Translate Toolkit & Pootle guide

guide:application_specific

Sat, 05 Jan 2008 17:55:05 +0000

1122884253	current
Line 1:	Line 1:
	+	====== Applications that need non-translation localisation ======

	+	These are applications that have country specific or language specific configurations. Ie there
	+	are things that you need to change to make the application more useful for user of your
	+	language or users in your country.;
	+
	+	The list merely highlights the issue but most of these have not been elaborated.
	+
	+	===== Country Specific =====
	+
	+	* KOrganizer --- public holidays
	+	* Evolution --- public holidays
	+	* Any Internet dialup logging
	+	* Mozilla dates and times (might be locale specific)
	+	* OpenOffice date and tiem formats (might be locale specific)
	+	* KDE country and regions selector in locale
	+	* KDE country.desktop file
	+	* Mozilla region packs
	+	* Default printer page sizes -- should be set in locale
	+	* Any weather application or user eg RSS feeds in Evolution
	+	* Validate weather service namign conventions might be out of date
	+	* Spellcheckers with:
	+	* Geographics names: province, cities, mountains, rivers, airports, harbours
	+	* People: statesman, hero's, common names
	+	* Animals: common names for fauna and flora
	+	* City locations for things such as KStars
	+	* Country specific shortcuts for Konqueror
	+	* Mozilla shortcuts already built
	+	* Mozilla local search engines and defaults already set
	+
	+	===== Language Specific =====
	+
	+	* KTuberling --- language wav files
	+	* Locale files
	+	* Mozilla, Konqueror --- default language preferences set
	+
	+	===== Both Language and Country specific =====
	+
	+	* OpenOffice --- locale files

guide:bootup

Sat, 05 Jan 2008 17:44:45 +0000

1122036167	current
Line 1:	Line 1:
	+	====== Direction setting based on the installation and bootup process ======
	+
	+	This document looks at the whole Linux installation and bootup process with these main sections:
	+
	+	- Installation
	+	- Booting
	+	- Init 5 (GUI)
	+	- Init 3 (Text)
	+
	+	The bootup and installation process are very visual and give a user the first impression of
	+	the level of localisation in their software. If we can give them a good impression with less
	+	work then it becomes easier to create a fully translated Linux distribution.
	+
	+	The idea is to layout the dependencies and try place translatable components in
	+	some relative order in the process. This is done so that we for instance can place
	+	the language locale file at the correct point in the bootup process. And thus at the
	+	correct point at which it becomes important to a localisation project.
	+
	+	FIXME This document is based on Fedora but probably applies to any Linux distribution.
	+
	+	===== Installation =====
	+
	+	^ Stage ^ Messages ^ Description ^
	+	\| CD boot messages \| (unkown) \| \|
	+	\| rhpl \| 134 \| lists mice keyboards and exceptions might be needed in bootup as it mentions failed mouse detection and not being able to do graphical install \|
	+	\| Comps \| 158 \| package groups and some descriptions used by anaconda I presume \|
	+	\| Anaconda installer \| 1473 \| \|
	+	\| Anaconda online help \| 263 \| \|
	+	\| Firstboot \| 73 \| \|
	+
	+	===== Booting =====
	+
	+	^ Stage ^ Messages ^ Description ^
	+	\| Grub \| - \| not sure how to localise (localisation not implemented only scheduled post 1.0 release) \|
	+	\| kernel \| \| The Linux kernel messages are localisable \|
	+	\| Locale files \| - \| needed by some initscripts to display date \|
	+	\| Initscripts \| 29 \| only 'sysinit' and some 'init.d/funtions' \|
	+	\| rhgb \| 29 \| \|
	+	\| Kudzu \| 70 \| \|
	+
	+	END OF BOOT PROCESS the next stage depends on the run level, most end-users will never see run level 3
	+
	+	==== Init 3 ====
	+
	+	^ Stage ^ Messages ^ Description ^
	+	\| /bin/login \| \| found in util-linux - should try and extract only /bin/login strings \|
	+
	+	==== Init 5 ====
	+
	+	^ Stage ^ Messages ^ Description ^
	+	\| GDM \| 695 \| lots of bumf but hard to seperate it out \|
	+	\| Wallpaper \| \| \|
	+	\| Fedora \| 5 \| artwork \|
	+
	+
	+	===== Post Booting =====
	+
	+	The booting process will take a computer to the place where the user is required to login. This section looks at what items
	+	are seen just after the user logs in.
	+
	+	==== Init 5 ====
	+
	+	* Wallpaper (if different from login wallpaper)
	+	* Fedora - menu's
	+	* GNOME / KDE
	+	* Kicker
	+	* K Menu bar general
	+	* Clock
	+	* Desktop
	+
	+	==== Init 3 ====
	+
	+	* /etc/issue not really possible to change easily
	+
	+	===== Config Tools =====
	+
	+	Once logged in a user will interact with the standard Free and Open Source applications that are usually localised
	+	outside of the distribution community. However each distribution has its own set of configuration tools that
	+	need localisation.
	+
	+	FIXME Create some category of which configuration tools are more important to localise then others.
	+
	+	On Fedora all the user centric and hardware related system-config tools constiture 7691 words of translation.

guide:british_english

Sun, 27 Nov 2005 07:44:33 +0000

1123107625		current
Line 29:		Line 29:

	<code bash>		<code bash>
-	pofilter -t stopwords en_GB en_GB-check	+	pofilter -t musttranslatewords --musttranslatefile=en_US-words-to-change en_GB en_GB-check
	</code>		</code>

-	FIXME stopwords was never committed to pofilter so we'll need to add that	+	If a msgid contains any of the words listed in the files //en_US-words-to-change// then these will be extracted to en_GB-check.
-		+
-	This will extract words to en_GB that potentially have American spellings.	+
	Correct them and then run:		Correct them and then run:

guide:cafetran

Thu, 23 Aug 2007 14:55:06 +0000

1187880517		current
Line 1:		Line 1:
	====== Cafetran CAT tool ======		====== Cafetran CAT tool ======

-	Cafetran is a CAT tool with translation memory. It is written in Java. Cafetran is not free, but a time-limited trial version can be downloaded from its web site. What makes Cafetran interesting is that it uses XLIFF for its native TM format. Documents translated in Cafetran automatically generate an XLIFF file as you go along.	+	Cafetran is a CAT tool with translation memory, glossary lookup, and web searching. It also supports sourceless translation using an autocomplete function.

	http://www.cafetran.republika.pl/		http://www.cafetran.republika.pl/
		+
		+	{{guide:cafetran1.jpg}}
		+
		+	Cafetran is written in Java. Cafetran is not free, but a time-limited trial version can be downloaded from its web site.
		+
		+	What makes Cafetran interesting is that it uses XLIFF for its native TM format. Documents translated in Cafetran automatically generate an XLIFF file as you go along.
		+
		+	Cafetran is fairly complicated to use, and the menu bar is very non-standard. The help file contained on the web site is not very comprehensive or well organised.
		+
		+	{{guide:cafetran2.jpg}}
		+
		+	Source files include HTML, OpenDocument, WordML (Microsoft Word 2007), etc.
		+
		+
		+

guide:calenders

Sat, 05 Jan 2008 17:43:30 +0000

1121973107	current
Line 1:	Line 1:
	+	====== Localising Calenders ======
	+
	+	===== KOrganizer =====
	+
	+	[[::kde_desktop\|KDE's]] KOrganizer stores country holidays in files designated holiday_XX where
	+	XX is the iso639 country code. In [[http://websvn.kde.org/\|KDE SVN]] these are located in
	+	kdepim/korganizer/plugins/holidays/holidays/holiday_XX

guide:clunky_glossary_creation

Mon, 24 Mar 2008 18:02:49 +0000

guide:common_cat_tools

Thu, 02 Aug 2007 15:02:03 +0000

	current
Line 1:	Line 1:
	+	====== Common CAT tools ======
	+
	+	Translators often translated documents using so-called CAT (computer aided translation) tools. Some of the more common CAT tools include Wordfast, OmegaT, Trados, Déja Vù, Metatexis, Isometry, and others. These tools are typically used to translate ordinary documents, but some of them can also be used to translate l10n formats like PO, CSV, XLIFF and other bilingual formats.
	+
	+	There are two types of CAT tools, namely those that work directly on the source text, and those that work indirectly on the source text by extracting the text from it.
	+
	+	===== Some CAT tools =====
	+
	+	==== OmegaT ====
	+
	+	OmegaT is written in Java and is GPL. It can translate OpenDocument files, well-formed XHTML, Java .properties files, key=value files and plaintext files. OmegaT extracts the text from the source documents, and the translator translates the text within the OmegaT environment. Text formatting is dealt with through special OmegaT tags.
	+
	+	==== Déja Vù ====
	+
	+	[[http://www.atril.com/\|Déja Vù]] is propriatory software (EUR 990 per licence). It can translate many other propriatory formats and open formats, including Gettext PO. It works, similar to OmegaT, by extracting text from source files so that the translator works within the program's own environment.
	+
	+	The disadvantage of programs like Déja Vù and OmegaT is that they can only handle existing, well-known text formats. The advantage is that they usually handle these formats very well or comprehensively.
	+
	+	==== Wordfast ====
	+
	+	Wordfast is a Visual Basic macro that runs inside Microsoft Word, and it is propriatory (EUR 250 per licence). It translates any file that can be opened in Microsoft Word, Excel and PowerPoint. It does not extract text, but instead it selectively allows the user to translate the portions of the file that needs to be translated.
	+
	+	==== Trados (bilingual RTF) ====
	+
	+	The way Trados handles files, is the same as the way Wordfast does it. In fact, Trados and Wordfast can read each other's files -- that is, the bilingual RTF files. This is the file type that is most useful for new or rare file formats. Trados is proprietary
	+
	+	==== Trados (TagEditor, TTX) ====
	+
	+	A newer method used by Trados is called TagEditor (it produces TTX files). TTX files are bilingual XML files that are opened translated in TagEditor, which is basically a user-friendly XML-viewer (but it can only view TTX XML files). In this sense TTX is similar to XLIFF or PO, because it is a bilingual file which is generated from the original file.
	+
	+	===== Features of CAT tools =====
	+
	+	CAT tools generally have additional features prized by translators, such as the ability to capture and use a translation memory, automatic on-the-fly fuzzy matching, automatic glossry recognition, various keyboard shortcuts to improve speed, and power reference search facilities.
	+

guide:comparingfiles

Thu, 04 Oct 2007 07:27:30 +0000

1145173551		current
Line 39:		Line 39:

	You can find [[http://wdiff.progiciels-bpi.ca/index.html\|more information on wdiff]].		You can find [[http://wdiff.progiciels-bpi.ca/index.html\|more information on wdiff]].
		+
		+	===== GUI software =====
		+
		+	==== Winmerge (MS Windows) ====
		+
		+	http://winmerge.org/
		+
		+	To compare two files, drag and drop one of them onto the Winmerge icon, and Winmerge will prompt you for the other file.
		+
		+	You can use Winmerge in combination with Tortoise SVN. In the SVN tree, right-click the folder with changes in it, and select Tortoise -> Check for Modifications. In the Modifications dialog, where modified files are displayed, simply double-click a file to have it opened in Winmerge.
		+
		+	You can make changes to a file in Winmerge.
		+
		+	Winmerge requires a BOM to recognise a Unicode file, so if your Unicode files don't have BOMs (which is the case with Toolkit produced files), you have to tell Winmerge to "fall back" to Unicode when opening a "non-Unicode" file (no, I'm not joking). Go Edit -> Options -> Codepage -> Custom Codepage and type in 65001 (which is the codepage for UTF8).
		+
		+	Winmerge can also be used to compare trees and directories.
		+
		+

guide:contributing

Sat, 05 Jan 2008 17:55:40 +0000

1122639740	current
Line 1:	Line 1:
	+	====== Contributing ======
	+
	+	We welcome contributions - if you are a localiser then this is your document.
	+	Please think of this as a living document. If you are using part of it and
	+	find that it is out of date then please update it so that others don't have to
	+	go through your pain.
	+
	+	If you have a new section to contribute or some corrections then please create
	+	an account and make changes. If you feel it might be controvercial then
	+	please email [[translate-devel@lists.sourceforge.net]]. If you have some criticism
	+	we also welcome it but if you accompany it with the changes then you are more
	+	likely to be listened to (its easy to complain, a little harder to contribute
	+	meaningfully)
	+
	+	We [[::credits\|credit]] those who contribute whole sections and significatant update.

guide:copyright

Sat, 05 Jan 2008 18:26:14 +0000

1199557574		current
Line 1:		Line 1:
	====== Copyright ======		====== Copyright ======

-	The localisation guide is released under the Creative Commons [[http://creativecommons.org/licenses/by-sa/2.0/\|Attribution-ShareAlike]] license.	+	The localisation guide is released under the Creative Commons [[http://creativecommons.org/licenses/by-sa/3.0/\|Attribution-ShareAlike]] license.

guide:costing

Sat, 05 Jan 2008 17:56:10 +0000

1122904368	current
Line 1:	Line 1:
	+	====== Budgetting for Paid Translators ======
	+
	+	Although we encourage the volunteer model there are projects that are funded
	+	and there are projects that receive funds from time to time to do a specific
	+	translation. This gives a rough outline of how to cost such a translation.
	+
	+	===== Calculation =====
	+
	+	Words = untranslated msgid words + fuzzy msgid words
	+	Cost = Words to translate * Rate per word
	+	Days = Words to translate / Words per days
	+
	+	==== Rule of thumb ====
	+
	+	Weeks = Strings to translate / 2000
	+	Days = Srings to translate / 400 or 500
	+	( Use 400 for large project 500 for a small project. This then evens out
	+	lost days or slow progress due to complicated section.)
	+
	+	Strings = Words / 3.5
	+	(This varies but is helpful)
	+
	+	===== Calculating Costs =====
	+
	+	FIXME convert these numbers to dollars or other 'international' currency
	+
	+	Translators will quote based on the number of source words:
	+
	+	R400 / 1000
	+	R40 / 100
	+
	+	which calculates to R0.40 per words
	+
	+	Note: these are just for illustration purposes
	+
	+	Using [[toolkit:pocount]] determine the number of source words. So this is the number of
	+	words in untranslated strings plus the number of words in fuzzy strings.
	+
	+	Words = untranslated + fuzzy
	+	Words = 40 000 + 5 000
	+
	+	Note: make sure you count the msgid or source words not the translations.
	+
	+	You final costing would then be:
	+
	+	Cost = 45 000 * 0.4
	+	= R18 000
	+
	+
	+	===== Calculating Time =====
	+
	+	Assuming that you need to translate 45 000 words do the following. Either,
	+
	+	* Find out how many words the translator can translate per day
	+
	+	1300 words per day
	+
	+	Days = 45 000 / 1300
	+	= 35 days
	+	= 7 weeks
	+
	+	* Convert to the rule of thumb guides above
	+
	+	Strings = 45 000 / 3.5
	+	= 12 857
	+	+- 13 000
	+
	+	Days = 13 000 / 400
	+	= 33
	+
	+	Weeks = 13 000 / 2000
	+	= 6.5 weeks

guide:creating_glossaries

Mon, 24 Mar 2008 18:09:14 +0000

1206381113		current
Line 7:		Line 7:
	===== Resources =====		===== Resources =====

-	* [[http://www.it46.se/kipot/\|KiPot]]	+	* [[http://www.it46.se/show_entry.php?id=128&lang=es\|KiPot]]
	* [[http://www.christophermayo.com/articles/2004/makeglossary.html\|How to make a simple glossary]]		* [[http://www.christophermayo.com/articles/2004/makeglossary.html\|How to make a simple glossary]]

Line 29:		Line 29:
	would lead to an enormous amount of mails sent back and fourth between the		would lead to an enormous amount of mails sent back and fourth between the
	projects participants. We needed an online communication tool that would		projects participants. We needed an online communication tool that would
-	facilitate the communication, and therefore, [[http://www.it46.se/kipot/\|KiPot]] was developed."	+	facilitate the communication, and therefore, [[http://www.it46.se/show_entry.php?id=128&lang=es\|KiPot]] was developed."

	===== Hacking a glossary from existing translation =====		===== Hacking a glossary from existing translation =====

guide:credits

Sat, 05 Jan 2008 17:56:38 +0000

1122640704	current
Line 1:	Line 1:
	+	====== Credits ======

	+	These are the people and groups who have made this document possible:
	+
	+	===== Contributors =====
	+
	+	The Writers, section contributors:
	+
	+	* [[http://www.translate.org.za/\|Translate.org.za]] --- The foundation which made this possible in the first place.
	+	* [[dwayne@translate.org.za\|Dwayne Bailey]] --- Capturing the experience of Translate.org.za to ensure that the pain for other localisers is considerably reduced.
	+	* [[javier@khmeros.info\|Javier SOLA]] --- Documenting the usage of the tools, instructions aimed at newbies and at Windows users and the guide for OpenOffice.org localisation
	+
	+	Page contributors:
	+
	+	* [[novican@mt.net.mk\|Novica Nakov]]
	+
	+	The techies.
	+
	+	* [[david@translate.org.za\|David Fraser]] --- Creation of mikiwiki to change this from a text only to a simple HTML document. Plus migration to the current wiki and various Pootle documentation.
	+
	+	Helpful input
	+
	+	The people whose input has helped shape the document.
	+
	+	* [[http://tacticaltech.org/\|Tactical tech]] --- Who seeded the idea of capturing this information.
	+	* [[http://www.doc.gov.za/\|South African Department of Communications]] --- Whose funding of Translate.org.za made the writing and capturing of this information possible.
	+
	+	===== Users =====
	+
	+	If you use and have found this document useful then please let us know.
	+
	+	* [[http://www.translate.org.za/\|Translate.org.za]]
	+	* [[http://www.khmeros.info/\|Khmer OS]]

guide:custom_web_searches

Mon, 24 Mar 2008 18:29:18 +0000

1206382932		current
Line 1:		Line 1:

	It may be useful to provide translators with a web-based archive of reference documents that pertain to their translation. This would help them do highly focused searches in online documentation etc.		It may be useful to provide translators with a web-based archive of reference documents that pertain to their translation. This would help them do highly focused searches in online documentation etc.
		+

	===== Create a custom Google search engine =====		===== Create a custom Google search engine =====
Line 12:		Line 13:
	* You can specify just one or several pages/sites to index, and you can specify them in interesting ways (eg only pages that contain the letters "about"). You can also exclude pages by a pattern.		* You can specify just one or several pages/sites to index, and you can specify them in interesting ways (eg only pages that contain the letters "about"). You can also exclude pages by a pattern.

