SOURCE TEXTIf you have questions on the Tutorial, exercises or WordFisher, please address them to Tibor Kornyei, WordFisher's creator.
The tutorial contains explanatory notes and step-by-step process on how to use some functions. Most of the information was taken from the official User Manual. However, the order follows that of the exercises. This tutorial is by no means a complete guide to WordFisher. Its goal is to explain what WordFisher is and show many of its functions in the context of translation. For a complete help file, please refer to the WordFisher's official User Manual.
In order to understand WordFisher better, it is preferable that you understand what CAT tools are, especially, and its concepts. It is also essential that you be familiar with Word and Windows standard functions.
WordFisher is an advanced macro toolkit for translators working with Microsoft Word documents. WordFisher is not a standalone program. It needs Word in order to work. This means that you work in Word and use WordFisher functions via its toolbars, which will be installed in the toolbar menu of Word. WordFisher is not a translation memory program, but it does offer several of the functions found in full-featured Translation Memory (TM) programs. So we can say that WordFisher is a CAT tool. It enhances Word and turns it into efficient software, which can help automate your translation process.
WordFisher was written in the macro language of Word 6 and Word 7 (or Word 95), WordBasic. However, Word 8 (or Word 97) and Word 9 (or Word 2000) convert the macros into their new macro language, VBA, without any problem.
This is an overview of some of the things you can do with WordFisher. It is by no means complete, but it gives a good idea of the amount of automation you can achieve in your daily work to make it easier and more efficient. For a more complete picture, please refer to the WordFisher's User Guide.
To better understand what WordFisher can do, it may be easier to put it back in the context of translation as we practice it everyday. To do so, we'll look at the various steps in the translation process, more precisely the preparation and production stages, which are our main activities.
The following steps are not necessarily made in the order in which they are presented, and there may be more steps in your translation workflow; those steps may be repeated, be done simultaneously, etc. We'll simply use them to show what WordFisher can do. Note that in this section, we are not demonstrating nor explaining how to use the toolbars and buttons. This will be done in the exercise section (or you can refer to the User Guide for a step-by-step help file).
Preparation Read/analyse source text:SOURCE TEXT
What do you do with your source text when you translate? You can print it and put it on a copyholder next to the computer to refer to it.
You can also have it on screen because it is possible to have several documents open in Word, each in its own windows. These windows can be rearranged at will. It is what I call the multiple document management. You can then go from one window to another with your mouse or a short key. But the source text doesn't synchronise your movement in the target text. That is you are translating the fourth paragraph in page 15 in your target file, but when you go to your source file to check the context, it is on page 12 where you last looked at it. So you have to go down manually, or search for the place with Word's search function.
WordFisher rearranges windows and synchronises your movements automatically between the target text and the source text.
If you prefer to see only your target text because you do not like to work in two panes, there is still a way for you to view the source text when needed by asking WordFisher to show you the source sentence where your cursor is positioned.
Usually, if you have received the source document in electronic format, you will probably open it, save it under another name (adding your language perhaps, for example FR to indicate French), and work in that document by overwriting the source text, say sentence per sentence, or typing before the source text and erasing the sentence once your have finished translating it. This ensures that you do not touch the original file.
WordFisher does something similar. It doesn't touch your original file. Briefly, WordFisher analyses the source text, cut it into segments (like TM does). Then it creates (save) two files, one for the source text and the other for the target text. You will work in the target file by overwriting the source text. Instead of overwriting the text, it is often better to type before the source text and erase it once you have finished translating the segment. WordFisher has a macro to automatically erase the text to the right of the cursor to the end of the segment and move your cursor to the next segment.
TERMINOLOGY AND
REFERENCES
By terminology I mean bilingual glossaries (yours, the client's, colleagues', etc.) and by references, I mean passages already translated (yours, the clients' colleagues', etc.), which WordFisher calls bilingual corpus. References can also be related material to your translation (monolingual and bilingual).
CREATE
TERMINOLOGY LIST (BILINGUAL GLOSSARY)
While you translate, it is quite useful to build a glossary as you go along. What you would do in Word is create a new document with a two-column table, which will contain the source term and its equivalent that you just found. To create your glossary, you would need to enter manually each entry. That is you would merrily copy and paste from the source document go to your glossary document and copy the source text in say the left column, then go to the target document copy the equivalent, go to the glossary file and paste the term in the right column. Then go back to what you were doing. Or you would simply type the entry directly in the table.
WordFisher allows you to create a glossary rather effortlessly. All you need to do is highlight the terms you want and click on some buttons.
Note that you may prepare questions to the client in forms of a glossary with WordFisher before you start translating, that is in the preparation stage of the translation workflow. You can do this also during the translation or after the first draft.
Note also that you can create a glossary with any other files than the ones from which you are translating (reference documents, previous translations).
You can also add entries to any existing glossary at any stage in the translation process. This means that you can also integrate other glossaries (the client's glossary for example to keep his preferences) to your own.
CREATE
REFERENCES (BILINGUAL CORPUS)
What WordFisher calls "bilingual corpus" is an aligned file containing segments in the source language on one side and the corresponding segment in the target language. Aligned files represent invaluable reference material for translators. You can use them to search solutions to translation problems. You can view the equivalent easily and the terms are in context.
WordFisher allows you to create bilingual corpus automatically in forms of Word table while you translate to be able to search what you have already translated in the current document. At the end of the project, you can create the final table and use it to pre-translate other files.
