One big TM/TDB or several little ones
Thread poster: Reed James
Reed James
Reed James
Chile
Local time: 21:13
Member (2005)
Spanish to English
May 31, 2006

I am a user of SDLX 2006. I have also worked with Deja Vu X and Wordfast. My question is: is it better to have several translation memories and term bases, i.e. one for each project or field of expertise, or is it better to have one huge one?

I remember back on the DVX Yahoo discussion group, there were many advocates of "Big Mama", meaning one big file for both TM and TDB.

I know from experience that Wordfast encourages the user to split these files up as small as p
... See more
I am a user of SDLX 2006. I have also worked with Deja Vu X and Wordfast. My question is: is it better to have several translation memories and term bases, i.e. one for each project or field of expertise, or is it better to have one huge one?

I remember back on the DVX Yahoo discussion group, there were many advocates of "Big Mama", meaning one big file for both TM and TDB.

I know from experience that Wordfast encourages the user to split these files up as small as possible; a TM for each project.

On one hand, it saves me time to just dump everything into one big pile of terms. On the other, I notice that there is a certain degree of clutter.

I would be interested to hear from other CAT Tool users in order to learn more. Thanks.

Reed
Collapse


 
mónica alfonso
mónica alfonso  Identity Verified
Local time: 21:13
English to Spanish
+ ...
IMHO Jun 1, 2006

several individual TM work better.
You can always use (at least with Trados) another TM as reference and when searching for concordance.
Individual TM are easier to load, and you update just what belongs to a specific job.
Personally, I have a big folder for TM, with a subfolder for each TM, and I have identified each by the subject plus the name of the agency or client who assigned that job. In case the same client comes with a similar job, I resource to the exact TM I need, a
... See more
several individual TM work better.
You can always use (at least with Trados) another TM as reference and when searching for concordance.
Individual TM are easier to load, and you update just what belongs to a specific job.
Personally, I have a big folder for TM, with a subfolder for each TM, and I have identified each by the subject plus the name of the agency or client who assigned that job. In case the same client comes with a similar job, I resource to the exact TM I need, and not to a clutter of options I do not need (or want). Again, I sometimes have some other TM ready as reference.
Collapse


 
Kevin Fulton
Kevin Fulton  Identity Verified
United States
Local time: 20:13
German to English
Large general categories Jun 1, 2006

I use 5 different translation memories in DVX depending upon the subject matter. In Trados (I have SDLX but haven't used it much), my TM is client-related, since my clients frequently supply a TM or MultiTerm database to be used. As long as you back up your databases regularly, it shouldn't matter what your approach is.

 
Viktoria Gimbe
Viktoria Gimbe  Identity Verified
Canada
Local time: 20:13
English to French
+ ...
Here's what I do: Jun 1, 2006

For each job, I create a new memory. Once the job is done, I make a copy of the newly created memory and import it into the master memory, which is one of the few memories I have that are subject-specific and that compile many jobs I've done throughout the years. In Trados, I use the newly created memory and specify the master memory as the read-only memory. This way, I have access to the new memory and the master memory at once, but I am still keeping the newly created memory separated, in case... See more
For each job, I create a new memory. Once the job is done, I make a copy of the newly created memory and import it into the master memory, which is one of the few memories I have that are subject-specific and that compile many jobs I've done throughout the years. In Trados, I use the newly created memory and specify the master memory as the read-only memory. This way, I have access to the new memory and the master memory at once, but I am still keeping the newly created memory separated, in case the client needs it. Once merged, memories cannot be un merged again. I have merged and single copies of all my memories, just in case.

As for SDLX and Wordfast, I am not sure how it goes, but I believe you can import in both of them, which means you can reproduce this workflow with them. However, if you cannot specify a second, read-only memory to use while you work, I would simply create a new memory, import the master memory into it, and work that way. However, I would also make sure to label all my TUs differently, so that, if the client wants their memory, I will be able to export just the TUs for that client, without exporting the entire master memory.

Hope this helps!
Collapse


 
Vauwe
Vauwe
Local time: 01:13
English to German
+ ...
Several TMs Jun 1, 2006

What you want is consistency of customer-specific terminology in follow-up texts of a specific project. When you first translate e.g. a marketing text of a digital camera, then a software guide and then a tunnel boring machine, I think the databases have nothing in common and there is no need to put all data in one great TM.

Best regards


 
Marc P (X)
Marc P (X)  Identity Verified
Local time: 01:13
German to English
+ ...
One big TM/TDB or several little ones Jun 1, 2006

Several little ones. Hundreds, even. A new one created at the end of each job.

Mix and match them. Sort them in folders by subject, by customer, etc. For each new job, just select the folder containing the TMs you want to use.

At least, that's the way we do things in OmegaT!