-	Unfortunately, the URL of the search engine is a bit difficult to remember, but you can easily link to it from any page. Here is a test search engine that will search only pages of this wiki:	+	Unfortunately, the URL of the search engine is a bit difficult to remember, but you can easily link to it from any page.
		+
		+	Here is a test search engine that will search only pages of this wiki, as well as pages from the Wikipedia that contain the letters "transl" in their URL:

-	[[http://www.google.com/coop/cse?cx=003167205026530159007%3Asrfg5ujnxog\|Translate Wiki Search Engine]]	+	* [[http://www.google.com/coop/cse?cx=003167205026530159007%3Asrfg5ujnxog\|Translate Wiki Search Engine]] (try doing a search for "machine").

	===== Adding a custom search in Opera =====		===== Adding a custom search in Opera =====

guide:cvs

Sat, 05 Jan 2008 17:57:03 +0000

1122886157	current
Line 1:	Line 1:
	+	====== A basic guide to CVS ======

	+	CVS is a version control system. It keeps track of files, changes and
	+	who made the changes. Some projects are starting to use [[::svn]] for
	+	managing their files.
	+
	+	===== Resources =====
	+
	+	* [[http://www.cvshome.org/docs/blandy.html\|Crash course]]
	+	* [[http://www.cvshome.org/docs/manual/\|Full course]]

guide:debian

Sat, 05 Jan 2008 17:57:29 +0000

1121980867	current
Line 1:	Line 1:
	+	====== Debian Translation ======
	+
	+	Debian is a Linux distribution.
	+
	+	===== Resources =====
	+
	+	* http://cvs.alioth.debian.org/cgi-bin/cvsweb.cgi/debian-installer/doc/translations.txt?rev=1.24&content-type=text/x-cvsweb-markup&cvsroot=d-i
	+	* [[http://people.debian.org/~seppy/d-i/translation-status.html\|Main Installer progress page:]]
	+	* [[http://people.debian.org/~barbier/d-i/l10n/nl/\|Example progess for Dutch]]
	+	* [[http://people.debian.org/~seppy/d-i/2nd-stage/nl.txt\|2nd stage translations]]
	+	* [[http://graal.ens-lyon.fr/~mquinson/debian/l10n-survey/\|An old document outlining the Debian process]]
	+	* [[http://www.debian.org/intl/l10n/\|Debian localisation stats]]

guide:developing_tools

Wed, 26 Nov 2008 13:51:53 +0000

1227707467		current
Line 1:		Line 1:
-	~~REDIRECT>spell checker case studies~~	+	~~REDIRECT>guide:spell_checker_case_studies~~

guide:document_translation

Fri, 21 Nov 2008 13:43:36 +0000

1227196963		current
Line 22:		Line 22:
	* Fonts or other styling changes		* Fonts or other styling changes
	* Style of generated content, such as numbers		* Style of generated content, such as numbers
		+	* Generated sections, such as Table of contents, list of figures, index, variables

guide:editing

Sun, 21 Oct 2007 13:20:20 +0000

1192972560		current
Line 49:		Line 49:

	This guide is under a creative commons license. Please be aware of that before adding your content and be aware that copy and paste might violate someone elses copyright - rather write from scratch or paraphrase the documents you want to copy.		This guide is under a creative commons license. Please be aware of that before adding your content and be aware that copy and paste might violate someone elses copyright - rather write from scratch or paraphrase the documents you want to copy.
		+

	===== Credit =====		===== Credit =====

-	Everyone want credit. Please add your name to the credits page.	+	Everyone wants credit. Please add your name to the credits page.

	===== Using Namespaces =====		===== Using Namespaces =====

guide:evaluating_spellcheckers

Wed, 04 Feb 2009 09:23:42 +0000

1227863755		current
Line 1:		Line 1:
-	=== Metrics ===	+	====== Evaluating spell checkers ======
-	to get an idea of the quality of a spellchecker we can use the information retrieval metrics of precision and recall.	+

-	Precision: is a measuer of the exactness of the spellcheckers responses, it basically tells you how much you should trust the spellchecker when it tells you this word is correct	+	===== Metrics =====

-	Recall: is a measure of the completeness of the spellchecker, it tells you how much of the language the spellchecker covers, the lower the value the more likely it is that the spellchecker will complain about correct words	+	To get an idea of the quality of a spellchecker, we can use the information retrieval metrics of precision and recall.

-	it might also be useful to measure the same useful to measure the same metrics but for the case where the spellchecker identifies the word as incorrect instead of correct.	+	Precision: is a measure of the exactness of the spellchecker's responses, it basically tells you how much you should trust the spellchecker when it tells you this word is correct.

-	in that case Recall is called specificity and it tells you how likely it is that the spellchecker will catch all incorrect words.	+	Recall: is a measure of the completeness of the spellchecker, it tells you how much of the language the spellchecker covers, the lower the value the more likely it is that the spellchecker will complain about correct words.

-	generally speaking the spellchecker should have very high Precision and specificity to be useful.	+	It might also be useful to measure the same useful to measure the same metrics but for the case where the spellchecker identifies the word as incorrect instead of correct.

-	we also calculate Accuracy which is derived from both precision and recall and is a general measure of the quality of the spellchecker.	+	In that case recall is called specificity and it tells you how likely it is that the spellchecker will catch all incorrect words.

-	the [[spell_quality.py]] script in the wordlist directory of the toolkit can be used to measure precision, recall, accuracy and specificity	+	Generally speaking the spellchecker should have very high Precision and specificity to be useful.
-	=== actual performance ===	+

-	now these metrics can be a bit too generic, to judge whether a recall of 30% is good or bad you need to know which 30% of the language does the spellchecker cover. or more importantly which correct words is it going to assume are incorrect.	+	We also calculate accuracy which is derived from both precision and recall and is a general measure of the quality of the spellchecker.

-	it's ok if the spellchecker doesn't know about obscure, archiac, deeply technical or other rarely used words. but it's a very big problem if it can recognize millions of words but fails to recognize a few widely used ones.	+	The [[spell_quality.py]] script in the wordlist directory of the toolkit can be used to measure precision, recall, accuracy and specificity.

-	we can manually evaluate coverage by generating [[false negative frequency tables]]	+	===== Actual performance =====
		+
		+	Now these metrics can be a bit too generic, to judge whether a recall of 30% is good or bad you need to know which 30% of the language does the spellchecker cover. Or more importantly which correct words is it going to assume are incorrect.
		+
		+	It's ok if the spellchecker doesn't know about obscure, archaic, deeply technical or other rarely used words, but it is a very big problem if it can recognize millions of words but fails to recognize a few widely used ones.
		+
		+	We can manually evaluate coverage by generating [[false negative frequency tables]].

guide:evolution

Sat, 05 Jan 2008 17:59:19 +0000

1199555939		current

guide:existing_glossaries

Wed, 26 Aug 2009 14:49:41 +0000

1199555983		current
Line 1:		Line 1:
	====== Existing Glossary Resources ======		====== Existing Glossary Resources ======

-	The following are existing glossary resources. Please first check that you are	+	The following are existing glossary resources. Please first check that you are indeed allowed to use them.
-	indeed allowed to use them.	+

		+	For Free and Open Source Software, the single most useful resource is probably at http://open-tran.eu/ which allows you to view translations from many of the major FOSS applications. Support for Open-Tran.eu is integrated into some translation tools, such as [[:virtaal/index\|Virtaal]].

		+	The FUEL project tries to coordinate terminology development amongst Indic languages.


	===== Glossaries =====		===== Glossaries =====

-	* [[http://search.novell.com/NSearch/SearchServlet?encoding=iso-8859-1&query0=glossaries&operator0=0&numhits=10&template=ResultListTerseTemplate.html\|Novell]]	+	* [[http://www.novell.com/company/glossary.html\|Novell]]
-	* [[http://www.microsoft.com/globaldev/tools/MILSGlossary.mspx\|Microsoft]] - you'll need to jump through the normal Microsoft hoops	+	* [[http://www.microsoft.com/language/\|Microsoft]]
	* [[ftp://ftp.apple.com/developer/Tool_Chest/Localization_Tools/Apple_Intl_Glossaries/Apple_Intl_Glossaries_TXT/\|Apple]] ([[ftp://ftp.apple.com/developer/Tool_Chest/Localization_Tools/Mac_OS_X_Glossaries/\|2]])		* [[ftp://ftp.apple.com/developer/Tool_Chest/Localization_Tools/Apple_Intl_Glossaries/Apple_Intl_Glossaries_TXT/\|Apple]] ([[ftp://ftp.apple.com/developer/Tool_Chest/Localization_Tools/Mac_OS_X_Glossaries/\|2]])
-	* [[http://members.microsoft.com/wincg/home.aspx\|Microsoft Online Glossaries]]	+	* [[http://members.microsoft.com/wincg/home.aspx\|Microsoft "community" Glossaries]]
-	* [[http://www.mozilla.org/projects/l10n/mlp_glossary.html\|Mozilla]] - old 2000'ish, look like they are simply TMs	+
	* [[https://developer.skype.com/Download/SkypeStrings\|Skype]]		* [[https://developer.skype.com/Download/SkypeStrings\|Skype]]

	===== Domain Specific =====		===== Domain Specific =====

-	==== Statistical Dictionaries ====	+	...
-		+
-	Useful to ensure that your translations of statistical fundtions in	+
-	spreadsheets are correct.	+
-		+
-	* [[http://europa.eu.int/comm/eurostat/research/index.htm?http://www.europa.eu.int/en/comm/eurostat/research/isi/&1\|ISI glossary of statistical terms]] --- available in various languages.	+

guide:false_negative_frequency_tables

Fri, 28 Nov 2008 13:27:52 +0000

	current
Line 1:	Line 1:
	+	frequency tables of words deemed incorrect by a spellchecker can be generated with a simple command line
	+
	+	<code>
	+	hunspell -d zu_ZA -l corpus.txt \| sort \| uniq -dc \| sort -nr \| head
	+	</code>

guide:fedora

Sat, 05 Jan 2008 18:00:10 +0000

1199555085		current
Line 26:		Line 26:
	* redhat-menus		* redhat-menus

-	Also look at the [[guide:bootup]] process to see other packages that might be important.	+	Also look at the [[bootup]] process to see other packages that might be important.

	===== Release Schedule =====		===== Release Schedule =====
Line 38:		Line 38:

	As usual check on the mailing list to see if anyone is translating your language. If someone is then work out		As usual check on the mailing list to see if anyone is translating your language. If someone is then work out
-	a way to combine forces. If not you will probably become the translation coordinator for your language. Either way you might want [[Fedora#Getting_CVS_access\|CVS access]]. In order to translate you don't need CVS access. But you will as soon as you want the data in the applications.	+	a way to combine forces. If not you will probably become the translation coordinator for your language. Either way you might want [[:fedora#Getting_CVS_access\|CVS access]]. In order to translate you don't need CVS access. But you will as soon as you want the data in the applications.

	===== Getting CVS access =====		===== Getting CVS access =====

guide:firefox

Thu, 16 Apr 2009 09:27:31 +0000

1238080727		current
Line 17:		Line 17:
	===== Settings =====		===== Settings =====

-	These are configurations not customisations. These are things that you must change.	+	These are configurations not customisations. These are things that you must change. Remember that you need to go through a formal procedure of review to commit changes to some of these files in the Mozilla repository, especially a file like browser/chrome/browser-region/region.properties. This usually means you need to file a bug, ask for review by setting flags and more things and put people in CC. Ask at Mozilla for the newest rules.

	- file: ''toolkit/defines.inc.po''		- file: ''toolkit/defines.inc.po''

guide:foss_fonts

Fri, 06 Feb 2009 11:00:58 +0000

1233917582		current
Line 12:		Line 12:
	* [[http://fontforge.sourceforge.net/\|FontForge]] --- Free font editing and creation tool		* [[http://fontforge.sourceforge.net/\|FontForge]] --- Free font editing and creation tool
	* Indrek Hein's [[http://www.eki.ee/letter/\|Letter Database]] --- what special characters are needed by your/a language, their Unicode codes, etc		* Indrek Hein's [[http://www.eki.ee/letter/\|Letter Database]] --- what special characters are needed by your/a language, their Unicode codes, etc
-	* Fontconfig [[http://cvs.freedesktop.org/fontconfig/fontconfig/fc-lang/\|Orthographies]] --- these fc-lang files contain the list of characters needed by various languages and are used to better guess which font to use to render a document.	+	* Fontconfig [[http://cgit.freedesktop.org/fontconfig/tree/fc-lang\|Orthographies]] --- these fc-lang files contain the list of characters needed by various languages and are used to better guess which font to use to render a document.
	* [[http://www.travelphrases.info/fonts.html]] --- want to see the scripts of various languages and their fonts, plus links to both free and proprietary fonts.		* [[http://www.travelphrases.info/fonts.html]] --- want to see the scripts of various languages and their fonts, plus links to both free and proprietary fonts.
	* Troubleshooting		* Troubleshooting

guide:foss_l10n_calendar

Fri, 24 Oct 2008 10:58:59 +0000

1224845628		current

guide:general_articles

Sat, 05 Jan 2008 18:01:32 +0000

1197541937		current

guide:gettext_hacks

Sat, 05 Jan 2008 18:01:52 +0000

1122916004	current
Line 1:	Line 1:
	+	====== Gettext Hacks ======
	+
	+	A few useful Gettext hacks.
	+
	+	===== Branch merge =====
	+
	+	Needs to merge translations from a branch back onto HEAD?
	+
	+	<code bash>
	+	msgmerge -C HEAD.po BRANCH.po HEAD.pot > NEW-HEAD.po
	+	mv NEW-HEAD.po HEAD.po
	+	</code>
	+
	+	Use //HEAD.po// as a compendium to make sure you get any translations that might
	+	have been done in HEAD. Then migrates //BRANCH.po// to be up-to-date with
	+	//HEAD.pot//. The output //NEW-HEAD.po// should then be moved to //HEAD.po// if the merge
	+	happened as planned.

guide:glossary

Sun, 21 Oct 2007 12:18:13 +0000

1145125579		current
Line 32:		Line 32:
	* PO --- Portable Object file. These are the files produced by xgettext (cf.) that a translator translates into their language. These files are compiled using Gettext (cf.) tools into an MO file (cf.)		* PO --- Portable Object file. These are the files produced by xgettext (cf.) that a translator translates into their language. These files are compiled using Gettext (cf.) tools into an MO file (cf.)
	* PO Editor --- Any one of a number of tools that can edit PO files. PO files are plain text files and in most cases a PO editor will hide the complexity from the translator and provide extra features like translation memory, etc.		* PO Editor --- Any one of a number of tools that can edit PO files. PO files are plain text files and in most cases a PO editor will hide the complexity from the translator and provide extra features like translation memory, etc.
		+	* TEnT --- Translation Environment Tool. A term suggested by some (Jost Zetzsche) as an alternative to CAT (cf.).
	* TM --- see Translation Memory		* TM --- see Translation Memory
	* TMX --- see Translation Memory Exchange format		* TMX --- see Translation Memory Exchange format

guide:gnome_desktop

Thu, 10 Jan 2008 10:20:38 +0000

1199556136		current
Line 1:		Line 1:
	====== GNOME ======		====== GNOME ======

-	GNOME is a feature rich Linux graphical desktop environment. The project has quite an	+	GNOME is a feature rich Linux graphical desktop environment. The project has quite an established localisation initiative.
-	established localisation initiative.	+
		+	Since this page has been created, many things in GNOME has changed, and they have good documentation and infrastructure to help translators.
		+
		+	The most important documentation for translating GNOME is therefore at the GNOME website:
		+	http://live.gnome.org/TranslationProject/
		+
		+	The rest of this document is kept in place for the time being, but a lot of it is now out of date and not 100% accurate any more.