Since you can create bilingual corpus as you translate, you could also use that table to ask questions to your client for example. This is convenient because that way you can also give your client the context. You can simply save the WordFisher table under another name in Word, edit it to keep only the problematic passages and send it to your client. You could also send the whole table and mark the problematic passage (assigning a colour, highlighting it in yellow, etc.).
WordFisher allows you also to take previous translations (yours, clients, colleagues, etc.) and align them automatically to form a bilingual corpus, which you can use to pre-translate future translation, search equivalents and solutions to translation problems. You often need to edit the table afterwards, however.
SEARCH TERMINOLOGY
AND REFERENCES
We do not only create bilingual glossaries and corpus, we search them too. Not only our own but glossaries that were sent by the client to follow a particular terminology, for example. Or you may wish to search translations you have done in the past or reference material sent to you by the client or found on the Net. In Word, you could open each file you need and use the normal search function to search through them. Or you could use the advanced search option to search in more than one file.
The Find function of WordFisher makes searching several files very efficient. It works similarly as a text retriever, giving you the number of hits and highlighting them in their context so you can browse through the results more easily. WordFisher uses the internal search engine of Word but enhances it. This means you can search more than one file and use Boolean operators (something you cannot do with the normal search function of Word). You can also use the Find function to search in the translation project you are working on.
EXTRACTING DATA
FROM REFERENCES (BILINGUAL GLOSSARIES AND CORPUS)
Once you have created glossaries and corpus, you want to be able to extract data from them to help you translate more quickly and use a more consistent terminology.
In Word, you would open your glossary file or your past translation file, search through them with the simple search function until you find the term, expression or passage you want. Then you would copy what you need from that file, go to your target file, and paste it.
It happens often while we are translating that certain passages are repeated within our text. In a normal Word environment, you'd have to remember where that passage was or use the normal search function of Word to find it, copy it then paste it to the place you were, then carry on. In a 30-page document, this can be time consuming, even using the splitting windows function of Word.
Wouldn't it be nice to be able to automatically extract data from a terminology list and from past translations or already translated passage within a document? Like translation memory software, WordFisher allows you to do that, albeit in a less powerful way. So instead of copy and paste the text manually, you can use WordFisher to pre-translate a new text with old material using the bilingual glossaries and corpus that you built over time.
PROOFREAD, CORRECT
AND EDIT
After finishing your first draft, you will need to correct and edit your translation.
You have received answers from the client and need to make changes, you may also want to change words, sentences, etc. The normal stuff we do when editing. WordFisher can help you go faster with its Search and Replace function.
When editing your final translation (or someone else's translation), you want to make sure that the terminology is consistent. That is always an important factor. Making sure a particular term has been translated the same way everywhere. This is a more demanding task if you are editing files that were done by several translators, for example, but it can be also quite demanding in your own file, especially in big documents. How do we go about this in Word? From memory, from searching in our glossaries or references, using search and replace, etc. WordFisher uses the bilingual glossary you created to check for consistency automatically.
You can use the table format (bilingual corpus created by WordFisher) at the end of your translation to verify if you haven't forgotten to translate any passage, checking segment by segment.
MULTIPLE FILE
PROJECT
You probably had to work on several Word files for the same project before. For example a 30,000-word manual divided in twenty chapters each in its own file.
In a normal Word environment, you will have to open each file separately and work on each one separately. You can have several files open at the same time, of course, depending on the capacity of your computer. But you still have to work on one at the time.
What if you want to change a term in more than one file? What about that passage that was translated in file05 and which comes back in file12? How about checking consistency of terminology?
With WordFisher you can search and replace, pre-translate and check consistency in several files simultaneously.
CONCLUSION OF
THE INTRODUCTION
There are more functions in WordFisher. Once you know it better, you can explore it to see how you can adapt it to your type of translation, your style of working and find use that other people may not have thought of before because they didn't have the need.
Even if you never use pre-translating functions or if you use only one function to help you increase your productivity and automate certain daily tasks (creating glossary, aligning files automatically, search references, or replace terms in several files), at its low price, WordFisher offers an excellent price to quality and usefulness ratio. It makes working in Word more bearable and more palatable.
Strictly speaking, WordFisher is not a translation memory software. Those are very complex software using large databases to store entries, algorithms for finding and matching segments, and many other functions. However, WordFisher uses advanced Word macros to create many functions that are similar to translation memory environment.
What WordFisher calls indexing, translation memory software calls segmentation.
What WordFisher calls bilingual glossary, translation memory software calls terminology database.
What WordFisher calls bilingual corpus, translation memory software calls memory database.
Like translation memory software, WordFisher analyses a text, cuts it into segment (using similar segmentation rules, which can be customised). It can also pre-translate using bilingual glossaries and corpus either in one file or several files in the same project simultaneously. However, it doesn't find fuzzy matches. It can only look for total match.
WordFisher only does pre-translation, it doesn't extract already translated passages or terms from the glossaries or corpus while you translate. It will not automatically present you with a list of already translated passages or terms from the bilingual glossary or corpus as you translate. You would need to search them with the search function and extract them manually.
Like translation memory software, WordFisher allows you to align previous translations using its alignment tool. WordFisher's alignment tool is very easy to use.
Like translation memory software, WordFisher allows you to search your bilingual corpus and glossaries with its powerful search function in a similar way as a text retriever does. You can search several files simultaneously. It will present the results of hits by highlighting the keywords within their context (i.e. segment).
There are other differences but this section was simply to give a broad idea of the sort of CAT tool WordFisher is and what you can gain with it.
