déjà Vu it is extremely quick and simple to enter words and phrases into the terminology database, making them well worth entering if a term is used more than once or twice. I don't know how it is in other CAT tools. Let's hear from a Wordfast user about how this works in Wordfast.


Direct link Reply with quote
 
David Turner  Identity Verified
Local time: 19:29
French to English
+ ...
AutoAssemble Mar 31, 2007

Viktoria Gimbe wrote:

You can insert a translation memory match (100% or similar) into the target segment, but you can't really insert the terminology database matches, because often, there are several terminology databse matches, as there may be several terms in the same sentence that are found in the terminology database. You can insert the database matches, but only one at a time, that is, not automatically, as opposed to the memory.



That's true with most CAT programs but Déjà Vu goes a stage further in allowing you to assemble sub-sentence portions (phrases and terms) automatically (AutoAssemble). While such chunks obviously depend on context, they provide far more matches than at sentence level, and with some texts can often be strung together to produce whole sentences automatically that require very little editing. This, combined with EBMT - Example Based Machine Translation - whereby a "fuzzy" match is turned into an exact match by substitution of a chunk, can save the translator a lot of time.
Regards,
David Turner


Direct link Reply with quote
 
David Turner  Identity Verified
Local time: 19:29
French to English
+ ...
Download the Wordfast manual Mar 31, 2007

Claudette Hepburn wrote:
So, are you telling me that all the words which are in my memory could appear singly in the terminology data bank - without being manually inserted by me???? WHERE are they then? And WHY do theuy not appear singly when needed?
I must be missing something.


To store a term or phrase in a termbase wiith Wordfast, you select (highlight) the source term or phrase, press control-alt + T, then select the target term and press control-alt + T again. Such terms will then be highlighted in light blue in the next source segment that contains them. You can select them using Ctl-alt + left/right and insert them into the target using Ctrl-alt + down. Not so elegant or ergonomic as Déjà Vu but better than nothing.

Regards,
David Turner


[Edited at 2007-03-31 07:50]

[Edited at 2007-03-31 14:39]


Direct link Reply with quote
 

sylver  Identity Verified
Local time: 02:29
English to French
Terminology... Mar 31, 2007

David Turner wrote:
Claudette Hepburn wrote:
So, are you telling me that all the words which are in my memory could appear singly in the terminology data bank - without being manually inserted by me???? WHERE are they then? And WHY do theuy not appear singly when needed?
I must be missing something.

To store a term or phase in a termbase wiith Wordfast, you select (highlight) the source term or phrase, press control-alt + T, then select the target term and press control-alt + T again. Such terms will then be highlighted in light blue in the next source segment that contains them. You can select them using Ctl-alt + left/right and insert them into the target using Ctrl-alt + down.

You also have many functions to propagate automatically in the target segment. Search for "Propagate" in Wordfast's manual.

Context search (Ctrl+Alt+C), which you probably know already. Combined with the very handy Alt+F12 shortcut, it fits the bill (When using Wordfast, you can copy the selection from any Ms Word window into the target segment. Assuming you find the proper term in the context/concordance window, Alt+F12 will insert it in the translation).

David Turner wrote:
Not so elegant or ergonomic as Déjà Vu but better than nothing.


Personally, I don't like DV all that much. It's a good tool but I like Wordfast better.... and I have no financial interest in either tool.


Direct link Reply with quote
 
David Turner  Identity Verified
Local time: 19:29
French to English
+ ...
Horses for courses Mar 31, 2007

sylver wrote:
Personally, I don't like DV all that much. It's a good tool but I like Wordfast better.... and I have no financial interest in either tool.


Wordfast is a fine little tool and certainly good value for money. However, having to work in Word does entail quite severe limitations (no filtering or sorting possibilities and ability to work on only one file at a time, to name but a few). I've tried both tools quite extensively and I find I can often do the same job in less than half the time with DV.

Regards,
David Turner


Direct link Reply with quote
 

sylver  Identity Verified
Local time: 02:29
English to French
Don't bet on that race! Apr 2, 2007

David Turner wrote:
sylver wrote:
Personally, I don't like DV all that much. It's a good tool but I like Wordfast better.... and I have no financial interest in either tool.

Wordfast is a fine little tool and certainly good value for money. However, having to work in Word does entail quite severe limitations (no filtering or sorting possibilities and ability to work on only one file at a time, to name but a few). I've tried both tools quite extensively and I find I can often do the same job in less than half the time with DV.

Regards,
David Turner

What do you call "filtering"? Filtering of what? Same for "sorting". What do you want to sort?

TM? Glossaries? Both can be sorted and filtered in Wordfast. Easily.

Only one file at a time? Automatically processing several files in one go is interesting -and can be done in Wordfast. But what's the point of working on many files at the same time? In Wordfast and in DV, you work segment by segment, so I fail to see how much of a difference that is.

Working in Word with open formats (txt TMs and glossaries) means that you can do exactly what you want with the files and you can use all Ms Word's functions as well (including advanced F&R, VBA macros and third party add-ins).

Beginner will only see the advantage of working in the very familiar Ms Word environment, but Wordfast offers plenty of things to keep advanced users satisfied as well.

Just like you skipped the whole "propagation" feature, you also may not have realized how much freedom there was in open formats. For instance, did you realize that you could easily run SQL queries on Wordfast TM and glossaries?*

You didn't know? How many other "details" did you miss?