	===== Essential Reading =====		===== Essential Reading =====

guide:golden_rules

Sat, 05 Jan 2008 18:22:29 +0000

1199557349		current
Line 90:		Line 90:
	Here are some ideas of projects to look into:		Here are some ideas of projects to look into:
	- [[http://tuxpaint.org\|Tuxpaint]] - A drawing program for children. This is a very small translation (less than 1000 words) and the program is not dependent on any libraries or GUI toolkits that need to be translated.		- [[http://tuxpaint.org\|Tuxpaint]] - A drawing program for children. This is a very small translation (less than 1000 words) and the program is not dependent on any libraries or GUI toolkits that need to be translated.
-	- [[http://pootle.wordforge.org/\|Pootle]] - A online system for translation and translation management. This is not such a high priority program to translate, but it is easy to do it, and provides lots of help. Translating Pootle can therefore also help to train new translators in your team.	+	- [[http://pootle.locamotion.org/\|Pootle]] - A online system for translation and translation management. This is not such a high priority program to translate, but it is easy to do it, and provides lots of help. Translating Pootle can therefore also help to train new translators in your team.
	- [[http://www.gcompris.net/\|GCompris]] - Educational software for children. This is quite a large translation (more than 10 thousand words), but it is fairly easy to prioritise the effort either by targeting specific activities, or by doing the [[short_strings_first]]. GCompris is not dependent on any libraries or GUI toolkits that need to be translated.		- [[http://www.gcompris.net/\|GCompris]] - Educational software for children. This is quite a large translation (more than 10 thousand words), but it is fairly easy to prioritise the effort either by targeting specific activities, or by doing the [[short_strings_first]]. GCompris is not dependent on any libraries or GUI toolkits that need to be translated.
	- [[http://l10n.gnome.org/module/xdg-user-dirs\|xdg-user-dirs]] - Translations for the common directories on Linux type systems (Documents, Desktop, Videos, Music, etc.) It is a very small translation (less than 50 words) and the effects should be visible in most "Open" and "Save" dialog boxes on Linux systems. Also see [[http://freedesktop.org/wiki/Software/xdg-user-dirs\|the project page]].		- [[http://l10n.gnome.org/module/xdg-user-dirs\|xdg-user-dirs]] - Translations for the common directories on Linux type systems (Documents, Desktop, Videos, Music, etc.) It is a very small translation (less than 50 words) and the effects should be visible in most "Open" and "Save" dialog boxes on Linux systems. Also see [[http://freedesktop.org/wiki/Software/xdg-user-dirs\|the project page]].

guide:google

Thu, 21 Jul 2005 20:41:18 +0000

	current
Line 1:	Line 1:
	+	====== Translating Google ======

	+	NOTE: It seems that as at 13 Oct 2004 Google says that they "cannot support more
	+	languages at this time". So I'm afraid your language will probably not be supported.
	+
	+	While not Open Source or Free Software the Google search engine is a powerful
	+	tool used by many on the Internet. Google can be localised for you language
	+	using their own custom interface.
	+
	+
	+	===== How to translate =====
	+
	+	The process is simple, [[https://services.google.com/tc/Welcome.html\|register]] and then follow
	+	the on screen instructions. One advantage of translating Google is that you can quicly see what your
	+	translation will look like in real life. That is an advantage of a web based applications.

guide:grammar

Sat, 05 Jan 2008 18:03:06 +0000

1132735828	current
Line 1:	Line 1:
	+	====== Grammar Checkers ======
	+
	+	Kvein Scannel is developing a grammar checker called //An Gramadóir//.
	+
	+	===== Resources =====
	+
	+	* [[http://borel.slu.edu/gramadoir/\|Project page]]
	+	* [[http://borel.slu.edu/gramadoir/manual/index.html\|Developer's manual]]

guide:gweather

Sat, 05 Jan 2008 18:03:28 +0000

1122750658	current
Line 1:	Line 1:
	+	====== Localising GWeather ======

	+	This email explains what needs to be done to localise gweather so that it has
	+	the right names for locations in your locale.
	+
	+	<code email>
	+	From: Davyd Madeley
	+	To: gnome-love@gnome.org, gnome-i18n@gnome.org
	+	Date: Tue, 30 Nov 2004 03:10:04 +0800
	+	Subject: GNOME Lovers Needed: l10n work for locations database
	+
	+	gweather ships a very comprehensive locations database. GNOME-Applets
	+	2.9.2 has some 3000+ more locations to this database[1] from syncing
	+	with the latest METAR list, but now we need your help.
	+
	+	The database now has support for tags (as well as ,
	+	and ). While region, country, and state are all easy to
	+	work out, city tags are hard, it's easy to figure out what locations are
	+	in a city (sometimes) but not what their local names are. As a result,
	+	many names are things like "Location 1" or "My Location (alternate)".
	+
	+	This is where you come in, spread around the world, and having a good
	+	knowledge of local, state and maybe even national geography, you will be
	+	able to clean up the locations file for your area.
	+
	+	How do you do this?
	+	1) Check out the latest version of Locations.xml.in [2]
	+	2) Fix up your area, the format for the XML file should be fairly
	+	evident, or you can consult the DTD [3].
	+	3) Use xmllint and the DTD to check that the file structure is still
	+	sane.
	+	4) File a bug in bugzilla under the gweather component of
	+	gnome-applets. Give it the subject, "Locations Love - $my_area". Attach
	+	a context diff of the Locations.xml.in file.
	+
	+	If you're a new contributor to GNOME, this is a really good way to get
	+	yourself in the Changelog and NEWS file. This type of bread and butter
	+	work is really easy, and you don't even have to build any software to do
	+	it ;)
	+
	+	Remember that all strings in the Locations.xml.in file should be in the
	+	C locale (US english). This means that locations should use their
	+	American names or anglicized names. Otherwise use the local name. If you
	+	use an anglicized name, add a comment to the XML so that translators
	+	know about the local name.
	+
	+	I would really like this done as soon as possible so that translators
	+	can get to work on getting the translations for the database up to
	+	scratch. I want to have a really rocking locations database ready for
	+	GNOME 2.10 (and eventually weather.gnome.org).
	+
	+	Also, maintaining the locations database has been proving to be an
	+	increasingly time consuming task for me. If you feel you might be up to
	+	scratch at maintaining this database, drop me a line, I could do with
	+	some help.
	+
	+	Happy Hacking,
	+	--davyd
	+
	+	[1] a log showing new locations/changed locations and removals is here:
	+	http://oracle.bridgewayconsulting.com.au/~davyd/misc/locations_changes.log
	+	[2] http://cvs.gnome.org/viewcvs/checkout/gnome-applets/gweather/Location=
	+	s.xml.in
	+	[3] http://cvs.gnome.org/viewcvs/checkout/gnome-applets/gweather/location=
	+	s.dtd
	+
	+	--
	+	Davyd Madeley http://www.davyd.id.au/
	+	</code>

guide:hyphenation

Wed, 18 Mar 2009 08:47:17 +0000

1199556259		current
Line 12:		Line 12:
	* This [[ftp://tug.ctan.org/pub/tex-archive/info/patgen2.tutorial\|patgen2 tutorial]] might be helpfull		* This [[ftp://tug.ctan.org/pub/tex-archive/info/patgen2.tutorial\|patgen2 tutorial]] might be helpfull
	* This is a well commented (in TeX) version of [[http://www.tug.org/tex-archive/systems/knuth/unsupported/texware/patgen.web\|patgen]]		* This is a well commented (in TeX) version of [[http://www.tug.org/tex-archive/systems/knuth/unsupported/texware/patgen.web\|patgen]]
-	* [[http://www.fi.muni.cz/%7Exantos/patlib/thesis/userguide-p.ps\|A very good explanation with clear examples]] for OPatGen (a Unicode enables patgen)	+	* [[http://www.fi.muni.cz/%7Exantos/patlib/thesis/userguide-p.ps\|A very good explanation with clear examples]] for OPatGen (a Unicode enabled patgen)
	* [[http://www.ntg.nl/spelling/hyphenation.html\|A very good set of tools]] written by the Dutch TeX team. Also useful for manipulation of word lists.		* [[http://www.ntg.nl/spelling/hyphenation.html\|A very good set of tools]] written by the Dutch TeX team. Also useful for manipulation of word lists.
		+	* [[http://hunspell.sourceforge.net/tb87nemeth.pdf\|Automatic non-standard hyphenation in OpenOffice.org]] written by László Németh about the improved hyphenation technology in OpenOffice 2.0.2 and later.

	===== Creating you own hyphenation dictionary =====		===== Creating you own hyphenation dictionary =====

guide:initranslator

Sat, 05 Jan 2008 18:04:40 +0000

1122744879	current
Line 1:	Line 1:
	+	====== Ini Translator ======
	+
	+	Ini Translator is a utility program to translate ini-style language files and
	+	has a look and feel reminiscent of [[::poedit]]. There are several plugins supplied
	+	with the install including parsers for Gettext PO format and others.
	+
	+	===== Resources =====
	+
	+	* [[http://initranslator.sourceforge.net/\|Home page]]

guide:introduction_to_wordfast

Mon, 06 Aug 2007 07:20:19 +0000

1186384505		current
Line 137:		Line 137:

	MS Word will give text files the file extension "TXT". If a different file extension is required, the files should be renamed in MS DOS or using some other renaming tool.		MS Word will give text files the file extension "TXT". If a different file extension is required, the files should be renamed in MS DOS or using some other renaming tool.
		+
		+	===== Marking text with styles =====
		+
		+	Many Wordfast users have never worked with documents that have been marked with untranslatable text. This shouldn't be a problem as long as they are told not to attempt to translate the grey text, and that they should use "Next Placeable", "Previous Placeable" and "Place Placeable" to copy the red text.
		+
		+	Placeables are pieces of text that Wordfast can grab in the source box and copy to the position of the cursor in the target box. Placeables can be placed using icons on the Wordfast toolbar, or using keyboard shortcuts (see the Wordfast manual for a comprehensive list of shortcuts).
		+
		+	Even if a Wordfast user has worked with such marked documents before (they are referred to as tagged texts), he may not know how to mark such a document himself. Ideally, therefore, the marking up of a document will be done by a project manager or senior translator.

guide:kbabel

Sat, 05 Jan 2008 18:05:00 +0000

1122662374	current
Line 1:	Line 1:
	+	====== KBabel ======

	+	KBabel is a KDE application for editing PO files. It is feature rich once you
	+	have got used to how it works. It will also shortly run on Windows.
	+
	+	===== Installing =====
	+
	+	* Linux --- Forms part of the kdesdk package. Install that and any dependencies.
	+	* Windows --- Porting still in progress ([[http://iidea.pl/~js/qkw/2004-11-25-kbabel_win32.png\|screenshot]])
	+
	+	===== Using KBabel =====
	+
	+	==== First Time ====
	+
	+	FIXME
	+
	+	==== The Editor ====
	+
	+	FIXME
	+
	+	==== Catalogue Manager ====
	+
	+	FIXME

guide:kde_desktop

Sat, 05 Jan 2008 18:05:25 +0000

1121971033	current
Line 1:	Line 1:
	+	====== KDE ======
	+
	+	The KDE Linux desktop is easily localisable and has a very mature localisation
	+	project.
	+
	+	===== Resources =====
	+
	+	* [[http://i18n.kde.org\|Localisation Project website]]
	+
	+	===== Suggested Priorities =====
	+
	+	These are the [[http://i18n.skulelinux.no/prioritering.html\|priorities]] as suggested by the SkuleLinux project.
	+
	+	There is no priority within each of these groups
	+
	+	FIXME this is probably out of date since the move to SVN version control
	+
	+	^ First ^ Second ^ Third ^ Fourth ^
	+	\| kdebase \| kdeaccessibility \| kdeaddons \| kdeextragear \|
	+	\| kdelibs \| kdeadmin \| kdeartwork \| kdeextragear-1 \|
	+	\| kdepim \| kdeedu \| kdegames \| kdeextragear-2 \|
	+	\| koffice \| kdegraphics \| kdetoys \| kdenonbeta \|
	+	\| \| kdekiosk \| docs \| others \|
	+	\| \| kdemultimedia \| \| \|
	+	\| \| kdenetwork \| \| \|
	+	\| \| kdesdk \| \| \|
	+	\| \| kdeutils \| \| \|
	+
	+	Of course you will probably want to move some applications found in the
	+	kdeextragear modules up. Or translate them when requested. Suggested
	+	candidates for bumping up include:
	+	amaroK, kbabel, superkaramba, Scribus, Kopete, Quanta+,
	+	Kaffeine, Juk, KDevelop, digiKam, kdebluetooth, kontact
	+	and kimdaba
	+
	+	===== Compiling your language =====
	+
	+	FIXME update for SVN
	+
	+	If you have made changes or wish to install the latest translations for your
	+	language from CVS do the following after checking out the kde-i18n module from
	+	CVS.
	+
	+	<code bash>
	+	$ cd kde-i18n
	+	$ echo XX > inst-apps # Where XX is your language's ISO code
	+	$ make -f Makefile.cvs
	+	$ ./configure # or ./configure --prefix=/usr if you do not want files installed is /usr/local
	+	$ make
	+	$ su -c "make install"
	+	</code>
	+
	+	===== Updating all translations manually =====
	+
	+	You can update all translations manually instead of waiting for KDE's automatic
	+	update bot called scripty.
	+
	+	<code bash>
	+	$ cd kde-i18n
	+	$ cvs up templates
	+	$ VERBOSE=yes make -f Makefile.am.in merge
	+	</code>
	+
	+	If you specify your language in a file called "inst-apps" then it will probably only update
	+	your language, which is a good thing. Have not checked this though.
	+
	+	===== Comitting to CVS =====
	+
	+	FIXME fix for SVN
	+
	+	Always check the PO files by running:
	+
	+	<code bash>
	+	$ cd kde-i18n
	+	$ check_po_files xx
	+	</code>
	+
	+	Where xx is your language code. Also running msgfmt in check mode
	+
	+	<code bash>
	+	$ msgfmt --check -o /dev/null some-file.po
	+	</code>
	+
	+	===== Checking accelerators =====
	+
	+	You now need to add the following to your kdeglobals in order to use the F12
	+	accelerator check feature:
	+
	+	[Development]
	+	CheckAccelerators=F12

guide:kde_locale

Sat, 05 Jan 2008 18:05:47 +0000

1199556327		current
Line 1:		Line 1:
	====== KDE Locales ======		====== KDE Locales ======

-	[[guide:KDE desktop\|KDE]], because it runs on platforms other than Linux, has created its own locale information environment. The data stored is similar to other locales, including date and time formats, etc.	+	[[kde_desktop\|KDE]], because it runs on platforms other than Linux, has created its own locale information environment. The data stored is similar to other locales, including date and time formats, etc.

	===== Resource =====		===== Resource =====

guide:keyboards

Sat, 05 Jan 2008 18:06:08 +0000

1160517717	current
Line 1:	Line 1:
	+	====== Keyboards ======

	+	For some languages you will need to create a keyboard. For many Latin languages this is not that difficult
	+	for others it can get quite elaborate and time consuming. Its easy if you are simply capturing the
	+	keyboard operation from an existing platform eg migrating a Microsoft based keyboard to Linux. However,
	+	if you are creating a whole new keyboard then you have a number of issues such as acceptance of the keyboard
	+	and standardisation processes.
	+
	+	===== Resources =====
	+
	+	* SIL documents on keyboards and input methods
	+	* [[http://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&item_id=WSI_Guidelines_Sec_7\|Guidelines for Writing System Support: Technical Details: Data Entry and Editing]]
	+	* [[http://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&item_id=KeybrdDesign\|An introduction to keyboard design theory: What goes where?]]
	+	* [[http://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&item_id=AfricanKeyboard1\|SIL Keyboarding Chart for Africa]]
	+	* [[http://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&item_id=FLOSS\|Resources for Writing Systems Implementation using Copyleft and FLOSS (Free/Libre and Open Source Software)]]
	+	* [[http://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&item_id=inputtoollinks\|Some tools and resources for character input]]
	+	* A list of [[http://www.datacal.com/alternative-language-support.htm\|Alternate Input Methods]] for Windows
	+
	+	===== Creating keyboards on Microsoft =====
	+
	+	==== Microsoft Keyboard Layout Creator (MSKLC) ====
	+
	+	* [[http://www.microsoft.com/globaldev/tools/msklc.mspx\|MSKLC]] --- Microsoft Keyboard Layout Creator
	+
	+	This will only work on Windows 2000 and XP. It WILL NOT work on Windows 95, 98, ME or NT 4
	+
	+	=== Using MSKLC ===
	+
	+	Firstly load an existing keyboard layout if you simply want to extend and existing layout.
	+
	+	You will notice that you can define keys for for each of the shift states and for the other modified state (usualy AltGr). For dead keys you must define one of the states as being a dead key and then define each of the combinations. Quite easy but tedious.
	+
	+	Once you are happy with your keyboard then select 'Project - Test Keyboard Layout...' you will see a text entry field and can type and test the keyboard. If all works well then select 'Project - Validate Layout', this will give you a report about any problems that were detected in your keyboard.
	+
	+	Want a picture of your new keyboard? 'File - Save As Image' - they're ugly but they get the message across.
	+
	+	The last step is to build your keyboard for that select 'Project - Build DLL and Setup Package'. This will create and MSI installer and a keyboard DLL, see the section below for steps on how to actually make the installer work.
	+
	+	=== Repackaging Installers with WiX ===
	+
	+	The MSKLC creates .msi files that don't contain the dll (they put it in a separate file).
	+
	+	In the keyboards/za module in CVS there is an example of how to rebuild these files correctly (for the South African keyboards). You need to run Make xx-xx.msi (and have the WiX binaries installed, see below) to rebuild the msi file for a language. You will probably need cygwin to run Make (it requires sed as well).
	+
	+	Manual way to fix this:
	+	- download WiX binaries from http://sourceforge.net/projects/wix/ and unpack them somewhere.
	+	- Disassemble the .msi file: ''/path/to/wix/dark.exe -x . path/to/us-za.msi us-za.wxs''
	+	- Make the changes to the .wxs file listed below
	+	- Make the wix object file: ''/path/to/wix/candle.exe us-za.wxs''
	+	- Build the new installer: ''/path/to/wix/light.exe us-za.wixobj''
	+
	+	You should see that the built installer has increased in size.
	+
	+	== Changes need to the MSKLC to get it to make a standalone installer ==
	+
	+	Edit the .wxs file before the candle and light commands:
	+	- Add Compressed="yes" as an attribute to the Package element at the start
	+	- Add EmbedCab="yes" Cabinet="Keyboard.cab" to the Media element a bit further down
	+	- Move the Media entry to between the Package and Property sections.
	+
	+	=== Using MSKLC automatically ===
	+
	+	It's nasty using a GUI tool :-)
	+
	+	http://lists.sharif.edu/pipermail/persiancomputing/2004-June/001425.html seems to be an approach to making it command-line...
	+
	+	==== Janko's Keyboard Generator ====
	+
	+	* [[http://solair.eunet.yu/~janko/engdload.htm\|Janko's Home Page]]
	+
	+	This will only work for Windows 95, 98 and ME.
	+
	+	Please note that if your languages characters are not in any Microsoft codepages then you will not be able to
	+	create a keyboard with those characters. Michael Kaplan (the Microsoft localisation guru) explains why not in this
	+	[[http://blogs.msdn.com/michkap/archive/2005/04/28/412977.aspx\|blog entry (look at the second one)]]. Essentially Windows
	+	95 is an old product which MS has stopped supporting and they'd rather you move to something new, they're both
	+	technical and marketing considerations.
	+
	+	===== Creating keyboard layouts for Linux =====
	+
	+	==== X.org and XFree86 ====
	+
	+	FIXME various pointers
	+
	+	==== Getting a nice picture of your layout ====
	+
	+	This prooves rather harder than expected. These pages will give you some pointers that are hopefully helpfull.
	+
	+	* [[http://www.meduna.org/sw_a2ps_en.html\|Ogonkify patch for processing xkbprint output]]
	+	* Look at [[http://www.xfree86.org/current/xkbprint.1.html\|xkbprint]] man page
	+
	+	==== Keyboard Mapping for Linux (KMFL) ====
	+
	+	* [[http://sourceforge.net/projects/kmfl\|Sourceforge project page]]
	+
	+	This project is trying to bring the kyman keyboard mapings to Linux it is/will be GPL'd
	+
	+	==== Changes needed to your Linux distribution ====
	+
	+	The various Linux distros have different configuration applications for setting things like keyboard. These few notes give a quick guide as to what files you need to change and where you need to look.
	+
	+	=== Fedora/Red Hat ===
	+
	+	The application ''system-config-keyboard'' sets the keyboard for your setup. It is simply a configuration tool for the normal X keyboard mappings. The file ''keyboard_models.py'' which is part of rhpl needs to be edited to add your keyboard.
	+
	+	===== Creating keyboard layouts for Mac =====
	+
	+	SIL has created a nice tools called [[http://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&item_id=ukelele\|Ukelele]] designed to create the XML files needed by the Macs keyboard system.

guide:kstars

Sat, 05 Jan 2008 18:06:31 +0000

1199556327		current
Line 1:		Line 1:
	====== Localisation of KStars ======		====== Localisation of KStars ======

-	[[http://edu.kde.org/kstars/\|KStars]] is a Desktop Planetarium for [[guide:KDE Desktop\|KDE]]. It provides an accurate graphical simulation of the night sky, from any location on Earth, at any date and time. The display includes 130,000 stars, 13,000 deep-sky objects,all 8 planets, the Sun and Moon, and thousands of comets and asteroids.	+	[[http://edu.kde.org/kstars/\|KStars]] is a Desktop Planetarium for [[kde_desktop\|KDE]]. It provides an accurate graphical simulation of the night sky, from any location on Earth, at any date and time. The display includes 130,000 stars, 13,000 deep-sky objects,all 8 planets, the Sun and Moon, and thousands of comets and asteroids.