When you first start using WordFisher to translate, the first thing to do is to create a project. Various elements need to be explained before doing this.
To work with WordFisher, you need to have a source file and any reference material in electronic format. While WordFisher can search reference material in every format that Word can open (doc, txt, RTF, html, etc.), the document to be translated has to be in Word format (.doc). Thus if you receive a file to be translated in text or RTF format, you will need to convert it into a doc file first. This can be done easily in Word.
For example, if you received a text file.
If you receive a file in another format, for example a PDF file, you will also need to find a way to convert it to Word format (.doc).
If you receive a paper or faxed document, you will have to determine if it is worth your while to spend the time to convert it into an electronic file (using a scanner and OCR software).You must know how to use Windows Explorer to manage your file. It is important that you learn how to create folders, rename, copy, move files and folders, and be very comfortable with these functions before using WordFisher. You will probably have to devise a good file management system before using WordFisher. If you already have a good file management system, you may have to make some changes to accommodate this new way of working.
Before you create a project in WordFisher, you have to create a folder (in Windows Explorer) into which you will put the file or files you have to translate. (If it is a file that you have received by email, it is possible that the original file is in the attachment folder configured in your email software). Note: Only put in this folder the file or files that need to be translated; if you put any other file like reference materials, they will also be segmented (indexed) and prepared for translation by WordFisher. The tree structure in Windows Explorer should look something like this:
Why create a folder before starting to use WordFisher? When you use WordFisher for translating, you deal with a translation project. The project is associated with a folder containing the source documents. What WordFisher calls a project is simply the job you have to do (that is the text to be translated). It is easier to call it a project because one project can contain several files. For example, if you are working on a 30,000-word manual, it is possible (most likely) that the manual to be translated will have been divided into several files.
Once you have created your folder and moved your file to be translated in that folder, you can open you Word application and create a project with WordFisher. Once this is done, WordFisher will automatically create three subfolders in the folder you have just created for your project. The tree structure in Windows Explorer should look something like this:
TABL contains the bilingual (source-target) tables created by WordFisher. File names are identical to the names of the original source files (at this stage, this subfolder should be empty, it will only contain files when you will have created bilingual corpus tables).
TRAN contains the exported translations (stripped of indexes, i.e., colour codes). File names are identical to the names of the original source files (at this stage, this subfolder should be empty, it will contain files when you have finished and exported your translation).
WFPR contains the indexed source and target project documents (original file names with SRD and TRD extensions, respectively), the project's settings file (PROJECT.WF), and the replacement log file (REPLACED.DOC).
Now that you have a good idea of what WordFisher can do and what is a CAT tool,
it is time to move onto how to do it concretely. Let's start with the toolbars:
WordFisher has five toolbars and a button to activate and deactivate them. They
should appear in your menu bar in Word. If not, click on the 
button to make them active. The use of each button will be explained as we move
forward in the tutorial.
- wf
WordFisher
- wf Main
- wf Glossary Builder
- wf Quick Alignment
- wf
Project
- wf
Aligner
Note: The 
(wfWordFisherToggle)
button toggles all the other WordFisher toolbars. You can use this button to
hide the above toolbars quickly when you do not want to use them. When you click
it again, it will show those WordFisher toolbars that were shown last time.
You can select the toolbars in Word's View/Toolbars... menu. It is worth
arranging all WordFisher toolbars (except wf WordFisher) into a separate
row. Note that WordFisher shows/hides the wf Project toolbar automatically,
and you are not supposed to select it manually.
Note that some button has several functions and will present you with different
panels depending on the file that is open at the time. For example, the 
(wfQuickAlignment) button can be used to synchronise SRD and TRD files
while working on the translation project. It is also the button for aligning
files when your source and target files to be aligned are open. The 
(wfReplace) button opens the replace panel when you are in a SRD or TRD
file. When in any other file, the replace panel will also include a Pretranslate
button. All these functions will be explained later.
Is WordFisher difficult to learn?
Not if you know Word very well, can function with ease in Windows and understand the concept of translation memory. WordFisher may seem complicated or difficult at times because there are several things to configure regularly, some panels vary according to various functions. But on the whole, WordFisher is a fairly easy program to use. You have to get used to the concept of working in a new environment within Word, and not confuse functions in Word and the ones in WordFisher.
As explained above, what WordFisher calls a project is simply the job you have to do (that is the text to be translated). It is more accurate to call it a project because a single project can contain several files to be translated.
To create a new translation project (exercise 1)Note: Only put in this folder the file or files that need to be translated; if you put any other file, like reference materials, they may also be indexed and prepared for translation by WordFisher. For the Tutorial, you should only put the file wfen.doc in this folder for now.
(wfProjects) button.Note: if you click on Browse, a window will open. It may not be the standard Windows to which you are accustomed. However, this window works in similar ways. The box may first list two dots in square brackets [..] (double clicking on this will take you one level up), followed by [C:\] (the letter represents your hard drive on your computer, thus if you have more than one hard drive, you will have more than one letter in square brackets in alphabetical order (i.e. [D:\], [E:\], etc.). The folder or folders are listed in alphabetical order usually under the letter(s). To select a particular letter or folder, you need to double click on it. You can see the path writing itself automatically above the box as you choose a letter or folder. Check the name next to Directory Path:. Before clicking on OK, make sure you have chosen the right folder (the one containing the project file you wish to translate).