We all have preferences and if you like DV (and have financial interests in DV) that's fine, but Wordfast is certainly not a "little tool".

I don't like to be stuck with proprietary formats and be restricted to the existing functions. With Wordfast, if I want a feature and it doesn't exist, I can just do it myself because it's all open.

According to DV's website, "The External View format is specifically designed for proofing or checking unresolved issues outside of Déjà Vu X." (a Word/Excel export). This feature starts in the "Pro version".
Question: What do you do with "unresolved issues" when you only have a "Standard" edition?

Open formats also means that no matter what the problem is, there is always a solution.

Wordfast is not perfect, and DV is a fine tool as well. Don't get me wrong, I have a lot of respect for DV. But each to his own. I'll use Wordfast over DV any day.

- Sylver

___________________
*Piece of cake. Loading tab delimited text (all Wf TMs and glossaries as tab delimited) in a SQL database takes a few minutes and then you can run all the SQL queries you want. No limits


Direct link Reply with quote
 
David Turner  Identity Verified
Local time: 19:29
French to English
+ ...
Seeing things differently Apr 4, 2007

sylver wrote:
What do you call "filtering"? Filtering of what? Same for "sorting". What do you want to sort?
TM? Glossaries? Both can be sorted and filtered in Wordfast. Easily.


No, the files themselves. You just select any sequence of one or more words (a difficult term for example), press a button and DV will list all the sentences containing that term or expression and how it is used in the one or hundreds of files in your project. This is a very powerful feature (unique among CAT tools I believe) and is very useful in helping you understand a difficult term or expression and how it is used, and then ensuring that you translate it consistently.


Only one file at a time? Automatically processing several files in one go is interesting -and can be done in Wordfast. But what's the point of working on many files at the same time? In Wordfast and in DV, you work segment by segment, so I fail to see how much of a difference that is.


Even with a TM, it's hard to be consistent when working on several files. With DV you can import a whole web site containing hundreds of files and work on them in a unified, all-files view seamlessly as if they were one file. You can also import mixed file formats (Excel, Word and Powerpoint, for instance) and then display and sort sentences alphabetically or in natural order, or display all duplicate segments, all fuzzy match sentences, all perfect match sentences, all sentences over or under a certain length, all sentences containing numbers. Just about anything. You have far more visibility and flexibility than in Word. A big document often seems daunting in Word with countless pages stretching ahead. Once it's imported into a DVX project, life seems so much easier.


We all have preferences and if you like DV (and have financial interests in DV) that's fine, but Wordfast is certainly not a "little tool".


I don't have any financial interests in DV actually other than giving a DV training course every couple of months or so for the local Atril office. I earn my living as a freelance translator.


I don't like to be stuck with proprietary formats and be restricted to the existing functions. With Wordfast, if I want a feature and it doesn't exist, I can just do it myself because it's all open.


DVX can handle over 25 different formats including Office 2007, Indesign, Open Office, RESX, SDLX ITD, Rich Text Format, Windows Help, Adobe FrameMaker MIF, Adobe PageMaker tagged text, QuarkXPress XTG, Interleaf ASCII, HTML, ASP/ASP.NET, PHP, JavaScript, VBScript, HTML Help, SGML, XML, RC, C/C++/Java, Java properties, IBM TM untranslated segments, Trados WorkBench documents, TradosTag TTX, Trados TagEditor BIF, Trados TagEditor TMX, GNU PO and POT files, and plain text files, so I don't think it could really be accused of being proprietary, restrictive or not open. You must be thinking of SDL Trados.
And on top of that, SQL gives the user a lot of power to customize features


According to DV's website, "The External View format is specifically designed for proofing or checking unresolved issues outside of Déjà Vu X." (a Word/Excel export). This feature starts in the "Pro version".
Question: What do you do with "unresolved issues" when you only have a "Standard" edition?


The External View is really only used for sending work to be proofread by someone who doesn't have DV.


Wordfast is not perfect, and DV is a fine tool as well. Don't get me wrong, I have a lot of respect for DV. But each to his own. I'll use Wordfast over DV any day.


So be it. No tool is perfect and I'm not saying DV is. Having made considerable efforts to learn a tool, many translators tend to stick with the one they started out with, even with its imperfections. But judging from your comments, I would say that you probably haven't really had a good enough look at DV to be so critical of its merits. There's a new version 7.5 just out. Why don't you download a free 30-day trial version and see for yourself?

Regards,
David Turner

[Edited at 2007-04-04 19:04]

[Edited at 2007-04-05 07:27]


Direct link Reply with quote
 


To report site rules violations or get help, contact a site moderator:


You can also contact site staff by submitting a support request »

Basic CAT-related concepts for a beginner

Advanced search


Translation news related to CAT tools





SDL MultiTerm 2017
Guarantee a unified, consistent and high-quality translation with terminology software by the industry leaders.

SDL MultiTerm 2017 allows translators to create one central location to store and manage multilingual terminology, and with SDL MultiTerm Extract 2017 you can automatically create term lists from your existing documentation to save time.

More info »
Anycount & Translation Office 3000
Translation Office 3000

Translation Office 3000 is an advanced accounting tool for freelance translators and small agencies. TO3000 easily and seamlessly integrates with the business life of professional freelance translators.

More info »



All of ProZ.com
  • All of ProZ.com
  • Term search
  • Jobs