	You can localise KStars by adding cities for your country to the file [[http://websvn.kde.org/trunk/KDE/kdeedu/kstars/kstars/data/Cities.dat?view=markup\|Cities.dat]].		You can localise KStars by adding cities for your country to the file [[http://websvn.kde.org/trunk/KDE/kdeedu/kstars/kstars/data/Cities.dat?view=markup\|Cities.dat]].

guide:locale_resource

Wed, 14 Oct 2009 15:04:45 +0000

1213186182		current
Line 4:		Line 4:
	* [[http://www.inference.phy.cam.ac.uk/dasher/download/alphabets/ALPHABETS.html\|Alphabets]] (what characters appear in a languages alphabet)		* [[http://www.inference.phy.cam.ac.uk/dasher/download/alphabets/ALPHABETS.html\|Alphabets]] (what characters appear in a languages alphabet)

-	The following are useful general resources that can be used during the construction of	+	The following are useful general resources that can be used during the construction of your locale files:
-	your locale files:	+

	* [[http://www.geonames.de/months.html\|Month names in various languages]]		* [[http://www.geonames.de/months.html\|Month names in various languages]]
Line 18:		Line 17:
	* [[http://www.loc.gov/standards/iso639-2/englangn.html\|ISO 639 language codes]] ([[http://www.loc.gov/standards/iso639-2/php/iso639-2form.php\|Request a new code]])		* [[http://www.loc.gov/standards/iso639-2/englangn.html\|ISO 639 language codes]] ([[http://www.loc.gov/standards/iso639-2/php/iso639-2form.php\|Request a new code]])

-	Collation:	+	===== Collation =====

	* [[http://developer.mimer.com/collations/charts/index.tml\|Language collation or character sort orders]]		* [[http://developer.mimer.com/collations/charts/index.tml\|Language collation or character sort orders]]
	* [[http://www.geonames.de/alphtz.html\|Some other alphabets]]		* [[http://www.geonames.de/alphtz.html\|Some other alphabets]]

-	Language Specific	+	===== Language Specific =====

	* Venda:		* Venda:
Line 29:		Line 28:
	* Font that can be used for Venda chars		* Font that can be used for Venda chars
	* http://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&item_id=Gentium_linux		* http://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&item_id=Gentium_linux
		+
		+	===== Operating System Specific =====
		+
		+	==== Microsoft Windows ====
		+	* [[win_lang_ids\|Language IDs]]
		+	* [[http://msdn.microsoft.com/en-us/goglobal/bb964664.aspx\|Locale IDs]] (LCIDs)

guide:locales_x11

Mon, 25 May 2009 17:36:12 +0000

1243272807		current
Line 71:		Line 71:
	===== New Locales: known problems =====		===== New Locales: known problems =====
	====== gdm ======		====== gdm ======
-	Your locale works in any terminal you open from the desktop, but won't be used in the desktop itself. This happens because of a copy of locale.alias that gdm holds in /etc/X11/gdm	+	Your locale works in any terminal you open from the desktop, but won't be used in the desktop itself. This happens because of a copy of locale.alias that gdm holds in /etc/X11/gdm (at least in gentoo).

	Add you locale to this file or it will be ignored by gdm.		Add you locale to this file or it will be ignored by gdm.

guide:locfactoryeditor

Sat, 05 Jan 2008 18:07:13 +0000

1135663874	current
Line 1:	Line 1:
	+	====== LocFactoryEditor ======

	+	===== the XLIFF and PO editor for Mac OSX =====
	+
	+	[[http://www.triplespin.com/en/products/locfactoryeditor.html\|LocFactoryEditor]], or LFE for short ;) is a reasonable recent entry to the translation arena. Originally intended for the professional translation market, it has been extended to handle PO files, and is in fact available free as a dedicated PO editor.
	+
	+	Mike of [[http://www.triplespin.com/\|Triplespin]] has worked closely with at least one open-source translator (me ;) ) to provide a really solid and indispensable translation tool, not only for volunteer translators, but for any translator using Mac OSX.
	+
	+	His aim is to create a full toolset for Mac OSX translators, firmly based on the current standards.
	+
	+	I've spent a lot of time testing and using his software, and I recommend it strongly. I'm impressed with a lot of things: the standards support, the built-in XLIFF/PO conversion, the TMX/Compendium translation memory, etc., but the one central thing that grabs me when I work in LFE is the interface.
	+
	+	It's intuitive. Nothing crowds the eye, and everything you need is right there, plus keyboard shortcuts if you prefer (and, of course, in OSX you can assign your own as well ;) ).
	+
	+	Smart folders mean simply one click to show only the untranslated or fuzzy strings, and counters at the top of the page alert you if spaces or symbols aren't consistent between original and translated string.
	+
	+	The Notes facility makes sense of the stodge of PO headers, and allows you to set your own Notes, input them with shortcuts, and edit them directly. Everything converted neatly back into PO format in the output file.
	+
	+	We have plurals support, of course, can show or hide obsolete strings, and LFE maintains a backup file automatically (to protect you from yourself ;) ).
	+
	+	If you also translate Mac applications via AppleGlot, LFE handles the entire process, from first pass to final test, work glossaries and .loc files.
	+
	+	Try it. The PO-only version won't cost you anything, and it'll save you a lot of time and hassle. I bought the full version, which is very affordable, and have had my money back in sheer features and convenience many times over.
	+
	+	As a Mac OSX translator, I had several options:
	+
	+	* emacs — very cluttered interface and (to me) non-intuitive operation: drove me around the bend
	+	* kbabel — finally installed it via fink after months of struggling with error messages and forums/mailing lists, and X11 for Mac OSX won't input Vietnamese ... !!
	+	* gtranslator — wasn't able to install it, despite many tries
	+
	+	I also use Pootle online, but in general and most heart-feltedly, I give thanks daily for LFE!
	+
	+	Clytie Siddall
	+
	+	Vietnamese translator for open-source projects

guide:mediawiki

Mon, 26 May 2008 01:06:31 +0000

1199556457		current
Line 6:		Line 6:

	* [[http://meta.wikimedia.org/wiki/Localization_statistics\|Localization Stats]]		* [[http://meta.wikimedia.org/wiki/Localization_statistics\|Localization Stats]]
-	* [[http://meta.wikimedia.org/wiki/MediaWiki_localisation\|MediaWiki localisation]]	+	* [[http://www.mediawiki.org/wiki/Localisation\|MediaWiki localisation]]

guide:message_marking

Sat, 05 Jan 2008 18:07:57 +0000

1121983843	current
Line 1:	Line 1:
	+	====== Profiling through message marking ======
	+
	+	If you can quickly determine where a string in the user interface originates
	+	from then you can determine what strings and packages are more important to
	+	translate.
	+
	+	To achieve this you need to be able to visually identify the strings origin.
	+	This is achieved by using [[toolkit:podebug]] from the [[toolkit:index\|Translate Toolkit]].
	+
	+	===== Advantages =====
	+
	+	It is simple to use and can also be used for correcting errors in the translations.
	+	By simply running an application or computer environment in the same way as
	+	your target audience you can quickly determine what parts of the system are
	+	being called and therefore what aspects you will want to translate.
	+
	+
	+	===== Using the podebug method =====
	+
	+	[[toolkit:podebug]] is simple to use and will take a directory of translated PO files and
	+	create a directory of tagged PO files:
	+
	+	podebug -f "[%cb] " -i af -o af-debug
	+
	+	Once the debug PO files are created then the normal build process is followed
	+	for that project.
	+
	+
	+	===== Existing profiling builds =====
	+
	+	The following are builds that have already been profiled. Simply download,
	+	install and run the program as normal, making notes of the source files
	+	identified in the various screens.
	+
	+	* [[http://www.kilinux.org/kiblog/downloads/OOo_1.1.3_LinuxIntel_install_src_debug.tar.gz\|OpenOffice.org]]
	+

guide:monolingual

Wed, 20 May 2009 16:02:15 +0000

1239792726		current
Line 9:		Line 9:
	* [[toolkit:subtitles\|Video subtitles]]		* [[toolkit:subtitles\|Video subtitles]]
	* [[toolkit:rc\|Windows .rc files]]		* [[toolkit:rc\|Windows .rc files]]
-	* Document translation (such as for [[toolkit:ODF]])	+	* Document translation (such as for [[toolkit:ODF]] or DocBook)
	* [[toolkit:symb2po\|Symbian translation files]]		* [[toolkit:symb2po\|Symbian translation files]]
	* [[toolkit:PHP]] files used for translations		* [[toolkit:PHP]] files used for translations

guide:mozilla

Thu, 16 Apr 2009 10:08:02 +0000

1239876380		current
Line 51:		Line 51:
	* [[https://wiki.mozilla.org/L10n:Home_Page\|Mozilla L10n Home]]		* [[https://wiki.mozilla.org/L10n:Home_Page\|Mozilla L10n Home]]
	* [[http://www.mozilla.org/foundation/trademarks/l10n-policy.html\|Localisation policy/trademarks, etc]] ([[http://www.mozilla.org/foundation/licensing.html\|2]])		* [[http://www.mozilla.org/foundation/trademarks/l10n-policy.html\|Localisation policy/trademarks, etc]] ([[http://www.mozilla.org/foundation/licensing.html\|2]])
		+	* [[toolkit:mozilla_l10n_scripts\|Scripts used by Translate.org.za]]

	==== Deprecated information ====		==== Deprecated information ====

guide:mozillacvs

Mon, 21 Jan 2008 13:11:34 +0000

1199556532		current
Line 3:		Line 3:
	===== Introduction =====		===== Introduction =====

-	Starting with Firefox 1.0, and later with Firefox 1.5 and thunderbird 1.5, Mozilla Foundation has the policy to allow only localizations from registered teams. The localizations must be based on their CVS tree, that has an specific structure, and the localization must be given back in their same proper structure, and uploaded in the CVS. The installers and packs are automatically generated via their own tools.	+	Mozilla provides [[http://wiki.mozilla.org/L10n:Starting_a_localization\|instructions on starting a localisation]].

-	Mozilla Foundation always welcomes contributors, but expects (maybe "requires") that new localizations are based on the newest tree, or not based in old trees (at this time, the 1.0.x tree is considered old, and they don't accept new localizations for that tree). More info at [[http://wiki.mozilla.org/Draft:L10n:Creating_a_new_Localization]].	+	Starting with Firefox 1.0, and later with Firefox 1.5 and Thunderbird 1.5, Mozilla Foundation has the policy to allow only localizations from registered teams. The localizations must be based on their CVS tree, that has an specific structure, and the localization must be given back in their same proper structure, and uploaded in the CVS. The installers and packs are automatically generated via their own tools.

-	This document aims to have some bash scripts that, with potools installed, automate all the tasks, so you don't have to worry about the inner directories needed for the translation.	+	Mozilla Foundation always welcomes contributors, but expects (maybe "requires") that new localizations are based on the newest tree, or not based in old trees (at this time, the 1.0 and 1.5 trees are considered old, and they don't accept new localizations for those trees). Make sure you know what the best target is to base your work on . More info at the Mozilla Wiki: [[http://developer.mozilla.org/en/docs/Creating_en-X-dude\|1]]
		+	[[http://wiki.mozilla.org/Draft:L10n:Creating_a_new_Localization\|2]].
		+
		+	This document aims to have some bash scripts that, with the translate toolkit installed, automate all the tasks, so you don't have to worry about the inner directories needed for the translation.

	===== Getting the source =====		===== Getting the source =====

guide:ms_core_fonts

Thu, 21 Jul 2005 15:02:23 +0000

1121957891		current
Line 1:		Line 1:
-	Microsoft's Core Fonts	+	====== Microsoft's Core Fonts ======

-	An ufortunate legacy of the browser wars is that many sites are IE specific and	+	An unfortunate legacy of the browser wars is that many sites are designed specificly
-	more importantly are designed around a few Microsoft specific fonts. These	+	for Microsofts Internet Explorer and more importantly are designed around a few
-	fonts were released for free on the web, the Tahoma font does require a valid	+	Microsoft specific fonts. Fortunately Micrsoft released these fonts gratis on the web.
-	Windows License.	+	Note however, that the Tahoma font does require a valid Windows License.

		+	===== Installation instructions =====

-	Installation instructions	+	The [[http://corefonts.sourceforge.net/\|corefonts]] project has links to the required
		+	files and instructions for building and installing the fonts on Linux.

-	The site http://corefonts.sourceforge.net/ has links to the required files and	+	Windows users without these fonts can simply [[http://sourceforge.net/project/showfiles.php?group_id=34153&package_id=56408\|download]] the required files.
-	instructions for building and installing the fonts on Linux.	+
-		+
-		+
-	Download	+
-		+
-	Windows users without these fonts can simply download these files:	+
-	http://sourceforge.net/project/showfiles.php?group_id=34153&package_id=56408	+

guide:nonpo

Sat, 05 Jan 2008 18:09:46 +0000

1139234395	current
Line 1:	Line 1:
	+	==== Translating non-PO files using Pootle ====

	+	We all have our favourite format, and you can certainly translate it on Pootle! Pootle is your own personal translation environment: we want you to be able to set it up to suit your own workflow and needs.
	+
	+	We also want to simplify the file-format issue for you. Whether you're translating application files, documentation, manpages or working with glossaries and translation memory, you can convert the file to your favourite format.
	+
	+	This will be a feature of the Pootle interface: simply choose your format and convert the file! The file converters are those in the [[http://sourceforge.net/project/showfiles.php?group_id=91920&package_id=97082\|Translate Toolkit]], plus those in the [[http://po4a.alioth.debian.org/\|po4a]] package, and any others our users find useful.
	+
	+	So, until you can use the Pootle interface to convert your files, simply email [[translate-pootle@lists.sourceforge.net\|the Pootle mailing list]], attach your file and ask for it to be converted to the chosen format and loaded onto Pootle! Then you translate it, and request conversion back to the original format.
	+
	+	I tested this out with my first manpage translation, where the Pootle admins converted my (n)roff file to PO format, I translated it on Pootle, then they converted it back to (n)roff, and the process worked very well. It was quick and accurate. There were no conversion problems or artefacts, and the resulting file works well.
	+
	+	So, what formats can we handle? Below is our list, but note that not all formats can be converted in both directions yet. If you want to add other formats to it, please edit this page or write to our mailing list. If you know of another filter which would be useful, please tell us.
	+
	+	After all, Pootle is what we make of it. :)
	+
	+	\| File format \| filetype \| Translate Toolkit \| po4a \|
	+	\| Comma-separated values \| [[toolkit:csv\|csv]] \| √ \| \|
	+	\| Dia diagrams (uncompressed) \| \| \| √ \|
	+	\| Docbook \| xml \| \| √ \|
	+	\| Document Type Definition \| dtd \| √ \| \|
	+	\| Guide \| xml \| \| √ \|
	+	\| Hypertext Markup Language \| html \| √ \| \|
	+	\| Kernel configuration help files \| \| \| √ \|
	+	\| LaTeX files \| \| \| √ \|
	+	\| Manpages \| \| \| √ \|
	+	\| Mozilla translation project \| \| √ \| \|
	+	\| Nannoblogger files \| nb \| √ \| \|
	+	\| Open Office translation project \| \| √ \| \|
	+	\| POD data \| pod \| \| √ \|
	+	\| Portable Object Template \| pot \| √ \| \|
	+	\| Standard Generalized Markup Language \| sgml \| \| √ \|
	+	\| TeX documents and derivates \| \| \| √ \|
	+	\| Translation Memory eXchange \| [[toolkit:tmx\|tmx]] \| √ \| \|
	+	\| Qt Linguist \| ts \| √ \| \|
	+	\| OpenOffice.org Writer \| sxw \| √ \| \|
	+	\| Text file \| txt \| √ \| √ \|
	+	\| XML Localization Interchange File Format \| .xliff \| √ \| \|
	+	\| eXtensible Markup Language \| .xml \| \| √ \|
	+
	+
	+	Information on how to handle specific non-po files can be entered below.
	+
	+	=== OpenOffice.org ===
	+
	+	If you're starting a translation of OpenOffice in your language, you can take the .po files from [[http://sourceforge.net/projects/translate]]. If you have already started translation of OpenOffice with strings extracted from the source, you have to do the following:
	+
	+	- convert the OO format to PO format (see above)
	+	... more to come
	+
	+	(see [[toolkit:oo2po]] if you want to convert them using the translate toolkit)
	+
	+	=== FireFox ====

guide:omegat

Thu, 14 Feb 2008 10:09:14 +0000

1199556632		current
Line 12:		Line 12:
	* [[http://www.omegat.org/omegat/omegat.html\|Home Page]]		* [[http://www.omegat.org/omegat/omegat.html\|Home Page]]