You will see WordFisher do something very quickly. What it does is similar to translation memory software: it analyses the text and splits it into segments indicated by colour codes (green by default) to be able to match them later.
You will see that it doesn't look like an ordinary file, it now has green codes (WordFisher calls them index). This can be quite confusing at first if you are not used to work in this type of environment. The green codes indicate where the text was split into segments. Why the funny green codes? To understand this, you have to understand how translation memory and segmentation work.
In WordFisher, indexes are specially formatted (Small Caps, Double Underline) colour strings in the form of {!--SNN--}, where NN is a unique number. Green is the colour by default. Note that once you have finished and exported your translation, the green codes of your target file will disappear leaving only the translated file in the original format.
The original source documents in the main project folder are left intact. Instead, WordFisher creates two files for each file to be translated within the same project, and assign them extension SRD for source and TRD for target. It puts them in the WFPR folder (which should be under your WFEXER folder (see File Management section).
If you click on the
(wfQuickAlignment) button while in the TRD file, it will open
the source file (SRD) while keeping the target file (TRD) opened, both in their
respective window, and arrange the two windows horizontally. At this stage,
you will see that your TRD file has exactly the same content as the SRD
file.
You overwrite the original text in the target project document (TRD), that is between each segment (green codes) while leaving the indexes unaltered.
You have to be careful because the green codes are not protected and so you can erase them (though WordFisher will warn you when you finish a project if indexes don't match and you will be able to correct them manually). It may be best not to use the Overtype function of Word.
There is one small problem that may occur and this is due to Word's "features", not Word Fisher's. When you type in Word, if you change the font, colour, etc. of characters, the next character you will type will carry these new formats over. For example, if you have a bolded word in the middle of a text and you place your cursor just after the last bolded character, the next character you will type will be bold. It is the same with the green codes in WordFisher. If your cursor is placed just after a green code, then the first letter you will type will take on the fonts and colour of the codes. To avoid this, you can select the first word and overtype your translation. This happens mostly with the first segment or perhaps in titles.
If you like working with both your source and target text open simultaneously, you will gain speed and ease by using the Quick Alignment toolbar of WordFisher as it allows to synchronise your movements between source and target files. The various buttons of the Quick Alignment toolbar are explained below.
As mentioned in the presentation of the toolbars, some
buttons have more than one function and panels depending on the file open when
clicking on it. The
(wfQuickAlignment) button is such a case.
If you click on the
(wfQuickAlignment) button while in the TRD file, it will open
the source file (SRD) while keeping the target file (TRD) opened, both in their
respective window, and arrange the two windows horizontally. At this stage,
you will see that your TRD file has exactly the same content as the SRD
file. Note that this button is also used to align files that are not TRD and
SRD (see Aligner).
After clicking 
(wfQASynchronize), the scrollbar position of the second window is synchronised
with that of the first one. Experience shows that this kind of alignment is
really useful in a number of cases. A great advantage is that you do not need
to prepare or modify the files for alignment, as is usually necessary in other
similar programs.
The 
(wfQAScrollDown/Up) arrows on the toolbar offer synchronised scrolling
of the matching documents.
To go to the next segment (exercise 2b)
Instead of overwriting, in most cases you can write the translation in front
of the source sentence. After the closing punctuation mark, click the 
(wfNextSegment) button. The rest of the segment will be deleted, and
the cursor jumps into the beginning of the next segment.
To check the source sentence after overwriting it (exercise 2c)
If you prefer to work with the target file in full window, you can still see
the source sentence once you have erased a segment after translating. The fastest
way is to click the 
(wfShowSourceSentence) button. A dialog box appears in the upper left
corner of the screen containing the source sentence. If you want to check the
context in the source file you can click the Align button, otherwise press Escape
or Enter to remove the box.
Note: If the sentence does not fit in the box, click into the box, and use the arrow keys to navigate in the source sentence. If the sentence is longer than 252 characters, or some segmentation problems are encountered in tables, WordFisher automatically splits the screen and aligns the project source and target documents.
To insert an index pair into the source and target project files (exercise 4)
If a sentence seems to be too long, you can insert an additional index at any place in the source and target documents in order to segment it further.
(wfInsertIndex) button. A new index is inserted at the cursor position,
and WordFisher opens the corresponding project document and tries to place
the cursor at the same place.
(wfInsertIndex) button to insert the index. A confirmation dialog appears.To delete a superfluous index pair from the source and target project files (exercise 30)
Conversely, if you find that WordFisher has split a segment where it should not have, you can erase that index, joining the two segments.
(wfDeleteIndex) button.Note: If the cursor is not within an index, the next index from the source and target files will be deleted after confirmation. If there is a selection, all indexes will be deleted, but only from the active document.
There will be more explanation about segmentation later on.
There are various types of references, provided by the client or collected by you:
These files can be anywhere on your hard disk. You can use WordFisher's 
(wfFind) function to search them. However, you first need to define the
files in which you want to search. This can be done in the 
(wfSetup) window.
Click the
(wfSetup) button. The panel contains two list boxes. The upper one is
the Working List. This box will list the file or files in which you want to
search. Once you have chosen the files (see below), you have to save the group.
This is done in the lower box Groups. The panel should look like this the first
time you set up a group of file to search:
As you can see, the title of the upper box is called 1 - Working List. The number in brackets next to the title indicates the number of files listed and in which you wish to search. At this stage it should be (0). Next to the title, you have a button called Save List as New Group and it should be greyed out (that is inactive). The first and only line in the Working List box at this stage should be [Empty].