Marc


 
Samuel Murray
Samuel Murray  Identity Verified
Netherlands
Local time: 01:13
Member (2006)
English to Afrikaans
+ ...
It depends on your tool... Jun 1, 2006

Reed D. James wrote:
I know from experience that Wordfast encourages the user to split these files up as small as possible; a TM for each project.


Well, the fact that WF itself (without the server) can use only one TM at a time, forces me to use a big momma TM when using WF. OmegaT on the other hand, can use any number of TMs simulataneously, which means I get as much leverage from individual TMs as I would get from a big momma TM. The same can be said of Isometry (multiple TMs).

For private clients, I use one TM per client. I never mix them, unless I have a new client in a subject field for which I have several smaller TMs available. In such a case, I compile a large TM from the smaller ones, but after the job is done, I extract only the TUs that really occurred in that job and place them in a separate TM.


 
Nicolette Ri (X)
Nicolette Ri (X)
Local time: 01:13
French to Dutch
+ ...
One per end client Jun 1, 2006

In most cases I have one TM per end client, because I don't want to mix up client-specific vocabulary and styles.


[Edited at 2006-06-01 14:09]


 
Oliver Walter
Oliver Walter  Identity Verified
United Kingdom
Local time: 00:13
German to English
+ ...
More information about Wordfast Jun 1, 2006

Samuel Murray wrote:
Well, the fact that WF itself (without the server) can use only one TM at a time, forces me to use a big momma TM when using WF.

With WF you can only use one TM at a time in the usual way but you can also have a "background TM" at the same time. This will, however, only find 100% matches. That's how I understand it, though I haven't in fact used this facility.
Oliver


 
Reed James
Reed James
Chile
Local time: 21:13
Member (2005)
Spanish to English
TOPIC STARTER
What about repetition? Jun 1, 2006

I like the sound of this approach but am not sure how to set it up... If I have 1 TM/TDB per client, how do I keep from repeating the same words in each dB? Though many subjects have specific terminology, there are several recurring terms. What about the small, everyday words like "and", "however", "his" etc.? Thanks.

Reed

MarcPrior wrote:

Several little ones. Hundreds, even. A new one created at the end of each job.

Mix and match them. Sort them in folders by subject, by customer, etc. For each new job, just select the folder containing the TMs you want to use.

At least, that's the way we do things in OmegaT!

Marc


 
Marc P (X)
Marc P (X)  Identity Verified
Local time: 01:13
German to English
+ ...
One big TM/TDB or several little ones Jun 1, 2006

Reed D. James wrote:

Though many subjects have specific terminology, there are several recurring terms. What about the small, everyday words like "and", "however", "his" etc.? Thanks.


TMs don't contain terminology. They contain segments (sentences and their translations).

Marc


 
Nicolette Ri (X)
Nicolette Ri (X)
Local time: 01:13
French to Dutch
+ ...
They do Jun 1, 2006

MarcPrior wrote:

Reed D. James wrote:

Though many subjects have specific terminology, there are several recurring terms. What about the small, everyday words like "and", "however", "his" etc.? Thanks.


TMs don't contain terminology. They contain segments (sentences and their translations).

Marc

Of course they do, if you do a context search. This feature searches in the TM, and in all the TMs in the same directory, if you want.


 
Marc P (X)
Marc P (X)  Identity Verified
Local time: 01:13
German to English
+ ...
One big TM/TDB or several little ones Jun 1, 2006

Nicolette Richy wrote:

Of course they do, if you do a context search. This feature searches in the TM, and in all the TMs in the same directory, if you want.


This is correct, but it has no bearing on Reed's question. The purpose of of TMs is to store segments, not terminology.

Marc


 
Marc P (X)
Marc P (X)  Identity Verified
Local time: 01:13
German to English
+ ...
One big TM/TDB or several little ones Jun 1, 2006

Reed D. James wrote:

If I have 1 TM/TDB per client, how do I keep from repeating the same words in each dB? Though many subjects have specific terminology, there are several recurring terms. What about the small, everyday words like "and", "however", "his" etc.?


The answer is that you can't, and you don't. There are multiple occurrences of terms even within a single, "Big Mama" TM. These multiple occurrences are beneficial, since they enable you to see how the same term has been used in different contexts. Whether you are using multiple TMs or a single one has no bearing on this.

Marc


 


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


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

One big TM/TDB or several little ones







Protemos translation business management system
Create your account in minutes, and start working! 3-month trial for agencies, and free for freelancers!

The system lets you keep client/vendor database, with contacts and rates, manage projects and assign jobs to vendors, issue invoices, track payments, store and manage project files, generate business reports on turnover profit per client/manager etc.

More info »
Trados Studio 2022 Freelance
The leading translation software used by over 270,000 translators.

Designed with your feedback in mind, Trados Studio 2022 delivers an unrivalled, powerful desktop and cloud solution, empowering you to work in the most efficient and cost-effective way.

More info »