		+

	===== Issues =====		===== Issues =====

-	* OmegaT can handle PO files, but only if they've been prepared using Gettext tools (possibly also toolkit tools).	+	* OmegaT's handling of PO files is non-standard. A partially translated PO file opened in OmegaT will lose all its translations. The PO file created by OmegaT will contain only translations that also occur in OmegaT's main TM.
-	* OmegaT is not intelligent, but simply replaces the text in a file with the translation.	+
	* OmegaT translates identical source segments with identical targets.		* OmegaT translates identical source segments with identical targets.

guide:open_language_tools

Thu, 23 Aug 2007 15:01:31 +0000

	current
Line 1:	Line 1:
	+	====== Open Language Tools ======
	+
	+	XLIFF editor
	+
	+	https://open-language-tools.dev.java.net/
	+
	+

guide:openoffice.org

Tue, 10 Nov 2009 09:54:47 +0000

1205748470		current
Line 13:		Line 13:
	* [[http://l10n.openoffice.org/languages.html\|Current and in progress localisations]]		* [[http://l10n.openoffice.org/languages.html\|Current and in progress localisations]]

-	===== What is your languages LCID =====	+	===== What is your language's LCID =====

-	Microsoft defines LCIDs for various locales. You need to know this so that OpenOffice.org can work well on Windows and also	+	Microsoft defines LCIDs for various locales. You need to know this so that OpenOffice.org can work well on Windows and also so that documents you create can move seamlessly between MS Word and Office Writer as the language identifier is correct.
-	so that documents you create can move seamlessly between MS Word and Office Writer as the language identifier is correct.	+

	There are a number of places that you can use to identify the LCID. For most languages they will all agree but in some cases (See 1072/Sutu/Sesotho) it helps to look at all list to help clarify what exactly Microsoft meant.		There are a number of places that you can use to identify the LCID. For most languages they will all agree but in some cases (See 1072/Sutu/Sesotho) it helps to look at all list to help clarify what exactly Microsoft meant.

guide:other_fonts

Sat, 05 Jan 2008 18:11:24 +0000

1199556064		current
Line 1:		Line 1:
	====== Other fonts ======		====== Other fonts ======

-	If you cant't get what you need from the [[guide:FOSS fonts]] then here is where you can look.	+	If you cant't get what you need from the [[foss_fonts]] then here is where you can look.

	===== License unkown =====		===== License unkown =====

guide:plurals_programmers

Sat, 05 Jan 2008 18:11:46 +0000

1122840558	current
Line 1:	Line 1:
	+	====== Plurals ======

	+	What is a plural? 1 file, 2 files, 3 files: file is singular, files plural.
	+	In this simple example we see the English plural form. Other languages
	+	have simpler or even more elaborate plural forms.
	+	Both KDE and Gettext have a method of managing plurals for PO files so that
	+	plurals can be used in languages that do not follow the English convention.
	+
	+	===== Why do you need this system? =====
	+
	+	Why not just write a piece of code like this:
	+
	+	<code c>
	+	if ( n == 1 ) then
	+	print "1 file"
	+	else
	+	print "%d files", n
	+	</code>
	+
	+	Well the simple answer is that not all languages follow the pluralisation
	+	format of English. Some use the singular form for both 0 and 1. Others have
	+	no singular/plural form. And yet others have complex systems that have at
	+	least 4 plural forms. For English speakers think of it as the different
	+	suffixes you see here: "1st, 2nd, 3rd, 4th ... 21st, 22nd, 23rd, 24th..."
	+
	+	So in order to cater for the various plural forms a system was developed.
	+
	+	===== When to use plurals in your application =====
	+
	+	If you have numbers in the singular and plural form then use the plural
	+	construct:
	+
	+	"%n file"
	+	"%n files"
	+
	+	However, if you do not have number in the construct then rather keep them as
	+	separate strings:
	+
	+	"Save the file"
	+	"Save the files"
	+
	+	===== What do plurals look like in the PO file? =====
	+
	+	On KDE with a 3 plural form language:
	+
	+	<code po>
	+	msgid ""
	+	"%n file\n"
	+	"%n files"
	+	msgstr ""
	+	"%n form 1\n"
	+	"%n form 2\n"
	+	"%n form 3"
	+	</code>
	+
	+	Gettext with a two form language:
	+
	+	<code po>
	+	msgid "%n file"
	+	msgid_plural "%n files"
	+	msgstr[0] "%n form 1"
	+	msgstr[1] "%n form 2"
	+	</code>
	+
	+	Also in Gettext style plurals the PO file header has an entry for the plural
	+	form which defines the number of plurals as well as the mathematical function to
	+	determine which one to use depending on the input number. Here is an example
	+	which would work for English:
	+
	+	<code po>
	+	Plural-Forms: nplurals=2; plural=n == 1 ? 0 : 1;
	+	</code>
	+
	+
	+	===== How to use the plural construct =====
	+
	+	==== KDE ====
	+
	+	FIXME this looks set to change for KDE4 with the Gettext method most probably being adopted
	+
	+	<code cpp>
	+	i18n( "One item", "%n items", count);
	+	</code>
	+
	+	Note that because KDE uses a \n to split the different plural forms you cannot
	+	use \n in the actual message. To work around this you can use KDE's RichText
	+	features and the
	+	tag.
	+
	+	==== Gettext ====
	+
	+	FIXME unsure how this is used

guide:po_with_wordfast

Sat, 05 Jan 2008 18:12:06 +0000

1138267634	current
Line 1:	Line 1:
	+	====== Translating PO files with WordFast ======
	+
	+	> I'd like to make some notes that can be reused for someone using
	+	> WordFast if they wanted to translate PO files.
	+
	+	Well, the important thing to remember is that Wordfast is not
	+	intelligent. The intelligence lies with the translator and/or the guy
	+	writing the tagger. Thankfully PO is quite simple, and if the lines are
	+	not wrapped, a translator can tag it almost manually.
	+
	+	What is tagging?
	+
	+	Well, in an MS Word document, you can apply a "style" to certain pieces
	+	of text. Default styles include Heading 1, Paragraph, etc. You can
	+	also set the behaviour of a style, so that all text with that style will
	+	look the same way. For example, I could set "Heading 1" to be Arial
	+	font, size 12, red text on a blue background, and all text with the
	+	style Heading 1 will look that way. It's kinda like HTML. You can also
	+	create your own "styles", and make them look like any thing you want to.
	+	The nice thing about styles is that you can tell a macro or a
	+	Find/Replace operation to perform its actions only on certain styles
	+	instead of the entire text. Very few people who use MS Word are aware
	+	of the great possibilities that exist thanks to "styles".
	+
	+	If I were to translate an HTML document in MS Word, I could do it as
	+	follows: First copy the text (the HTML code) to an empty MS Word
	+	document. Then create a unique style with a unique name. Then use
	+	Find/Replace to change the style of all the non-translatable code to
	+	that unique style. Then write a macro which extracts the translatable
	+	text by ignoring everything which is in that unique style. Or, write a
	+	macro which hides all text which is in that unique style (so that only
	+	the translatable text is visible). Or, well, you can see the
	+	possibilities here.
	+
	+	In Wordfast (and in Trados) there are two such unique styles, with which
	+	the macros interact in a special way. The one style is called
	+	tw4winExternal and the other is called tw4winInternal. In an HTML file
	+	(saved as an MS Word document), the External style is applied to
	+	block-level code and the Internal style is applied to line-level code.
	+	The Wordfast macro ignores everything which has the External style, and
	+	creates a temporary text box with each sentence of translatable text in
	+	which the translator types the translation. Text with the Internal
	+	style has a red colour, so that the translator can see which pieces of
	+	line-level HTML he should watch out for. In this way, the translator
	+	doesn't need to know HTML and yet he can translate directly within an
	+	HTML document (using MS Word) without any fear of "breaking" the code.
	+
	+	Marking the untranslatable code with External and Internal styles, is
	+	known as "tagging".
	+
	+	So, a translator faced with a tagged PO file in MS Word will not
	+	necessarily know how the PO format works. He will be able to see
	+	everything in the PO file, but he'll ignore it because Wordfast will
	+	automatically allow him access only to the text to the translated.
	+
	+	And now for the complication...
	+
	+	In Wordfast, the source text is overwritten with the target text.
	+	This means that when the translator gets the file from the client, the
	+	file must already be in such a way that the source text can be
	+	overwritten by the target text. This is why I insist on a "duplicated"
	+	PO file (however, I can duplicate it in MS Word too, so if the client is
	+	unable to provide a "duplicated" file, I can do it myself anyway).
	+
	+	Now, if you think about it carefully, you'll see that a translator
	+	cannot deal with PO fuzzys in Wordfast. Not possible.
	+
	+	However, most of the time when fuzzys occur, it is because an existing
	+	string (which had been translated before) has changed slightly, right?
	+	So it follows that the old string and its correct translation is still
	+	available somewhere. If this old string and its translation isn't in
	+	the translation memory (TM) already, it can easily be added to it, right?
	+
	+	Therefore, a simple workmethod is to delete all the proposed
	+	translations of fuzzys and turn the fuzzys into untranslateds (and then
	+	"duplicate" them), and translate them as usual.
	+
	+	Translating Opera
	+
	+	When I get PO files from my contact at Opera, the translated, fuzzy and
	+	untranslated strings are in a single file. First I used Gnu Gettext
	+	tools to separate these into three files. Then I turn the fuzzy strings
	+	into untranslated strings. Then I do the translation on the two
	+	"untranslated" files, and then I merge them back into the original file
	+	(the translated strings remain untouched by this process).
	+
	+	With Dwayne's PO files, I have to be more careful, and that is why I
	+	prefer that he does the gymnastics (then if something goes wrong and it
	+	is a conversion issue, then it's his fault). The gymnastics I'm
	+	referring to here, is separating the translated from the untranslated,
	+	and putting them into separate files. The "duplication" I can do
	+	flawlessless in MS Word (as long as the lines are unwrapped).
	+
	+	What is duplication (just in case you don't know)?
	+
	+	Here is an unduplicated PO string:
	+
	+	# This is a comment.
	+	# This is another comment.
	+	msgstr "This is the source text."
	+	msgid ""
	+
	+	Here is a duplicated PO string:
	+
	+	# This is a comment.
	+	# This is another comment.
	+	msgstr "This is the source text."
	+	msgid "This is the source text."
	+
	+	Here is a translated PO string:
	+
	+	# This is a comment.
	+	# This is another comment.
	+	msgstr "This is the source text."
	+	msgid "Hierdie is die bronteks."
	+
	+	These examples are oversimplified, of course.
	+
	+	Any questions?
	+
	+	The reason why some of the first PO files I did for Dwayne, didn't quite
	+	work out well, was because I didn't know how to do regex back then, so I
	+	didn't know how to do tagging, so instead I did a roundtripping which
	+	turned out to be only 99% lossless. Ouch (and again: sorry).
	+
	+	Samuel

guide:poedit

Sat, 05 Jan 2008 18:24:16 +0000

1186493099	current
Line 1:	Line 1:
	+	====== poEdit ======

	+	PoEdit is a PO editing tool that will run on Linux or Windows. It has a clean
	+	yet simple interface as well as quite a good translation memory. Its catalogue
	+	manager does not seem logical at first but is powerful in that it allows you to
	+	manage multiple projects.
	+
	+	===== Issues =====
	+
	+	* PoEdit knows a file's encoding from the PO header. Therefore, if a PO file has no header, PoEdit opens a file using the system's default encoding. This means that if you're using Windows, and you're opening UTF-8 PO files without headers, PoEdit will open them in Windows plaintext encoding, and so break the files.
	+
	+	===== Resource =====
	+
	+	* [[http://www.poedit.net/\|Home page]]
	+	* [[http://www.poedit.net/download.php\|Download]]
	+
	+	===== Using poEdit =====
	+
	+	==== Setup ====
	+
	+	When you first run poEdit you will be asked for some information. These items
	+	are stored in the header of the PO files and include your name, email address
	+	etc. The tabbed dialogue that appears needs the following information:
	+
	+	* Personalize: Supply your name and email address
	+	* Editor: you can safely leave these unchanged
	+	* Translation memory: leave unchanged for now
	+	* Parser: unchanged
	+
	+	Click OK and you are presented with the PO editing interface.
	+
	+	==== The first time with poEdit ====
	+
	+	Once POEdit is started you will see the translation interface.
	+
	+	\| {{poedit_windows.jpg}} \|
	+
	+	Click file open to open and existing PO file. If you have no PO file or only a
	+	POT file. Then simply copy the POT file so that it ends in .po
	+
	+	With a PO file open you will see all strings in the top part of the interface.
	+	Strings are shown in different colours:
	+
	+	\| Cyan \| untranslated \|
	+	\| Yellow \| fuzzy - needs to be examined and might need to be retranslated \|
	+	\| White \| translated \|
	+
	+	You translate in the lower half of the interface. The original string or
	+	English string appears in the left. You type your translations in the bottom
	+	right hand section. It seems that keyboard navigation does not work properly
	+	so you will have to select each new string with the mouse.
	+
	+	Don't forget to save regularly using Ctrl-S
	+
	+	Press F1 at anytime for the POEdit user manual. We recommend that you browse
	+	it now.
	+
	+	==== Using poEdit ====
	+
	+	FIXME
	+
	+	==== Using the Catalogue Manager ====
	+
	+	The catalogue manager allows you to manage projects with multiple PO files.
	+
	+	File -> Catalog manager
	+
	+	Click on create new translation project. Then for each directory that has PO
	+	files add it to the project.
	+
	+	==== Using Translation Memory ====
	+
	+	FIXME