The title in the lower box, just under the Working List is called 2 - Groups. The number in brackets next to the title indicates the number of groups created. At this stage it should be (0). Next to the title, you have a button called Clear Working List and it should be greyed out (that is inactive). The first and only line in the Groups box is [New Group OR Clear Working List].
To select the file or files in which you want to search:
You setup panel should look like this now:
You will see that the Clear Working List button and the Save List as New Group button are now active. In the Working List box, the file wfref-bil.doc should be the first and only file listed. It replaced the Empty.
Now that you have chosen the file you want to search, you have to create the group for it. One group can contain several files or folders to be searched simultaneously. We will add a file later. To create a new group, click on the Save List as New Group button. A small box opens and let you choose a name for your group. By default, WordFisher will give the name Group1 (current date). Give the name wfexer to the group and click OK.
You can see that your group wfexer has been added to the list. You will notice also the number in brackets in both Working List and Groups is now (1).
Click on Close, to close the window.
Now that you have setup a group of files to search, you can search it using the Find function.
To search in glossaries and documents (i.e., in the group of files selected in Setup in the previous exercise). (exercise 6)
Click the (wfFind)
button to initiate searches. If you have not selected any search string in the
document, the macro selects the word at the cursor position and puts it into
the Text Pattern 1 box. If it is not what you want to search, type your
chain of characters. Before sending your search, you have to select the Group
in which you wish to search.
The first element in the Find panel is Search the Group:. By default, WordFisher gives [Active document]. To access the list of Groups you have created (in the Setup panel), click on the arrow at the right of the box. Select the Group in which you wish to search and click OK.
You will see WordFisher do things very quickly. It is searching the chain of characters that you typed in the file or files that you have selected. Once it finished searching, it will present you with the search results.
Search results, that is, paragraphs or table rows containing the search pattern(s),
will be collected into a Word document, named FOUND.FNH. WordFisher highlights
in red the keywords. WordFisher keeps the last 25 files (Found1.fnh, Found2.fnh,
etc.), and any one of them can be retrieved after clicking History in
(wfFind).
It is very easy to jump into the corresponding document from FOUND.FNH. Position
the cursor in a word in the hit paragraph and click the 
(wfExpressJump) button. The macro opens the document and puts the cursor
before the word you indicated in FOUND.FNH.
To return to the translation after browsing the results
Click the 
(wfGoBack) button. If you select a term in the hit list before clicking
the (wfGoBack)
button, a small dialog box appears offering to insert the term into the translation.
To add documents to be searched in a Group. (exercise 7)
(wfSetup) button.When a Group is highlighted, the Clear Working List button above the Working List box changes to Append Group to List (the files and folders in the highlighted Group are appended to the end of the upper list if you click this button) and the Save List as New Group button above the Groups box changes to Update Group with List.
To add a file to the Working List.
Click on Add. Select the file you wish as we did above. Then click on Update Group with List to include the latest file in the Group. Click on Close. Then you can use the Find function to search terms as explained above.
GLOSSARY BUILDER (exercise 9-12)
This module can add entries to any existing glossary. If you want to start a new one, first create a new Word document for it, the same way your would for any other document. For example, in the File menu, click on New. This will open a new document. Then in the File menu, click on Save. Give the name you want to the file (in the Tutorial, glossaire1.doc). You can either close the document or leave it open. If you close it, WordFisher will reopen it automatically when needed.
Before adding entries to your glossary, you need to tell WordFisher in which
glossary to add the entries to. To do so, click on the 
(wfGlossaryTools) button, then click on Setup at the bottom of
the panel. The first time you create a glossary, the window should look like
this:
The first item you will see it GLOSHIST.TXT contains 0. In the box, the first and only item is Empty. This means your list of glossaries is empty. To choose a glossary, click on New button. A window will appear where you can select the glossary file you wish to build. In the Tutorial, glossaire1.doc.
Then you will see a simple dialog box, where you can define the separator and the optional stamp string.
Click OK after setting the necessary parameters.
To add entries to a glossary
(wfBuffer1).(If the buffer is not empty, a panel appears with the options of Replace, Append or [=]+Append. Using the Append option, you can put together a term from elements scattered in the text. The [=]+Append does the same, but adds the separator as well, allowing the creation of multilingual glossaries. If you add an entry you do not want in your glossary, you can simply edit it the same way you would any Word file and save the changes.)
(wfAddAndActivate) button. The glossary file will open and you can
see the entry you have just added. You can edit it if you wish. You can go
back to the translation the same way you would in a multiple document environment
in Word, either via the menu (click on Windows, then choose the document you
want) or using a shortkey (by default Word's shortkey for toggling between
windows is Ctrl+F6). If you do not need to edit the glossary and do not want
to go to the glossary file all the time, you can add the new entry in the
background by clicking the 
(wfAddInTheBackground) button. Note: The macro always checks the first
part of the entry to be added, and warns the user if it already exists in
the glossary.To add a glossary to the list of glossaries (exercise 12)
Click on the
(wfGlossaryTools) button, then click on Setup at the bottom of
the panel. You will see a simple dialog box, where you can define the separator
and the optional stamp string. Click on Change. The Glossary Builder
Setup panel appears. You can see the list of glossaries contained in the
GLOSHIST.TXT. At this stage, you should have only one called glossaire1.
To add a glossary, click on New and choose the file you want. Click OK.
