Mobile menu

Counting words with TagEditor
Thread poster: Therrien

Therrien  Identity Verified
Canada
Local time: 00:14
English to French
+ ...
Jun 17, 2008

Hello Forum:

I am new to the TagEditor and I am probably running into a simple problem.

I am a new to TagEditor, and CAT tools all-together.

I have been using TagEditor to complete a translation of .xml documents, all rendered
into a .ttx format by my handy associate.

I am now completing the translation work and would like to make a final word count, but I cannot find the function to do so. Mr. Help File states that "Word counts are displayed in the ITD tab of the Properties dialog box". Unfortunately, I don't see an ITD tab in the properties box.

I thought of simply taking the painful time to copy/paste the translated material from the preview pane and counting the lot using a word processor. I realized I am loosing precious words from my word count as some of the material is not in view at the end of the line, such as keywords. And I am sure there is a much simpler solution right under my nose.

Also, I am currently operating on a computer that only has Open Office, which does not work with Trados et al.

So, what do I seem to be missing?
Help would be very much appreciated.

Best:

Alexandre


Direct link Reply with quote
 

Eleni Makantani
Greece
Local time: 09:14
English to Greek
+ ...
Analyse Jun 17, 2008

Hello,

I think that Tageditor is really unfriendly to work on, but equally necessary when having to deal with different file types. Why don't you try Workbench--> Tools--> Analyse? I think you will get a reliable wordcount this way. It is also useful while translating, because it tells you how much work there is still to be done...

Regards,
Eleni


Direct link Reply with quote
 

Annelise Meyer  Identity Verified
France
Local time: 08:14
English to French
+ ...
A final wordcount of what? Jun 17, 2008

Hi Therrien,

What wordcount do you want, the target one or the source one? If you are using TagEditor, then you must have Translator's workbench opened, which means that you can perform an analysis (simply choose Tools>Analyse and pick the file after closing it) which will give you all the total source wordcount now that you've completed translation. Is this what you are looking for?

Best,
Annelise


Direct link Reply with quote
 

Therrien  Identity Verified
Canada
Local time: 00:14
English to French
+ ...
TOPIC STARTER
I stand humbled. Jun 17, 2008

Thank you very much for your reply. Yes, it was right under my nose!
As I effect the analysis, I see that the word count is there. Thank you!

Here's another question that I would like to ask for future reference if I could take another minute of your time:

If a quote is given with a sliding scale, according to the percentage match of the words in the document (fuzzy matches, etc), what is the tool or method used to view that in the target language?

After this translation the analyser gave me 100%, throughout the document. (Of course, I translated it and it saved the entries in the workbench)

How can I effect this word count in reference to this sliding scale, now having a different quantity of words in the target language?

Best:

Alex


Direct link Reply with quote
 

Eleni Makantani
Greece
Local time: 09:14
English to Greek
+ ...
Before the translation Jun 17, 2008

Dear Alex,

I'm not sure there is a way to get the sliding scale matches after the end of the translation. For the simple reason that now, all segments have an equivalent, and are considered by the program to be 100% matches (or repetitions, I'm not sure). What I usually do at the end of the working day is perform an analysis of the file and then I can see how much work there is left - as my work progresses, the percentage of 100% matches increases.

If you want to see the sliding scale, it's a good idea to perform the analysis before translating anything and printing out or saving separately the log file, for your own reference.

Maybe there is also a way to do this after the end of the work, but I don't know it...

Eleni


Direct link Reply with quote
 

Erik Freitag  Identity Verified
Germany
Local time: 08:14
Member (2006)
Dutch to German
+ ...
Analyse against empty TM Jun 17, 2008

Hi!

If you want a word count with the percentages of fuzzy matches, repetitions etc., all you have to do is analyse the original file against an empty TM.

Kind regards,
Erik


Direct link Reply with quote
 

Annelise Meyer  Identity Verified
France
Local time: 08:14
English to French
+ ...
Precisely Jun 17, 2008

Yes, either use an empty TM or the original TM (without the new segments) if you received one from your client.

Best,
Annelise


Direct link Reply with quote
 

Therrien  Identity Verified
Canada
Local time: 00:14
English to French
+ ...
TOPIC STARTER
L'Envoi Jun 17, 2008

I understand what you mean now.
And thank you very much for the info.

I must understand why on the dispatch I received, there was a price-tag put on fuzzy matches. As I understand, there was a .TMW file created with similar material and the contractor analyzed the documents and there was fuzzy matches above 50%. From what I understand: since those segments will be edited, this is a completely dependant variable.

The only arithmetic I can tinker with this would be to deduct the original fuzzy match and 100% match count from the final word count and apply the difference on the invoice as "no-match" (full price).

Unless there's a basic concept I'm missing, this sounds like the modus operandi.

With Best Regards:

Alex


Direct link Reply with quote
 

Katalin Horváth McClure  Identity Verified
United States
Local time: 02:14
Member (2002)
English to Hungarian
+ ...
Are you invoicing based on source or target wordcount? Jun 18, 2008

I am a bit concerned that you may be easily ripped off if you don't fully understand these things.
First of all, are you invoicing based on source or target word count?
Depending on the language pair, there will be expansion/reduction in wordcount when going from source to target.
It seems you were given a TM (a .tmw and a few other files), the source file and a list containing percentage bands for fuzzy matches, 100% matches and repetitions.

I will not go on a tangent about why it should be YOU doing the analysis and setting those bands up and decide how much YOU want to charge for each, before you accept the work, and why it is NOT the outsourcer's role to dictate those terms - I will just try to explain the mechanics.

The analysis is done on the source file.
As others already pointed out, if you have a copy of the original TM and the original source, you can run the Analysis in Workbench to see the matches.
You will get a list containing percentage bands for fuzzy matches, 100% matches and repetitions. For a single segment, if it is determined to be a 50% fuzzy match, that means that for the given source segment, there is a similar segment in the TM, and the level of their similarity is 50%. So, for example out of 10 words, 5 are the same. So, if this segment was 10 words long, that will add 10 to the 50% fuzzy wordcount in the analysis. I hope this is clear.

The log file, which you can save as a CSV file and open up in Excel, too, will have the matches as segments, words, and also as percentages. (For example, 12% of the text is 100% match.)

If you are invoicing based on target wordcount, you can apply the percentages to the target word count. It is not 100% accurate, as you don't know how the expansion affects the particular sentences, but it is a pretty close estimation, to assume that expansion/reduction affects all sentences about the same.

Another way of getting match statistics in target count terms would be to reverse the original TM (export it and import it into a newly created TM that has the source and target languages reversed), and analyze the cleaned up, target language translation against that. I am sure it would not be the same as the source statistics, but in some cases it may be more fair, as it would show how much of the TM was really used.

Imagine, that the TM was garbage, and you had to edit it a lot, even 100% or 99% matches. The source statistics do not show that, you would get paid the same for a good 99% match and a bad 99% match that you had to rewrite.
If you do the reverse TM + target analysis, you would get a lower match and thus, a higher payment for the edited 99% match.


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 »

Counting words with TagEditor

Advanced search


Translation news related to SDL Trados





WordFinder
The words you want Anywhere, Anytime

WordFinder is the market's fastest and easiest way of finding the right word, term, translation or synonym in one or more dictionaries. In our assortment you can choose among more than 120 dictionaries in 15 languages from leading publishers.

More info »
PerfectIt consistency checker
Faster Checking, Greater Accuracy

PerfectIt helps deliver error-free documents. It improves consistency, ensures quality and helps to enforce style guides. It’s a powerful tool for pro users, and comes with the assurance of a 30-day money back guarantee.

More info »



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