guide:pofilter_examples

Sat, 05 Jan 2008 18:23:57 +0000

1123140471	current
Line 1:	Line 1:
	+	====== Common Translation Errors ======

	+	The following are common errors detected by the [[toolkit:pofilter]] tool. It is usefull
	+	to see these errors in order that you don't commit them yourself. We have
	+	given examples of some false positives also.
	+
	+
	+	===== Double Spacing =====
	+
	+	Notice the double space in "lefoko la".
	+
	+	<code po>
	+	#: NoColorError
	+	#, fuzzy
	+	#_ doublespacing: checks for bad double-spaces by comparing to original
	+	msgid "Click on a color or enter a valid HTML color string"
	+	msgstr ""
	+	"tobetsa mmala kgotsa tsenya lefoko la mmala la html le le letleletsweng"
	+	</code>
	+
	+	The following is a false postive, notice that the double space ("account. There")
	+	is in the msgid.
	+
	+	<code po>
	+	#: accountSettingsDesc.label
	+	#, fuzzy
	+	#_ doublespacing: checks for bad double-spaces by comparing to original
	+	msgid ""
	+	"The following is a special account. There are no identities associated with "
	+	"it."
	+	msgstr ""
	+	"Se se latelang ke akhaoto e kgetegileng. Ga go kitsiso e kamano le yona."
	+	</code>
	+
	+	===== End Punctuation =====
	+
	+	The sentence is missing a fullstop.
	+
	+	<code po>
	+	#: MissingSiteNameError
	+	#, fuzzy
	+	#_ endpunc: checks whether punctuation at the end of the strings match
	+	msgid "Please enter a name for this publishing site."
	+	msgstr "ka kopa tsenyetsa letlakala le la phatlalatso leina"
	+	</code>
	+
	+	The sentence is missing and elipse (...)
	+
	+	<code po>
	+	#: AdvancedProperties
	+	#, fuzzy
	+	#_ endpunc: checks whether punctuation at the end of the strings match
	+	msgid "Advanced Properties..."
	+	msgstr "ditiro tsa maemo a kwa godim"
	+	</code>
	+
	+	Missing colon (:)
	+
	+	<code po>
	+	#: JobTitle.label
	+	#, fuzzy
	+	#_ endpunc: checks whether punctuation at the end of the strings match
	+	msgid "Title:"
	+	msgstr "Setlha"
	+	</code>
	+
	+	The end contains a space before the colon which shouldn't be there.
	+
	+	<code po>
	+	#: addressMessageTo.label
	+	#, fuzzy
	+	#_ endpunc: checks whether punctuation at the end of the strings match
	+	msgid "Address message to:"
	+	msgstr "Attresetsa molaetsa go :"
	+	</code>
	+
	+	Notice that the msgid is incorrect (it only has 2 dors) and the msgstr has been corrected. Leave
	+	this as it is as the msgstr is now correct. Also report this as an error against the application.
	+
	+	<code po>
	+	#: saveAll.label
	+	#, fuzzy
	+	#_ endpunc: checks whether punctuation at the end of the strings match
	+	msgid "Save All.."
	+	msgstr "Boloka tsotlhe..."
	+	</code>
	+
	+	There should be a space before "->"
	+
	+	<code po>
	+	#: fieldMapExport.add
	+	#, fuzzy
	+	#_ endpunc: checks whether punctuation at the end of the strings match
	+	msgid "Add Field ->"
	+	msgstr "Atisa lebala->"
	+	</code>
	+
	+	The msgstr should end in ? but it ends in .
	+
	+	<code po>
	+	#: PKCS12PasswordInvalid
	+	#, fuzzy
	+	#_ endpunc: checks whether punctuation at the end of the strings match
	+	msgid ""
	+	"Could not decode PKCS #12 file. Perhaps the password you entered was "
	+	"incorrect?"
	+	msgstr ""
	+	"Ga ya kgona go sirolola PKCS #12 file. Gongwe lefoko-phetiso leo o le "
	+	"tsentseng le fosagetse."
	+	</code>
	+
	+	The closing bracket should be a round bracket not a curly bracket.
	+
	+	<code po>
	+	#: UnknownCertIssuer
	+	#, fuzzy
	+	#_ endpunc: checks whether punctuation at the end of the strings match
	+	msgid "(Unknown Issuer)"
	+	msgstr "(Mofi ga e itsewe}"
	+	</code>
	+
	+	===== End Whitespace =====
	+
	+	Remove the extra whitespace at the end.
	+
	+	<code po>
	+	#: deleteCardCmd.label
	+	#, fuzzy
	+	#_ endpunc: checks whether punctuation at the end of the strings match
	+	#_ endwhitespace: checks whether whitespace at the end of the strings matches
	+	msgid "Delete Card"
	+	msgstr "Ntsha karata "
	+	</code>
	+
	+	===== Start Punctuation =====
	+
	+	False positive. The translation is correct, although you might want to define a policy on how you handle this form of plural.
	+
	+	<code po>
	+	#: SuccessfulP12Backup
	+	#, fuzzy
	+	#_ endpunc: checks whether punctuation at the end of the strings match
	+	#_ startpunc: checks whether punctuation at the beginning of the strings match
	+	msgid "Successfully backed up your security certificate(s) and private key(s)."
	+	msgstr ""
	+	"(Di/Se)tifikeiti tsa gago tsa pabalelo le (di/se)lotlolo (t)sa bosephiri di/"
	+	"se bolokilwe thuso-morago ka tswelelopele."
	+	</code>
	+
	+	===== Start Whitespace =====
	+
	+	The space at the beginning of the message should be deleted.
	+
	+	<code po>
	+	#: DisableDlgTitle
	+	#, fuzzy
	+	#_ startwhitespace: checks whether whitespace at the beginning of the strings matches
	+	#_ startpunc: checks whether punctuation at the beginning of the strings match
	+	msgid "%S Quick Launch"
	+	msgstr " %S Quick Launch"
	+	</code>
	+
	+	The space at the beginning of the message should be deleted.
	+
	+	<code po>
	+	#: ko
	+	#, fuzzy
	+	#_ startwhitespace: checks whether whitespace at the beginning of the strings matches
	+	#_ startpunc: checks whether punctuation at the beginning of the strings match
	+	msgid "Korean"
	+	msgstr " SeKorea"
	+	</code>
	+
	+
	+	===== Numbers =====
	+
	+	The number 1252 does not appear in the translation.
	+
	+	<code po>
	+	#: windows-1252.title
	+	#, fuzzy
	+	#_ numbers: checks whether numbers of various forms are consistent between the two strings
	+	msgid "Western (Windows-1252)"
	+	msgstr "Turkish (Windows-1254)"
	+	</code>
	+
	+	Might not be a problem if 2 has been written in full
	+
	+	<code po>
	+	#: SSL2Disabled
	+	#, fuzzy
	+	#_ endpunc: checks whether punctuation at the end of the strings match
	+	#_ numbers: checks whether numbers of various forms are consistent between the two strings
	+	msgid "You cannot connect to %S because SSL version 2 is disabled."
	+	msgstr "Ga o kgone go lomagana go %S ka gonne SSL e ntshitswe bokgoni.."
	+	</code>
	+
	+	===== Variables =====
	+
	+	&vendorShortName; should not have been translated.
	+
	+	<code po>
	+	#: throbber.tooltip
	+	msgid "Go to the &vendorShortName; home page"
	+	msgstr "E ya ko &vendorKortNaam; go letlakala la le gae"
	+	</code>
	+
	+	" should not have been translated
	+
	+	<code po>
	+	#: incomingServerNameDesc.label
	+	#, fuzzy
	+	#_ variables: checks whether variables of various forms are consistent between the two strings
	+	msgid "Enter the name of your incoming server (for example, &qout;mail.example.net")."
	+	msgstr "Tsenya leina la moridi~Zi wa gago yo a go romelago molaet~Za (ka mohlala, &tsopolo;poso.mohlala.net&tsopolo;)."
	+	</code>
	+
	+
	+	===== Punctuation Spacing =====
	+
	+	There should be a space after , in "Ka sekai,netscape". A single quote is
	+	also missing.
	+
	+	<code po>
	+	#: abbreviateOn.label
	+	#, fuzzy
	+	#_ puncspacing: checks for bad spacing after punctuation
	+	#_ endpunc: checks whether punctuation at the end of the strings match
	+	msgid "Full names (For example, 'netscape.public.mozilla.mail-news')"
	+	msgstr "Maina ka botlalo (Ka sekai,netscape.public.mozilla.posol-dikgang')"
	+	</code>
	+
	+	Space missing after colon.
	+
	+	<code po>
	+	#: unreadMsgStatus
	+	#, fuzzy
	+	#_ puncspacing: checks for bad spacing after punctuation
	+	msgid "Unread: %S"
	+	msgstr "Ga ja balwa:%S"
	+	</code>
	+
	+	Space missing after comma "kenna,o sutlhe"
	+
	+	<code po>
	+	#: defaultcharactersetBidiCmd.label
	+	#, fuzzy
	+	#_ puncspacing: checks for bad spacing after punctuation
	+	msgid ""
	+	"Use my default character set, overriding the document-specified character set"
	+	msgstr ""
	+	"dirisa ditlhaka jaaka ditlhophilwe kenna,o sutlhe tse di tlhophetsweng "
	+	"tokomane e"
	+	</code>
	+
	+	Space missing after semi-colon "da dk;faele". Also note missing minus between
	+	da and DK.
	+
	+	<code po>
	+	# LOCALIZATION NOTE GROUP : DO not localize the entities below; test case
	+	#: da-DK-file.label
	+	#, fuzzy
	+	#_ puncspacing: checks for bad spacing after punctuation
	+	msgid "Danish (da-DK; file)"
	+	msgstr "danish (da dk;faele)"
	+	</code>
	+
	+	===== Short =====
	+
	+	There is a missing sentence
	+
	+	<code po>
	+	#: SIClueless
	+	#, fuzzy
	+	#_ endpunc: checks whether punctuation at the end of the strings match
	+	#_ endwhitespace: checks whether whitespace at the end of the strings matches
	+	msgid ""
	+	"There are unknown problems with this digital signature. You should not trust "
	+	"the validity of this message until you verify its contents with the sender."
	+	msgstr "Gona le mathata ao a sa itsiweng ka mosaeno o wa dinomoro. "
	+	</code>
	+
	+	===== Long =====
	+
	+	The translation looks too long, it might be right but it is unlikely.
	+
	+	<code po>
	+	#: directionBidiMenu.label
	+	#, fuzzy
	+	#_ endpunc: checks whether punctuation at the end of the strings match
	+	msgid "Default Direction"
	+	msgstr ""
	+	"tshupetso e e diragadiwang fela fa ditshupetso tse dirulagantsweng go "
	+	"diragalapele sitwa godiragala."
	+	</code>
	+
	+	<code po>
	+	#: mediaEncryption
	+	#, fuzzy
	+	#_ endpunc: checks whether punctuation at the end of the strings match
	+	msgid "Encryption:"
	+	msgstr "fetolelo go mokwalo o o fitlhang tshedimosetso ka ga lefoko la tlwaelo"
	+	</code>
	+
	+	===== Unchanged =====
	+
	+	The english has not been translated. The word plugin is translatable. In fact this example show two more errors, the start capital is missing and the colon is also missing. So this translators took an English string and replace it with an untranslated and badly formated string.
	+
	+	<code po>
	+	#: mediaPlugin
	+	#, fuzzy
	+	#_ endpunc: checks whether punctuation at the end of the strings match
	+	msgid "Plugin:"
	+	msgstr "plugin"
	+	</code>
	+
	+
	+	===== URLs and Emails =====
	+
	+	This is a well translated email address in this case it was used as an example so is translated
	+
	+	<code po>
	+	#: emailExample.label
	+	#, fuzzy
	+	#_ endpunc: checks whether punctuation at the end of the strings match
	+	msgid "(for example: user@example.net)."
	+	msgstr "(ka mohlala: modirisi@mohlala.net)"
	+	</code>

guide:preparing_files_for_wordfast

Tue, 07 Aug 2007 09:51:02 +0000

1186478994		current
Line 90:		Line 90:
	==== How to create Wordfast styles (if you can install external macros) ====		==== How to create Wordfast styles (if you can install external macros) ====

-	If you know how to install external macros (i.e. if you know where you should copy a file in MS Windows' hidden folder structure), you can install AndoTools into MS Word, which has a function to insert all tw4win styles into any document easily. Once you've installed AndoTools, in MS Word go Ando -> Documenet Operations -> Fonts and Language. Click "Add tw4win styles" to add them to the current document.	+	If you know how to install external macros (i.e. if you know where you should copy a file in MS Windows' hidden folder structure), you can install AndoTools ([[http://atools.dotsrc.org/\|here]]) into MS Word, which has a function to insert all tw4win styles into any document easily. Once you've installed AndoTools, in MS Word go Ando -> Documenet Operations -> Fonts and Language. Click "Add tw4win styles" to add them to the current document.

	===== How to prepare a l10n file for Wordfast =====		===== How to prepare a l10n file for Wordfast =====
Line 209:		Line 209:
	The macro is now added to normal.dot, and can be used for any document that is opened in MS Word. Incidently, the above macro does exactly what we did in the advanced find/replace operation above.		The macro is now added to normal.dot, and can be used for any document that is opened in MS Word. Incidently, the above macro does exactly what we did in the advanced find/replace operation above.

		+	=== Adding a macro from an existing document ===

		+	Adding a macro to normal.dot from an existing document is similar to what we did in the "easy way" for adding styles. I assume you have a document with a macro embedded in it. I've embedded the above macro for you, in a document {{guide:with_apple.doc\|here}}.
		+
		+	To add the macro to normal.dot, here's how:
		+
		+	- Open a blank document in MS Word.
		+	- In MS Word, go Tools -> Macro -> Macros.
		+	- Click "Organizer". It should show you the macros in with_apple.doc and normal.dot.
		+	- Select "apple" and click Copy to copy to normal.dot.
		+	- Close both files, and click OK etc.
		+
		+	And that's it. Now a macro called apple.apple is part of normal.dot, and can be used on any document you open in MS Word.
		+
		+	=== How to execute a macro on a document ===
		+
		+	When running the macros described above, it is assumed that you have tw4winExternal as a style in normal.dot, or in the document that you're about to tag. What we're going to do, is to run the macro apple or apple.apple, which will perform the find/replace operation mentioned previously. This will mark the necessary text as "untranslatable", so that Wordfast will ignore it.
		+
		+	- Open the l10n file in MS Word (described above)
		+	- In MS Word, go Tools -> Macro -> Macros.
		+	- Select the macro apple or apple.apple in the list, and click "Run" (if you can't see the macro, it is either not in the normal.dot, or the normal.dot is not selected in the dropdown list).
		+	- Click OK etc.
		+
		+	If everything went well, your document should now be tagged, as per the images above.
		+
		+	(next write a short intro, plus upload a number of macros for XLIFF, TMX, PO, etc.
		+
		+	==== Another way... ====
		+
		+	Some of what is written here, is re-inventing the wheel. The wheel we're talking about, is Tortoise Tagger, [[http://www.accurussian.net/tagger.htm\|here]].

guide:project_objectives

Sat, 05 Jan 2008 18:23:18 +0000

1199556660		current
Line 21:		Line 21:

	A real world example. The Translate.org.za project initially selected the		A real world example. The Translate.org.za project initially selected the
-	[[guide:KDE desktop]] as its first localisation project. The goal was a little more defined in that we	+	[[kde_desktop]] as its first localisation project. The goal was a little more defined in that we
	hoped to impact users at the same time as promoting Linux. When we readjusted		hoped to impact users at the same time as promoting Linux. When we readjusted
	our objectives we realised that if we wanted to impact language speakers then		our objectives we realised that if we wanted to impact language speakers then
-	we would have to be localising cross-platform projects such as [[guide:OpenOffice.org]]	+	we would have to be localising cross-platform projects such as [[openoffice.org]]
-	and [[guide:Mozilla]]. Thus our objectives have shifted.	+	and [[mozilla]]. Thus our objectives have shifted.

	Another example is that of a localiser who I met who reiterated the Eastern Blowfish		Another example is that of a localiser who I met who reiterated the Eastern Blowfish

guide:purpose

Sat, 05 Jan 2008 18:22:55 +0000

1122639279	current
Line 1:	Line 1:
	+	====== Purpose ======

	+	The purpose of this document is to provide one reference for localisers. You
	+	will find lots of information on localising and packaging on the web but not a
	+	single resource that can guide you. Most of the information is also domain
	+	specific ie it addresses KDE, Mozilla, etc. We hope that this is more general.
	+
	+	This document also goes beyond the technical aspects of localisation which
	+	seems to be the domain of other localisation documents. The document aims to
	+	help other localisation teams by consolidating the information of the
	+	[[http://translate.org.za\|Translate.org.za]] project and other localisation initiatives. So in this
	+	document we have added sections on selecting what to translate, budgeting,
	+	timing and more.
	+
	+	There are sections explaining how to use the [[toolkit:index\|Translate Toolkit]] and practical
	+	sections on common translation technical errors as well as ideas on how to
	+	translate or configure certain things like accelerators key, whether to
	+	translate acronyms, etc.
	+
	+	===== Duplication =====
	+
	+	We hope not to duplicate information and will point to other online resources
	+	if needed. But often we document in the process of creating our translation
	+	builds so there might be slight duplication.

guide:short_strings_first

Sun, 30 Aug 2009 00:07:12 +0000

1199557349		current
Line 1:		Line 1:
	====== One, two, short, long ======		====== One, two, short, long ======

-	How do you translate the most quickly for the biggest impact. If you look at a	+	How do you translate most quickly for the biggest impact? If you look at a GUI application you will notice that most strings are relatively short. Menu items are between one and three words long. Error messages are more descriptive at perhaps 5-10 words. So in general the items that users see in normal operation are relatively short.
-	GUI application you will notice that most strings are relatively short. Menu	+
-	items are between one and three words long. Error messages are more	+
-	desctiptive at perhaps 5-10 words. So in general the items that users see in	+
-	normal operation are relatively short.	+

-	Therefore, if you can initialy focus your translation on the short strings then	+	Therefore, if you can initially focus your translation on the short strings then you can make a quick visual impact. Afterwards you can follow up with longer and longer strings until the application is fully translated.
-	you can make a quick visual impact. Afterwards you can follow up with longer and longer	+
-	strings until the application is fully translated.	+

guide:skype

Wed, 02 Apr 2008 11:26:32 +0000

1207134869		current
Line 2:		Line 2:

	Arguably the most important VoIP solution that can be localised. This [[http://share.skype.com/sites/devzone/2006/03/localization_tricky_but_import.html\|Skype devzone]] article shows the importance of localisation - ony 28% of users of Skype make use of an English interface.		Arguably the most important VoIP solution that can be localised. This [[http://share.skype.com/sites/devzone/2006/03/localization_tricky_but_import.html\|Skype devzone]] article shows the importance of localisation - ony 28% of users of Skype make use of an English interface.
		+


Line 8:		Line 9:

	===== How to localise Skype =====		===== How to localise Skype =====
		+
		+	* [[http://forum.skype.com/lofiversion/index.php/f7.html\|Skype Translation]] forum
		+	* [[http://forum.skype.com/lofiversion/index.php/t111756.html\|Adding unsupported languages]] (Windows)

	Skype makes use of Qt [[toolkit:.ts]] files for localisation. Thus a compiled [[toolkit:.qm]] file should bring localisation to your Skype GUI. Unfortunately, certainly on the Linux builds, Skype does not seem to detect languages based on the presence of a language file, rather it seems to have a pre-set list of possible interface languages. Rather sad as this would have been an easy GUI to localise based on the power or Qt.		Skype makes use of Qt [[toolkit:.ts]] files for localisation. Thus a compiled [[toolkit:.qm]] file should bring localisation to your Skype GUI. Unfortunately, certainly on the Linux builds, Skype does not seem to detect languages based on the presence of a language file, rather it seems to have a pre-set list of possible interface languages. Rather sad as this would have been an easy GUI to localise based on the power or Qt.