Later you can choose another glossary at any time here by clicking the Change
button. You can select a glossary from the list of 60 previously chosen glossaries
(file names are stored in a text file named GLOSHIST.TXT in WordFisher's
configuration directory).
To create a table for the file in order to use the 'project table search' feature (exercise 15)
You searched a bilingual corpus table in an earlier exercise. This type of reference material was taken from previous translations. However, you may want to search the current file you are working on. For this you need to create a corpus table of what you have translated so far.
Make sure the cursor is in the last sentence you have completely translated.
Click the 
(wfProjectTools) button, then choose Create Corpus Table. Select
options and click OK. In a few seconds the table appears. You will notice
the first columns contains numbers, those correspond to the index strings (green
codes). You can go back to the TRD file by clicking the
(wfGoBack) button.
You can update that table regularly while translating and once you have finished your translation (see later).
To see how a term was translated in the project (exercise 16)
While you are in your target text translating, if you want to search a term
that you think you have already translated in the current file, select the term
or place the cursor in the word and click the
(wfFind) button, then the Proj-Tbl button. After one or two seconds
a document opens, FOUND.FNH, which shows the hits in red.
To return to the translation after browsing the results
Click the
(wfGoBack) button. If you select a term in the hit list before clicking
the (wfGoBack)
button, a small dialog box appears offering to insert the term into the translation.
To update a table (exercise 18)
You have to update the table after translating a while, in order to reflect
the changes you made. Make sure the cursor is in the last sentence you have
completely translated. Click the
(wfProjectTools) button, then choose Create Corpus Table. Choose
Current File and click OK. Leave the Append to 'To Index'
option selected and press Enter or click the OK button. In a few
seconds the updated table appears. (It is not necessary to update the table
after each paragraph. You can do so hourly.) You can go back to the TRD
file by clicking the
(wfGoBack) button.
As mentioned in the presentation of the toolbars, some buttons have more than
one function and panel depending on the file open when clicking on it. the
(wfReplace) button is such a case. Open your translation file TRD if
it is not already open. Click on the
(wfReplace) button. It will activate the wfReplaceInProject panel.
With this panel, you can replace strings in the active project target file or
all project target files.
To replace a string in project files (exercise 19)
Place the cursor in the word you wish to search (WordFisher will select it
automatically, you have to select only multi-word strings). Click the
(wfReplace) button. If the word automatically selected by WordFisher
is not the string you want, in the Find What: box type the string you
want to find. Then in the Replace With: box, type the string that will
replace the string in the Find What: box. (Note: changes will be irreversible,
unless you turned on Tools/Revisions.../Mark Revisions while Editing,
which can be done by checking the Mark Revisions On box). You have the
choice of replacing the string in the current file or in every file in the project.
If you click on the File button, the string will be replaced in the active
file only, while clicking on the Project button will replace it throughout the
project. Note: WordFisher only does global replacement; it
has the same limits with global replacement as does Word.
To recreate the table after modifying some parts in the translation (exercise 20)
Make sure that the cursor is somewhere in the last sentence you have translated
completely. Click the
(wfProjectTools) button, then choose Create Corpus Table. Select
the Recreate until 'To Index' option.
To close the project (exercise 21)
Make sure the cursor is in the last sentence you have completely translated.
Click the
(wfProjects) button and choose the Close button. All SRD/TRD project
documents will be saved and closed. The project bar will disappear.
WordFisher's Aligner does not touch the original source and target files. It creates two duplicates of the original source and target files and creates a table with the corresponding sentence pairs. In order to speed up the revision process, it splits large files into smaller units, and creates several tables instead of a huge one that would be difficult, or impossible, to handle in Word. While creating the table(s), it takes into account the segmentation rules and exceptions you collected in special files in order to apply your additional segmentation rules or intelligently ignore abbreviations ending with a period (this is explained later).
Once the table is created, you need to check and finalise the alignment. WordFisher provides all the necessary tools to do this effectively. Its wf Aligner toolbar contains buttons for joining and splitting cells, as well as for deleting cells and rows.
To create the alignment table
Open the source and target documents to be aligned. Then click the
(wfQuickAlignment) button, select the source document in the left window,
and the target document in the right window.
Then click the Aligner button. A dialog box appears. Choose the boundaries options then click the Align button.
Now an empty document appears and you will be asked to save it for the alignment table. Choose an appropriate folder and file name, and save the file. First, the program cleans the files. In order to speed up the alignment process, it splits large files into smaller units, and creates several tables instead of a huge one that would be difficult, or impossible, to handle in Word. Depending on the size of the files and the speed of your computer, this process will take a few seconds, though rarely more than a minute.
When the table, which may consist of several parts, has been successfully created, a message box appears saying so, and the wf Aligner toolbar becomes visible.
Now you should go through the table, from the very beginning, and check whether the sentences in the left and the right cells are exactly matching. The toolbar offers all necessary tools to quickly correct any errors. Once you have finished editing the matches, you may need to erase extra cells at the end of your table.
Tip: Instead of the toolbar buttons, you can start the macros by using keyboard shortcuts. Shortcut keys can be assigned in Word's Tools/Customize.../Keyboard. Aligner's macros begin with "wa".
Note: The editing functions can also be used with other tables, even if they contain more than two columns.
To join a cell with the cell below it
Put the cursor anywhere in the upper cell, and click the 
(waJoinCells) button. The content of the lower cell will be copied at
the end of the upper cell, and all the cells under this cell shift one cell
upwards.
