While we're on the topic of badly misspelled words, Sergio
Lozano wrote this:
Is there a way to remove words from Word's
spellcheck dictionary? I don't know how many times I have typed tow when I
meant two. I would rather have tow be flagged as a misspelled word, though
putting in an autocorrect entry is a good workaround. Nonetheless, I'm curious
as to whether you can mess with Word's default dictionary.
That's a good question and the quick answer is
that it is not possible to mess with the Word spelling dictionaries directly.
The emphasis here is on directly -- while you can't modify the main
dictionaries, it is possible to teach Word to ignore certain words by entering
them into an "exclusion"
dictionary. An exclusion dictionary
causes Word's spelling engine to ignore the entries in the main dictionary for
the words it contains. This means that you can tell Word to treat certain words
as misspelled, even though your spelling dictionary thinks they are OK.
(Sergio's example was one instance where this would be handy; other examples
might be theatre vs. theater, public vs. pubic, or abut vs. about).
First the procedures for
Office 2003 and below. In a document in Word or in a text editor,
enter the words you would like to exclude in lowercase letters, each followed
by a hard return. Here's what the words we mentioned above would look like:
tow
theatre
pubic
abut
Once you have added the
most annoying offenders, save the file to C:\Documents and Settings\<user
name>\Application Data\Microsoft\Proof (Word 2003 and below and Windows
XP) or C:\Users \<user name>\AppData\Microsoft\Proof (Word
2003 and below and Windows Vista).
Save this as a text file
(if you used MS Word, you'll need to select Text Only (*.txt)) under Files
of type in the Save as dialog and choose the following naming
convention: mssp3<two or three digit language code>.exc.For Spanish this would be
mssp3es.exc, French would be mssp3fr.exc, etc. (you can find the correct naming if you
look for the .lex files that are stored
under C:\Program Files\Common Files\Microsoft Shared\Proof).
For Word 2007 it
is a little less complicated because here the file already exists (it is
automatically created the first time you add a word to the custom dictionary).
You will be able to find a file with the long file name ExcludeDictionary<LanguageCodeLanguageID>.lex in the folder
C:\Documents and
Settings\<user name>\Application Data\Microsoft\UProof (Windows XP) or C:\Users \<user
name>\AppData\Microsoft\UProof (Windows Vista). (To see those
files and folders, you will need to make sure that you have the display of
hidden files and folders enabled under Start> (Settings>)
Control Panel> Folder Options> View.)
The U.S. English file
name would look like this: ExcludeDictionaryEN0409.lex. You will simply
have to open it in Notepad or another text editor (or Word) and
add the lower-case terms that you want to have excluded as described above.
Sound complicated? Well, just imagine how
complicated it was to research these things to present them to you. (-: Actually,
if you just try to follow your combination of Office and Windows
versions it isn't that hard, and the reward can be sweat -- I mean, sweet.
While we're at spell-checking, here are a couple
of other helpful hints.
Most of you know not to use the Automatic
Language Detection that you can activate in the Language dialog in Word
(you can access that by clicking on the language identifier down on the task
bar in Word). This was one of those things that were meant well but
brought about all kinds of issues, including bloated documents and a more
difficult processing of these documents in some TEnTs (translation environment
tools). What I've been looking for ever since Word 97 and never adequately
understood is the option right above it: Do not check spelling or grammar.
It's entirely possible that I was alone in this misunderstanding, but I always
took that to refer to the complete document (and, yes, I always wondered why it
didn't work right). Well, it only works for the passage that you select, of
course, meaning that you can define certain parts (or all parts if you select
all) to be not spell-checked.
Here's another one: Many of us don't do all our
translation work in an Office program; we use an independent TenT
instead. While some tools such as Wordfast, Multitrans, and MetaTexis
still primarily use the Word interface for their translation activities,
they are all about to go to an independent editor, as Trados has done
with TagEditor, or as tools such as Déjà Vu, Heartsome, MemoQ,
SDLX, Transit, and tons of other ones have always done.
The problem has traditionally been that it was a
shame to have to use the third-party spell-checking systems (for a relatively
limited number of languages) that come with those tools. Most of the vendors
had chosen the third-party spelling engine by Wintertree, which
typically was not up to par with what Word had to offer. Plus, the
entries in the custom dictionaries were not all automatically transferred, so
all the work that you might have spent to add hundreds of terms to the Word
spell-checker was not available when checking with SDLX, Déjà Vu,
Trados TagEditor, or any of the other tools that used Wintertree.
If you are intent on using the Wintertree
spell-checker, there is a trick to selecting the custom.dic file from Word:
in Déjà Vu and SDLX, select Dictionaries> Import in the
Check Spelling dialog, change the Files of Type to External
Files, and import the .dic dictionary that is typically under C:\Documents
and Settings\<user>\Application Data\Microsoft\Proof. In Trados
TagEditor, select Tools> Plug-ins> WinterTree Spelling Checker, select Properties, Custom Dictionaries, and Add.
Select the above described custom.dic, highlight it, and click on Set as default.
But all this was generally not good enough. So
most of the tools today (including the three mentioned above) have embraced one
of two strategies: they either allow you to use MS Word's spell-checker directly
(provided, of course, that it is installed!) or they use the same engine that OpenOffice.org
uses.
Here are some examples: MemoQ, OmegaT,
Swordfish, and Heartsome all use the current Hunspell engine
of OpenOffice, and Transit uses MySpell, the earlier
version of OpenOffice's spell-check engine. The benefit of this is that
these are free and the number of supported languages is not only very large but
also consistently growing.
And with Halloween coming up and all, I can't help
pointing to this silly cartoon again: www.internationalwriters.com/spellcheck.jpg.