guide:specific_phrases_in_afrikaans

Sat, 05 Jan 2008 18:21:47 +0000

1190038717	current
Line 1:	Line 1:
	+	====== Translation of specific phrases in Afrikaans ======
	+
	+	unable to
	+
	+	Unable to (v) = Kan nie (v) nie
	+	Unable to (v) (n) = Kan nie (n) (v) nie
	+
	+	error
	+
	+	Error (v)ing (n) = Kon nie (n) (v) nie
	+	Error occurred (v)ing (n) = Kon nie (n) (v) nie
	+	Error = Fout
	+	Error occurred = 'n Fout het voorgekom
	+	An error occured while = Kon nie ... volledig ... nie
	+	There were errors (v)ing (n) = (n) kon nie volledig (v) nie
	+	There were errors (v)ing (n) = Kon nie (n) volledig (v) nie
	+	(n) error = (n)fout
	+	(v) error = (v)fout
	+	Error in (n) = Fout in (n)
	+	returned an error = 'n fout aangegee
	+
	+
	+

guide:spell_checker_case_studies

Wed, 26 Nov 2008 13:46:25 +0000

	current
Line 1:	Line 1:
	+	====== A spellchecker for your language ======

	+	Does the current spellchecking software work well for your language? As software is translated into more and more languages, we are finding that writing aids software based on European-language structures is less and less effective.
	+
	+	So what can we do about it? It's time to review our current resources, and work together to create writing aids tools appropriate to a wider range of languages, or appropriate to specific languages. Some case-studies are discussed below.
	+
	+	This page will link to discussions on this subject in different mailing lists, to wiki pages detailing it for and in different languages, and to software being developed, or which can be used, to meet these needs.
	+
	+	Please add any information you think could be useful. :)
	+
	+	===== Case Studies =====
	+
	+
	+	==== Vietnamese ====
	+
	+	=== Nguyễn Thái Ngọc Duy ===
	+	I have been working on vspell, a Vietnamese spell checker, since 2003
	+	(there were huge gaps when I did not work on it at all, though). The
	+	source code can be found [[http://repo.or.cz/w/vspell.git\|here]] (click on the first "snapshot" link to get a tarball). The core idea is quite simple. It is
	+	trained with a word-segmented corpus. When a sentence is given
	+	(actually a phrase because it still does not understand "sentence"),
	+	it will generate similar sentences based on common spelling errors. It
	+	then uses statistics from the corpus to determine which sentence (the
	+	original one or one of the generated ones) is "better". If a generated
	+	one is better, then it assumes the original one is misspelled. That's
	+	all. The rest of work is matching the original one and the "right" one
	+	to see differences between them and tell users about that.
	+
	+	The result as of three months ago was not very promising: precise rate
	+	was about 60%-70% (I expected at least 80% to be useful). I was
	+	investigating to see why the precise rate was low, and had some
	+	technical difficulties.
	+
	+
	+	I've been working on a spellng checker for Vietnamese. The problems are:
	+	* It requires input as phrases, not words. As far as I can tell, hunspell and almost all spelling checkers take input as words.
	+	* It uses a lot of memory (≥ 256MB for itself)
	+	* It needs to be trained to be able to differentiate good phrases and bad phrases. There are some difficulties in my training method (I don't know, I may give up on my method, as it is becoming infeasible, something with number explosion).
	+	* It's not stable. I spent most of my time training it and testing it with a small sample. It liked to crash back then ;)
	+
	+	The first problem means we must interact with Openoffice without
	+	Hunspell. That's not easy to do.
	+
	+	But the main problem is the third. My approach is statistics-based. It
	+	requires lots of (correct) word-aligned sentences to be trained on.
	+	That kind of corpus for Vietnamese does not exist (at least freely).
	+	So my workaround is to take a raw corpus and train repeatedly to get
	+	better result each iteration. The workaround has number explosion
	+	problem. It breaks 32-bit integer limit easily and also "long double"
	+	limit. In short, until I find a feasible training method, my spell
	+	checker is no use.
	+
	+	=== The basic problem in Vietnamese ===
	+
	+	Vietnamese "words" are often composed of more than one word. Words are usually monosyllabic, so we can think of them as syllables of these longer words. However, current spellchecking tools treat each Vietnamese "syllable" as a separate word. This means that when you make a mistake that is still a valid word, e.g. typing màu hình instead of màn hình, current spellchecking tools will still recognize « màu » and « hình » as valid separate words, and not detect the error.
	+
	+
	+	=== Ivan Garcia ===
	+	(Note: Ivan's current [[http://code.google.com/p/hunspell-spellcheck-vi/\|Hunspell]] spellchecking dictionary is being used in OpenOffice.org, and in Firefox and Thunderbird.)
	+
	+	=== Duy ===
	+
	+	I know my statistical method isn't ideal. A rule-based approach would be more realistic. But the rule-based one requires human power to build the ruleset. A rule generation approach like [[http://stp.ling.uu.se/~torbjorn/Mutbl/bibliography.html\|TBL]] (Tranformation Based Learning) requires an annotated corpus, which I don't have.
	+
	+	Identifying composed words is also what I want my spellchecker to do.
	+
	+	I don't think a grammar-checker is viable, due to the complexity of Vietnamese grammar. My spellchecker is basically a spellchecker. Although it could also be able to detect some semantic/grammatic mistakes as well.
	+
	+	Even using current grammar-checking tools, you will have more troubles with Vietnamese ;) Before you discuss grammar, you must split a sentence into words (actually annotated words but that's not the point). It's already difficult to do that in Vietnamese. Now you are supposed to do that on a misspelt text.
	+	Good luck :D
	+
	+	European languages don't have this problem, as distinct words can be easily
	+	recognized. CJK languages do though, but I guess CJK spellchecker
	+	status in OOo is just the same as it is for Vietnamese.
	+
	+	To have an idea how hard it is to split a sentence into words, let's
	+	take a corner case: « Ông già đi nhiều quá ». You can understand this
	+	sentence in a couple of ways:
	+
	+	* An old man goes a long way (Ông-già đi nhiều quá, notice « Ông già » is a single word)
	+	* He gets very old (Ông già đi nhiều quá, notice « Ông già » is two separate words)
	+
	+	Now suppose « già » is mistakenly written as « dà », then pass the sentence
	+	to a grammar checker ;)
	+
	+	=== Could we train a spellchecker? ===
	+
	+	Is it possible to train a spellchecker, as one trains a Bayesian spamfiltering program like [[http://c-command.com/spamsieve/\|SpamSieve]] on OSX? If we have a group of volunteers each building up a corpus, and training the spellchecker each time it makes a mistake, perhaps we could amass the data we need.
	+
	+	=== This discussion ===
	+
	+	You can follow this discussion on the specially-created [[http://lists.hanoilug.org/listinfo/spell\|anoi-spell]] mailing list, or on the [[https://lists.sourceforge.net/lists/listinfo/translate-devel\|Translate Toolkit]] mailing list.
	+
	+	==== Quechua ====
	+
	+	=== Amos Batto ===
	+
	+	It is interesting hearing about the difficulties of
	+	spell checking in Vietnamese. I have also run into
	+	problems using Hunspell to spellcheck Quechua (an
	+	indigenous language of the Andes), but of a very
	+	different nature.
	+
	+	Quechua is an agglutinative
	+	language. Most words have 1 or 2 suffixes, but some
	+	can have as many as 8 or 9 suffixes. Most suffixes
	+	have to be added in a very specific order, but a few
	+	can appear in almost any order. Needless to say, the
	+	possible combinations of suffixes is almost infinite
	+	and almost impossible to list in an affix file. When I
	+	tried to write out all the possible combinations, I
	+	got up to 500 pages of combinations and that was only
	+	combining 1, 2, and 3 suffixes.
	+
	+	Hunspell is much
	+	better than aspell and ispell because it allows an
	+	infix and 2 levels of suffixes, but it is still
	+	woefully inadequate for languages like Quechua.
	+
	+	Another problem of Hunspell is a lack of a "sounds
	+	like" feature, as is found in Aspell. In Quechua, K,
	+	K', KH, Q, Q', QH are all easily confusable letters,
	+	along with their Spanish equivalents: QU and C. In
	+	order to properly spellcheck in Quechua a "sounds
	+	like" function needs to be added to Hunspell. The code
	+	for "sounds like" shouldn't be that difficult to
	+	write, but the Hunspell code looks pretty complicated
	+	and I haven't figured it out.
	+
	+	//Note: Kevin Scannell, who was one of the developers of
	+	Aspell contacted me to say that he intended to add the
	+	Aspell "SoundsLike" function to Hunspell. I hope that he
	+	finds the time to do this since it will help me greatly.
	+	Kevin also noted that the metaphone function in Hunspell
	+	could act like SoundsLike to some degree.//
	+
	+	If you have more interest in learning about the challenges
	+	of a Quechua spellchecker, see [[http://www.runasimipi.org/blanco-en.php?file=cartaHunspellPreguntas-en.txt\|this note]] I wrote to the
	+	Hunspell developer explaining our difficulties.
	+
	+	==== Zulu ====
	+
	+	=== Friedel Wolff ===
	+
	+	The issue of agglutinative languages is quite interesting for me since
	+	we are working on spellchecking in Zulu which is also of this nature.
	+	We are starting to work now on a program to help people review word
	+	lists and to identify word roots. This is all done under the assumption
	+	that identifying roots is the most important part of the work in terms
	+	of the words and word lists, and combined with proper affix rules
	+	(developed separately), we can create a usable spell checker (for an
	+	agglutinative language).
	+
	+	People interested can have a look [[http://translate.sourceforge.net/specifications/Spelt/\|here]] at our ideas for how this should
	+	work. This is currently really meant to be a small project that can be
	+	implemented quite quickly.
	+
	+
	+	==== Statistical Support ====
	+
	+	Kevin Scannell can generate word frequency lists and other useful statistics for many languages (more than 400 as of May 2008) using his web crawling software [[http://borel.slu.edu/crubadan/\|An Crúbadán]]. There's a good chance your language is already supported if it has a non-trivial presence on the web.
	+	Contact him (kscanne at gmail dot com) if you are beginning development of a spell checker and plan on releasing it under and open source license.
	+
	+	==== Your language ====

guide:spell_quality.py

Fri, 27 Feb 2009 10:19:03 +0000

1235729913		current
Line 11:		Line 11:
	</code>		</code>

-	The corpus files should include one single word per line. It can be prepared based on any given corpus with commands similar to	+	The corpus files should include one single word per line and should be in the UTF-8 encoding. It can be prepared based on any given corpus with commands similar to

	<code>		<code>

guide:spelling_checkers

Mon, 30 Nov 2009 07:12:24 +0000

1227706967		current
Line 16:		Line 16:

	* [[http://speling.org/\|Scandinavian spell checkers website with some useful tools]].		* [[http://speling.org/\|Scandinavian spell checkers website with some useful tools]].
-	* [[http://dict-common.alioth.debian.org/dsdt-policy.txt\|Debian Spellchecker packaging policies]]	+	* [[http://dict-common.alioth.debian.org/dsdt-policy.html\|Debian Spellchecker packaging policies]] ([[http://dict-common.alioth.debian.org/\|Website]])
	* [[http://lingucomponent.openoffice.org/spell_dic.html\|OpenOffice.org Lingucomponent]]		* [[http://lingucomponent.openoffice.org/spell_dic.html\|OpenOffice.org Lingucomponent]]
	* [[http://www.rhedadur.org.uk/index.php?lg=en\|Konjugator]] is a browser-based conjugator for Welsh verbs. This might be usefull for developing your own spelling and grammar checkers in your language.		* [[http://www.rhedadur.org.uk/index.php?lg=en\|Konjugator]] is a browser-based conjugator for Welsh verbs. This might be usefull for developing your own spelling and grammar checkers in your language.

guide:start

Sat, 10 Oct 2009 05:12:56 +0000

1255151576		current
Line 127:		Line 127:

	* [[guide:tools:glossary\|Glossary creation and management]]		* [[guide:tools:glossary\|Glossary creation and management]]
-	* [[http://mac.gettranslateit.com/\|TranslateIt]] on OSX provide system-wide translation into single or multiple languages, as well as glossary creation	+	* [[http://mac.gettranslateit.com/\|TranslateIt]] on OSX provides system-wide translation into single or multiple languages, as well as glossary creation

	==== Source Control ====		==== Source Control ====

guide:statistics

Sat, 24 Oct 2009 07:49:18 +0000

1227551967		current
Line 27:		Line 27:
	* http://i18n.kde.org/stats/gui/trunk/index.php		* http://i18n.kde.org/stats/gui/trunk/index.php

		+
		+	==== UPDATE 24.10.2009 ====
		+
		+	The scripts for the KDE statistics page as seen here:
		+
		+	* http://i18n.kde.org/stats/gui/trunk-kde4/team/
		+
		+	Are generated using the l10n-stats scripts, which can be obtained here:
		+
		+	* http://i18n.kde.org/about-stats.php

	===== Kumulate =====		===== Kumulate =====

guide:suse

Sat, 05 Jan 2008 18:20:29 +0000

1195980700		current

guide:svntips

Sun, 29 Jun 2008 06:27:03 +0000

1154850834		current
Line 11:		Line 11:
	How does this affect us translators? Source control means we have access to the latest .po file at any time.		How does this affect us translators? Source control means we have access to the latest .po file at any time.

-	//Note: a GUI front-end for SVN like [[http://www.lachoseinteractive.net/en/community/subversion/svnx/features/\|svnX]] makes using SVN even easier!//	+	//Note: a GUI front-end for SVN like [[http://www.lachoseinteractive.net/en/community/subversion/svnx/features/\|svnX]] makes using SVN even easier! Your [[http://www.cocoatech.com/\|file-browser]] might also have embedded SVN functions.//

	====How do I get the files?====		====How do I get the files?====

guide:testing

Mon, 02 Jul 2007 10:43:44 +0000

	current
Line 1:	Line 1:
	+	====== Testing Translations ======

	+	How do you test the translations that you have created? In some cases it is almost impossible to quickly test since you need to compile the application. The following are guides for various systems to allow you to quickly test your translations.
	+
	+	===== Cannot test =====
	+
	+	These cannot be quickly tested:
	+
	+	* OpenOffice.org - requires the application be recompiled
	+	* Mozilla Firefox and Thunderbird - requires the creation and installation of an XPI
	+
	+	===== Gettext =====
	+
	+	This works on Linux or any system where there is a default location for the compiled PO files.
	+
	+	<code>
	+	$ mkdir -p $HOME/.local/share/locale/$your_language_code/LC_MESSAGES
	+	$ pocompile $your_application.po $HOME/.local/share/locale/$your_language_code/LC_MESSAGES/$your_application.mo
	+	</code>
	+
	+	Substitute $your_language_code for your iso639 language code e.g. zu for Zulu. And substitute $your_application for the application that you are translating (actually this is the name of the text domain which might be slightly different from the application name but is usually the same as the PO filename).
	+
	+	Now we have a local repository for storing our custom .mo file. And we have compiled one and placed it in the correct location. You can substitute ''pocompile'' with ''msgfmt'' in the above example.
	+
	+	<code>
	+	$ export LANGUAGE=../../../$HOME/.local/share/locale/$your_language_code:$your_language_code
	+	$ $your_application
	+	</code>
	+
	+	This will firstly force us to look into our custom location for MO files and use those first then fallback to those in the system location. Now when you run the application you will see your translations and can quickly test them.

guide:thesaurus

Sat, 05 Jan 2008 18:20:12 +0000

1133861178	current
Line 1:	Line 1:
	+	When the thesaurus function is activated in a program, it supplies the user with possible replacements for the highlighted word or word group. It can indicate the word types, synonyms, related words, abbreviations or complete forms, more desirable forms, etc.

	+	===== OpenOffice.org =====
	+	* Download MyThes
	+	* create “dat” file with thesaurus information:
	+
	+	ISO8859-1
	+	talented\|1
	+	(adj)\|gifted\|untalented (antonym)
	+	talk\|7
	+	(noun)\|talking\|conversation (generic term)
	+	(noun)\|discussion (generic term)\|discourse (generic term)
	+	(noun)\|lecture (generic term)\|lecturing (generic term)
	+	(noun)\|lecture\|public lecture\|speech (generic term)
	+	(verb)\|speak\|communicate (generic term)
	+	(verb)\|spill the beans\|let the cat out of the bag\|keep quiet (antonym)
	+	(verb)\|lecture\|teach (generic term)\|instruct (generic term)
	+
	+	* run the indexing program that is part of MyThes
	+	* add both the “dat” file and the “idx” file to the “dict” directory in the installation
	+	* add a line like the following to “dictionary.lst”:
	+
	+	THES en US th_en_US_v2
	+
	+	* restart OpenOffice.org (and the quickstarter in Windows, if used)
	+
	+	Select a word (like "talented" or "talk") and activate the thesaurus. The thesaurus can be accessed under Tools->Language.

guide:thunderbird

Thu, 26 Mar 2009 15:22:44 +0000

1199557192		current
Line 1:		Line 1:
-	====== Localizing Mozilla Thunderbird ======	+	====== Thunderbird ======

-	FIXME These docs were migrated from the old Wiki. They have not been checkd and could easily be incorrect. Your fixes are most welcome	+	Thunderbird is the e-mail client that was born out of the [[Mozilla]] project.

-	FIXME {omgs: how to get po files from CVS for Firefox and Thunderbird 1.5 is in the new [[mozillacvs]] page}	+	This page contains Thunderbird-specific localisation notes and issues. You should read this in conjunction with the [[Mozilla]] pages as there are some things shared between Thunderbird and [[Firefox]] localisation.

-	===== Using Mozilla Firefox POT/PO files =====	+	As stated on the [[Mozilla]] page, there are more complete instructions about the requirements and processes involved in Mozilla localisation on the [[https://wiki.mozilla.org/L10n:Home_Page\|Mozilla L10n]] project page.

-	This section outlines few steps that can make localizing Mozilla	+	===== Links =====
-	Thunderbird e-mail client somewhat easier and faster. This	+	* [[http://thunderbird.dnsalias.net/instructions/thunderbird-l10n-OSX.html\|Packaging a localised Max OS X version]].
-	only makes sense if you already have a localized version of Mozilla	+
-	Firefox.	+
-		+
-	In short, a faster translation process is possible because Mozilla	+
-	Firefox and Mozilla Thunderbird share some of the files that need to be	+
-	translated, and thus a translator can reuse part of the work that is	+
-	already done in the localized Firefox.	+
-		+
-	Firefox and Thunderbird are developed to share the 'toolkit' folder. If	+
-	you've done some l10n work on Firefox than you probably know that the	+
-	files needed for a l10n build are separated in two folders: 'browser'	+
-	and 'toolkit'.	+
-		+
-	Firefox's toolkit folder contains the following folders:	+
-	autoconfig/ global/ global-platform/ global-region/ mozapps/ necko/	+
-	passwordmgr/ pipnss/ pippki/	+
-		+
-	and at this stage with little or no effort you can reuse everything	+
-	exept 'autoconfig' and 'passwordmgr'.	+
-		+
-	===== A step-by-step guide (on GNU/Linux) =====	+
-		+
-		+
-	<code bash>	+
-	bash$ cp /some-path/thunderbird/chrome/en-US-mail.jar tb-l10n/	+
-	# Copy the jar file to favored localization folder.	+
-	bash$ cd tb-l10n/	+
-	bash$ unzip en-US-mail.jar	+
-	# Unpack the jar file. The result is a 'locale' folder.	+
-	bash$ moz2po -P --progress=verbose -i locale/ -o locale-pot/	+
-	# Convert to POT templates.	+
-	bash$ cd locale-pot/	+
-	bash$ rm -r pippki pipnss necko global-region	+
-	# These are equal with Firefox ones.	+
-	bash$ mv mozapps mozapps-tb	+
-	bash$ mv global global-tb	+
-	bash$ mv global-platform global-platform-tb	+
-	# Just to be on the safe side here. Some files form these folders are	+
-	# needed. This may seem like an redundant step, because you can simply	+
-	# enter these folder and delete everything but the needed file. :)	+
-	bash$ mkdir -p mozapps global global-platform/win	+
-	bash$ cp mozapps-tb/pref-downloads.dtd.pot mozapps/	+
-	*bash$ cp global-tb/brand. global/**	+
-	*bash$ cp global-tb/brand. global/**	+
-	bash$ cp global-tb/charset.dtd.pot global/	+
-	bash$ cp global-platform-tb/win/nsWindowsHooks.properties.pot global-platform/win/	+
-	# Copying the few files that are different. And now you can delete those	+
-	# '-tb' folders.	+
-	*bash$ mv en-US/ .**	+
-	# Moving files form en-US to current directory. Don't know if this is	+
-	# really needed. I just recreated the directory structure from the Russian	+
-	# XPI.	+
-	</code>	+
-		+
-	These are the files that await your translating skills so you can now start your favorite PO editor.	+
-		+
-	If you want to have all the files in one place you can copy the missing	+
-	po and pot file from Firefox (just don't overwrite the few previously	+
-	selected files).	+
-		+
-	When your translation is finished is back to po2moz, and building a XPI	+
-	(you can use some other Thunderbird language pack XPI as a template for your own).	+
-		+
-	===== Aditional Migration =====	+
-		+
-	You can also now make use of pomigrate and friends. These are bash scripts that wrap the	+
-	Gettext tools. They can often increase the level of translationa as they will apply a translation	+
-	compendium to your files.	+

guide:tortoise

Sat, 05 Jan 2008 18:19:34 +0000

1199556883		current
Line 9:		Line 9:
	from your Windows machine.		from your Windows machine.