Note: By default, the contents of the two cells will be separated
by a space character. You can change this at any time the following way: Click
the
(wfQuickAlignment) button, then choose the Aligner button.
Change the setting under Edit Options - Join Cells with as you wish, then click the Close button.
To split a cell
Position the cursor exactly where you want to separate the cell, and click
the 
(waSplitCell)
button. An empty cell will be created below this cell, the text from the position
of the cursor until the end of the original cell will be copied into this new
empty cell, and all cells below will shift one cell downwards. All space characters
will be deleted from the beginning of the inserted cell.
To swap the cell with the cell below it
Put the cursor anywhere into the cell, and click the 
(waSwapCells) button.
To delete a single cell
Put the cursor anywhere into the cell to be deleted, and click the 
(waDeleteCell) button.
To delete cells
Select the cells to be deleted, and click the
(waDeleteCell) button.
To delete a single row
Put the cursor anywhere into the row to be deleted, and click the 
(waDeleteRow) button.
To delete rows
Select the rows to be deleted, and click the
(waDeleteRow) button.
To merge tables
WordAlign splits the alignment table for longer documents. Tables are preceded
by a "Part n of N" string, indicating that the following table is the
nth of all N tables. They need to be merged during the revision
process. When you reach the end of a table during the revision process, click
the 
(waMergeWithNextTable) button to merge the table with the next one. Then
continue revising.
To undo an action
Use Word's Undo command in the Edit menu or click the button.
Note: It may be necessary to click three or four times this button to undo a split or join operation.
To clean a table
In order to remove all tab characters, double spaces and leading/closing spaces
from all table cells click the 
(waCleanTable) button. The same result can be achieved if you set the
Clean Corpus Table after Alignment option in the above dialog box before
alignment.
To reopen the project (exercise 24)
Click the
(wfProjects) button and choose the Open button. A dialog box appears
with a list of the documents in the project. Choose the document you want to
work on and click the Open button. The document will open at the position
you left it last time. The project bar will reappear.
You can edit (add, erase and modify entries) a glossary any time in the document itself. Just open your glossary file and edit it as you would any other Word file.
You can also add entries to an existing glossary from another file (the same
way it was explained in the Glossary Builder section). You
may need to select the right glossary file before editing from another file.
WordFisher keeps in memory the last glossary that you have used or worked on.
To change a glossary file, click on the
(wfGlossaryTools) button, then click on Setup at the bottom of
the panel. You will see a simple dialog box, where you can define the separator
and the optional stamp string. Click on Change. The Glossary Builder
Setup panel appears. You can see the list of glossaries contained in the
GLOSHIST.TXT. Select the one you want and click OK. You can select a
glossary from the list of 60 previously chosen glossaries (file names are stored
in a text file named GLOSHIST.TXT in WordFisher's configuration directory).
To add a new document to the project (exercise 28)
(wfProjects) button, and choose the Close button. The project
bar will disappear.
(wfProjects) button and choose the Open button. WordFisher will
detect the new document, and it will prepare it for the project.Note: You may add any number of files to the project at any time.
Segmentation is the division of text into translation units, and it determines what strings will be presented in the rows of corpus tables (paragraphs, sentences, smaller units). WordFisher contains fully customisable segmentation rules, as well as separate tools to add or remove indexes any time in a translation project, making it one of the most flexible systems in the market.
The same segmentation rules/exceptions are used in wfProjects when a new project file is added to the project and in the Aligner in wfQuickAlignment when the source and target files are segmented for alignment.
WordFisher's wfProjects module inserts indexes at segment boundaries, while Aligner splits the documents into segments at the boundaries, and puts the corresponding segments into parallel table cells.
You can add your segmentation rules using a special file, RULES.DOC. Open it
by clicking the rules.doc button or choose Options in
(wfSetup) and click the rules.doc button.
To define your own segmentation rules
You can define further rules by adding strings that mark a segment boundary into RULES.DOC. You can use special combinations for characters found in Word's Find dialog, such as ^w for white space. For example, add the following line into RULES.DOC if you want full stop plus a quotation mark to end a segment:
."
By default, segments will be split before the string defined here. In order to change the split position, start the definition by SPLIT@n, (including the comma), where n is a number between 0 and 9. For example, to split after the second character in the above example, write:
SPLIT@2,."
Lines beginning with # will be ignored when processing this file. Thus you can temporarily leave out certain rules without deleting them.
To define exceptions to segmentation rules
WordFisher assumes that a word ending with a period always marks the end of
a sentence. In order to exclude abbreviations that end with a period like Mr.,
Ms. etc., you should collect the most frequent abbreviations in the source and
target documents into a special file, EXCLUDE.DOC. Open it by clicking the exclude.doc
button or choose Options in
(wfSetup) and click the exclude.doc button.
Note: Abbreviations are case sensitive. Special character combinations, such as ^w cannot be used here. However, they can begin with a space, for example, " p." (without the quotation marks). Lines beginning with # will be ignored when processing this file. Thus you can temporarily leave out certain rules without deleting them.
Tip: You can check abbreviations by using Word's Pattern Matching
in Edit/Find..., and searching repeatedly for the following string:
". [a-z0-1]" (without quotation marks). This will find periods that are followed
by a space and lowercase letters or a digit. You can add abbreviations to your
exclude list easily. Just select it, together with its closing period, and click
the 
(waAddToExcludeList) button on the wf Aligner toolbar.
(wfProjectTools)
is the project's tools centre.
Project Statistics displays a customisable table containing character/word/paragraph counts of the source and target files (when the target files have been exported).
During pretranslation a list of source language terms and expressions are replaced by their corresponding target language equivalents. This list can be any appropriate glossary or a bilingual table.
To pretranslate a project using a glossary (exercise 33)
If the client gives you a glossary for the job, or you have a glossary from a previous job, you can use these glossaries for pretranslating a project.
(wfGlossaryTools) button, choose the proper Source/Target Separator.
A single character must separate the source and target expressions. If they
are separated by two or more characters, for example " = " (with spaces),
then replace them in Word making sure that only a single character separates
them. It works better if you use the default separator of WordFisher, that
is the pipe separator | (code ANSI 0124). To get this symbol, press the ALT
key on the left of the spacebar and, while keeping the ALT key pressed, press
the keys 1,2,4 consecutively on the numerical pad (NumLock has to be activated).
(wfReplace) button, then the Pretranslate button. If the Mark
Revisions On option is checked, all replacements can be reviewed (accepted
or rejected) later using Word's own Tools/Revisions function (since
WordFisher does a global replacement it is better to check the Mark Revisions
On options, as global replacement can create undesirable translation).
Then click Replace.As mentioned in the presentation of the toolbars, some buttons have more than
one functions and panels depending on the file open when clicking on it. The
(wfReplace) button is such a case.
To pretranslate a project using a bilingual corpus (exercise 36-37)
You can also use an aligned file of previous translations to pretranslate in the same way you have done above with a glossary file. However, you first need to convert the table into WordFisher's glossary format. You can do this in Word by using Convert table into text in the Table menu, using the pipe as separator (|) (code ANSI 0124). To get this symbol, press the ALT key on the left of the spacebar and, while keeping the ALT key pressed, press the keys 1,2,4 consecutively on the numerical pad (NumLock has to be activated). If you do this with a bilingual corpus created with WordFisher after finishing a translation), you will need to erase the first column which contains the numbering of the index strings (it was added automatically by WordFisher when it created the bilingual corpus) before converting the table into text.
Once your file is converted into a glossary format, you can save it under a new name. Then use it to pretranslate following the same procedure as with the glossary file above.
Note: Some segments may be too long for WordFisher to handle. WordFisher will open a box warning you that your string is too long. You will have to erase or shorten those strings. Then try again to pretranslate.
You can use the corpus table of the current project created by WordFisher to help you edit your translation. You will then view your translation in a table format with the source segments on your left and the corresponding target segments on your right. This may seem a clumsy way to edit a file at first, but you may find some efficiency in doing it that way. Since every segment is aligned, you can easily see if you have omitted to translate a passage. If you need to change something in your translation, you will do so in the TRD file and not in the corpus table. WordFisher has buttons to take you back and forth easily. This way of editing may be better if you don't have too many changes to make. That is, it could be your last editing stage.
To revise your project (exercise 42)
(wfProjectTools) and choosing Create Corpus Table. The time
it takes depends on the size of the project and the speed of your computer.
WordFisher automatically closes all the files after doing this.
(wfExpressJump) button. After correcting the error you can return to
the table using the
(wfGoBack) button, and you can continue to edit your file.To check for consistency (exercise 44)
(wfGlossaryTools) button, choose the proper Source/Target Separator.
Make sure the radio button next to Sort for Pretranslation or Consistency
Check is checked (it is by default). Then click OK.
(wfGlossaryTools) button again, check the Consistency Check
box, and select the [Project Tables] group. Click OK.
(wfExpressJump) button (the same way you did when revising the file
with a corpus table).
(wfGoBack) button, and you can continue checking.Finishing a project
Once your have completely finished and edited your translation to your satisfaction, you need to export it. WordFisher will then take out the green codes and automatically save your finished translation in the Tran folder it created at the beginning of the project. The file will have the same name as the original file you received to translate.
(wfProjectTools) function. They will be placed into the TABL
subfolder. Once done WordFisher will close all files automatically. (wfProjectTools)
function. They will be placed into the TRAN subfolder. These are the
files you can send to the client. You have the option to export all files
in a single process.Participating in projects requiring a Translation Memory (TM) program
Almost all TM programs are capable of importing two-column tables or tab-delimited text files into their databases. WordFisher has a macro, waConvertTable, to convert a project table to Trados import format. This means that you can send the exported project tables to the client, who will be able to integrate them into the TM databases. As a result, you are able to deliver a job with WordFisher just as if you were the owner of a big TM system.
To convert a table for Trados import (exercise 48)
Click the 
(waConvertTable) button, if you want to feed the result of the alignment
into a Trados translation memory.
If you go to Windows Explorer in your project folder (WFEXER) you will se that now you have the original files in the main folder of the project (WFEXER), the finished translations in the TRAN folder (same file names) and the project tables in the TABL folder (same file names). You can move all these files into the folders where you usually collect translations and reference files for that client.
Note: Since the files all have the same name, you may need to rename them. For example, the translated file you will send to your client, you may want to add the language into which it was translated e.g. FR for French. The corpus table in the TABL folder are reference material, so you could add REF in their name. You may also wish to check the files in the WFPR folder to make sure there is no glossary files that you want to keep before erasing the folders.
After moving these important files in Windows Explorer, you can go back to
Word and remove the project from the project list if you wish in the
(wfProjects) function by clicking on Remove. This will not automatically
remove the subfolders and the files in your WordFisher project. It simply removes
it from its list of projects. You can delete the project's folder together with
its subfolders and files in Windows Explorer.