-	See also [[guide:WinCVS]] as an alternative to Tortoise CVS.	+	See also [[wincvs]] as an alternative to Tortoise CVS.

	===== Resources =====		===== Resources =====

guide:translateathon

Fri, 20 Feb 2009 02:53:31 +0000

1235097209		current
Line 122:		Line 122:

	This gives you 4 hours of translation time with none more than an hour long.		This gives you 4 hours of translation time with none more than an hour long.
-	Adjust as needed. If you have a mixture of new and mature translators then it	+	Adjust as needed. If you have a mixture of new and experienced translators then it
-	might be nice to arrange the venue so that mature translators don't have to	+	might be nice to arrange the venue so that experienced translators don't have to
	listen to any of the talks.		listen to any of the talks.

guide:translateathon_halloffame

Thu, 14 Aug 2008 14:49:34 +0000

1210782921		current
Line 11:		Line 11:
	\| April 2007 \| AITI, Accra, Ghana \| Yoruba, Twi, Wolof, Hausa, Gun, Akan, Vai, Yemba, N'ko, Krio, Amharic \| FOSSFA, Translate.org.za\| Tuxpaint, Worpress subset \| +- 600 per language\| :-) \|		\| April 2007 \| AITI, Accra, Ghana \| Yoruba, Twi, Wolof, Hausa, Gun, Akan, Vai, Yemba, N'ko, Krio, Amharic \| FOSSFA, Translate.org.za\| Tuxpaint, Worpress subset \| +- 600 per language\| :-) \|
	\| May 2007 \| Rhodes University, Grahamstown, South Africa \| Xhosa \| SANTED, Translate.org.za \| Horde and IMP \| 10 000 words \| :-) \|		\| May 2007 \| Rhodes University, Grahamstown, South Africa \| Xhosa \| SANTED, Translate.org.za \| Horde and IMP \| 10 000 words \| :-) \|
		+	\| Aug 2008 \| Makerere University, Kampala, Uganda \| Luganda \| Makerere, Rhodes, Translate.org.za \| Firefox \| ? words \| :-) 300 students \|

	===== Comment on your event =====		===== Comment on your event =====

guide:translation_comments

Sat, 05 Jan 2008 18:19:13 +0000

1122924419	current
Line 1:	Line 1:
	+	====== Translation Comments ======

	+	Why do you need translation comments?
	+
	+	Consider this simple case; you have a menu item called "Manual". You know what
	+	it means, but when the translator sees this they will wonder did you mean:
	+
	+	- a document or help manual, or
	+	- a manual process?
	+
	+	This is the simplest case where a translation comment such as "The installation manual" helps to clarify
	+	the situation and makes a translator more productive.
	+
	+	===== More examples of the need for translation comments =====
	+
	+	Real world examples are best. This is a discussion over the use of the word
	+	"Forward" in Northern Sotho:
	+
	+	"When you go forward. You go 'Pele', But when you forward the document, O
	+	'Fetišetša pele', so if you just say forward, we don't know what you are
	+	talking about, it is better if it is in a sentence. But in this case i think
	+	we will use 'pele' because on the string no:86 and
	+	88 there is "show previous page in history" and "show next page in history"
	+
	+	Was there translators guess correct? I think so, but it makes it so much easier if
	+	they don't need to be super sleuths as well as translators.
	+
	+
	+	===== What should be commented? =====
	+
	+	==== Ambiguous words ====
	+
	+	Words that can be used as a noun or verb in English but may have different
	+	names in another language
	+
	+	==== Environment variables, Config files entries ====
	+
	+	Is it really obvious that TERM should not be translated?
	+
	+	==== Filenames ====
	+
	+	Filenames that you do not want translated eg .Xauthority. Especially if they
	+	appear translatable.
	+
	+	==== Program Names ====
	+
	+	Some sentences do not make it clear that the text refers to a program name
	+	which you probably don't want to translate.
	+
	+	==== Protocols ====
	+
	+	Especially when they appear as URIs eg help:index. Also protocol responses eg.
	+	"Server returned HELO" very tempting to translate. Or "KEEPALIVE response".
	+	Should KEEPALIVE be translated?
	+
	+	===== Types of comments =====
	+
	+	The types of comments depend very much on the localisation system implemented:
	+
	+	==== KDE style comments ====
	+
	+	The appear to the translator like this:
	+
	+	<code po>
	+	msgid ""
	+	"_: The installation manual\n"
	+	"Manual"
	+	</code>
	+
	+	One advantage is that they can double as context strings. In Gettext PO files
	+	you can only have one translation of the source English word. So if your
	+	language translates an English word differently depending if it is a noun or a
	+	verb then you will have a problem. Context information in a KDE style comment
	+	essentially means that the stings are different and thus allows you to
	+	distinguish between verb and nouns spelled the same way in English.
	+
	+	==== Gettext comments ====
	+
	+	These appear in the # comments of a message block. They are sometimes
	+	overlooked by translators as usually the PO editing tool does not make them
	+	obvious.
	+
	+	===== Adding comments =====
	+
	+	FIXME not sure how to do it please refer to the Gettext manual and KDE programmers manual

guide:translation_guidelines

Sat, 15 Nov 2008 14:09:49 +0000

1222020766		current
Line 19:		Line 19:
	* Translation of [[specific_phrases_in_afrikaans\|specific phrases]]		* Translation of [[specific_phrases_in_afrikaans\|specific phrases]]
	* [[translation_guidelines_german\|German]]		* [[translation_guidelines_german\|German]]
		+	* [[translation_guidelines_hebrew\|Hebrew]]
	* Ndebele, South		* Ndebele, South
	* Northern Sotho		* Northern Sotho

guide:translation_guidelines_german

Tue, 23 Sep 2008 11:01:09 +0000

1222166446		current
Line 96:		Line 96:
	Kann ich keine empfehlen (da ich keine benutze). Im Zweifelsfall einfach mal bei einer anderen Anwendung nachgucken. Auch Wikipedia hilft oftmals bei der Suche nach dem passenden Begriff weiter.		Kann ich keine empfehlen (da ich keine benutze). Im Zweifelsfall einfach mal bei einer anderen Anwendung nachgucken. Auch Wikipedia hilft oftmals bei der Suche nach dem passenden Begriff weiter.

-	Wer einfach nur ein Online-Wörterbuch such ist bei [[http://www.leo.org/\|leo.org]] ganz gut aufgehoben (Vorsicht bei Google-Sprachtools & Co.)	+	Pootle listet bei der Übersetzung möglicherweise relevante Wörter mit ihren Übersetzungen aus dem Terminologie-Projekt rechts auf. Oftmals wurden dort alternative Übersetzungen als Anmerkung hinterlegt welche als Tooltip verfügbar sind (den Mauszeiger für eine kurze Zeit über dem Wort positionieren).
		+
		+	===== Wörterbücher =====
		+	Wer ein Online-Wörterbuch sucht ist bei [[http://www.leo.org/\|leo.org]] ganz gut aufgehoben (Vorsicht bei Google-Sprachtools & Co., da sie oftmals nur eine Übersetzung für das Wort auflisten und nicht wie leo.org auch alternative Bedeutungen)
		+
		+	„Analoge“ Wörterbücher sind für die Übersetzung von Anwendungen meist eher weniger geeignet, da Online-Wörterbücher oftmals einen größeren Wortschatz in Richtung EDV aufweisen. Auch hier sollte man auf einen ausreichend großen Wortschatz achten und auch alternative Bedeutungen eines Wortes durchlesen.

guide:translation_guidelines_hebrew

Sat, 15 Nov 2008 14:11:47 +0000

		current
Line 1:		Line 1:
		+	Everyone, use the [[http://carmel.whatsup.org.il/he/\|Carmel word list]]. It will ease your job.

guide:translation_guidelines_template

Sat, 05 Jan 2008 18:18:34 +0000

1122975534	current
Line 1:	Line 1:
	+	====== Translation Guidelines (Template) ======

	+	This is a template, use it to start your own guidelines document. Also update it with improvements you make in your document.
	+
	+	===== Capitalisation =====
	+
	+	==== Title Case ====
	+
	+	Title Case if where you make the first letter of each word capital except words like: is, are, a, an.
	+
	+	This is Title Case
	+	This is sentence case
	+
	+	If you see something in title case do you change it so sentence case?
	+
	+	===== Plurals =====
	+
	+	==== Gettext Plural Header ====
	+
	+	What should your gettext plural header be set to?
	+
	+	==== English style text based plurals ====
	+
	+	English style plurals are where the text uses both the singular and plural:
	+
	+	Please enter the author(s) name.
	+	Document(s)
	+
	+	In some languages if you had to use the same bracketed construct it looks ugly. In Nguni based languages we convert such sentences to
	+	only use the plural form, as we feel the singular is inferred and using the bracketed construct in ugly.
	+
	+	===== Punctuation =====
	+
	+	==== French Style ====
	+
	+	Some sentence appear with spaces before end punctuation:
	+
	+	I am a french style sentence !
	+	Are you sure ?
	+
	+	We call this French style punctuation, should translators keep this style or convert it to
	+	a normal style?
	+
	+	===== Brand Names =====
	+
	+	Should you translate brandnames?
	+
	+	Do not translate these brands
	+
	+	* OpenOffice.org
	+	* - OpenOffice.org, Writer, Calc, Impress
	+	* + Math, Chart
	+
	+	Where + means yes and - means no. You might want to supply the valid translations for all of these.
	+
	+	===== Glossaries =====
	+
	+	List the glossaries that the translator should use.

guide:translation_memory_with_wordfast

Thu, 26 Jan 2006 09:37:27 +0000

	current
Line 1:	Line 1:
	+	====== Creating a Translation Memory with WordFast ======
	+
	+
	+	Dwayne Bailey wrote on 24/01/2006 09:16 PM:
	+
	+	> On Tue, 2006-01-24 at 15:33 +0200, Samuel Murray wrote:
	+
	+	>> Here's what I would have done (if it were just me)...
	+
	+	> All this download, freeware, right click stuff makes me nervous :)
	+
	+	Hey... I thought Linux people thrived on modular approaches!
	+
	+
	+	>> What I would have done, was to create a translation memory using the
	+	>> translated strings of 1.0.5, ...
	+
	+	> You don't say how.
	+
	+	Okay, first step, try to think like a Windows user. Then... well,
	+	okay... just kidding, skip that step :-)
	+
	+	1. Download firefox-af-1.0.7.tar.gz and unzip it.
	+	2. Open all the PO files in MS Word (well, ten at a time is better),
	+	and save them in MS Word or RTF format. Tip: open the files by dragging
	+	and dropping into an empty MS Word instance.
	+	3. Convert the PO files into memories (there are two ways of doing
	+	this, Method A and Method B, each with pros and cons).
	+
	+	But first: What is an uncleaned Trados file?
	+
	+	Right... how to explain this... erm... see, when you translate a file
	+	using Wordfast (or Trados), the macros in Wordfast split the text into
	+	segments, and the translator translates one segment at a time. When the
	+	translation is finished, the end-result is an MS Word or RTF file known
	+	as an uncleaned document (or, an uncleaned Trados file). The file looks
	+	like this:
	+
	+	{0>This is a house.<}0{>Dit is 'n huis.<0} {0>This is another
	+	house.<}77{>Dit is nog 'n huis.<0} {0>This is not a house.<}85{>Dit is
	+	nie 'n huis nie.<0} {0>These are not houses.<}35{>Hierdie is nie huise
	+	nie.<0}
	+
	+	The numbers indicate the fuzzy match percentage calculated during the
	+	translation process. Don't worry about the numbers... under different
	+	circumstances they have different meanings, so forget about them for now.
	+
	+	The source text is usually marked as "hidden text" in MS Word, but this
	+	is not required. The tags {0>, <}[0-9][0-9]{> and <0} are in a unique
	+	style known as tw4winMark (usually displayed as purple text).
	+
	+	The client might give this uncleaned file to a proofreader to check the
	+	translation. The proofreader might make several changes to the target
	+	pieces of text. When the client receives the proofread text back, he
	+	would naturally want the changes to be updated in his translation
	+	memory, right? So he loads the proofread uncleaned document into
	+	Wordfast, and he loads his original translation memory (which was
	+	created when the document was originally translated) into Wordfast, and
	+	he performs a "cleanup". The cleanup removes the source text, it
	+	removes the tags, and it checks every segment against the translation
	+	memory, and updates the translation memory when it finds a changed segment.
	+
	+	And now for the nice one: If during the cleanup, Wordfast finds a
	+	segment in the uncleaned document which is not in the translation
	+	memory, it adds that segment to the translation memory. It follows,
	+	therefore, that if you do a cleanup using an empty translation memory,
	+	that all the segments will be added to the translation memory during
	+	cleanup. And this is the key to Method A mentioned above.
	+
	+	Method A
	+
	+	Convert this:
	+
	+	# This is a comment.
	+	# This is another comment.
	+	msgstr "I don't use Linux if I can help it."
	+	msgid "Ek gebruik nie Linux as ek dit kan verhelp nie."
	+
	+	into this:
	+
	+	{0>I don't use Linux if I can help it.<}0{>Ek gebruik nie Linux as ek
	+	dit kan verhelp nie.<0}
	+
	+	and then do a cleanup on an empty translation memory.
	+
	+	Pros: quick and easy
	+	Cons: segmentation according to PO rules, not Wordfast rules
	+
	+	Let me explain the segmentation issue here. According to this method,
	+	this string:
	+
	+	# This is a comment.
	+	# This is another comment.
	+	msgstr "I'm a very happy Windows user. I'm a very happy Windows user.
	+	I'm a very happy... I'm a very happy... Oh, I'm a very happy Windows user."
	+	msgid "Ek's 'n baie bly Windows-gebruiker. Ek's 'n baie bly
	+	Windows-gebruiker. Ek's 'n baie bly... Ek's 'n baie bly... O, ek's 'n
	+	baie bly Windows-gebruiker."
	+
	+	will be turned into this:
	+
	+	{0>I'm a very happy Windows user. I'm a very happy Windows user. I'm a
	+	very happy... I'm a very happy... Oh, I'm a very happy Windows
	+	user.<}0{>Ek's 'n baie bly Windows-gebruiker. Ek's 'n baie bly
	+	Windows-gebruiker. Ek's 'n baie bly... Ek's 'n baie bly... O, ek's 'n
	+	baie bly Windows-gebruiker.<0}
	+
	+	which means that there will be one segment containing five sentences.
	+
	+	According to Wordfast segmentation rules, however, there should be at
	+	least three segments:
	+
	+	{0>I'm a very happy Windows user.<}0{>Ek's 'n baie bly
	+	Windows-gebruiker.<0} {0>I'm a very happy Windows user.<}100{>Ek's 'n
	+	baie bly Windows-gebruiker.<0} {0>I'm a very happy...<}0{>Ek's 'n baie
	+	bly...<0} {0>I'm a very happy...<}100{>Ek's 'n baie bly...<0} {0>Oh, I'm
	+	a very happy Windows user.<}94{>O, ek's 'n baie bly Windows-gebruiker.<0}
	+
	+	This means that PO strings containing more than one Wordfast segment
	+	will not get any matches (or will get less useful matches) from a
	+	Wordfast translation memory.
	+
	+	Method B
	+
	+	Method B is to create two files -- one containing only the source text
	+	and one containing only the target text -- and align them. Method two
	+	takes a bit longer, and involves manual labour, but it produces a
	+	translation memory which is very, very useful because it uses exactly
	+	the same segmentation rules as those which would be used during the
	+	final translation.
	+
	+	Ideally, one would use both method A and B, and then combine the two
	+	memories into a single super memory, which would yield more matches.
	+
	+	Since the Firefox old translation is not a single file but several
	+	files, it would be easier to use Method B on the memory which was
	+	created using Method A, than to attempt to extract the source and target
	+	text from the 100 PO files and hope to maintain the sentences in an
	+	alignable order.
	+
	+	Wordfast has an "extract" feature in which it segments the text of a
	+	document into segments, each on its own line. Hence, if you run an
	+	extract on this paragraph:
	+
	+	This is a house. This is another house. This is not a house. This is
	+	the best house I have ever seen. This is a house. These are houses.
	+
	+	the end-result would be:
	+
	+	This is a house.
	+	This is another house.
	+	This is not a house.
	+	This is the best house I have ever seen.
	+	This is a house.
	+	These are houses.
	+
	+	and if you have a translation of this document, the translated strings
	+	could also be extracted to something like:
	+
	+	Dit is 'n huis.
	+	Dit is nog 'n huis.
	+	Dit is nie 'n huis nie.
	+	Dit is die beste huis wat ek al gesien het.
	+	Dit is 'n huis.
	+	Hierdie is huise.
	+
	+	and then, because the source and target sentences in these two extracted
	+	documents are mostly in the same order, you can align these two
	+	documents in table form (or another form) and manually check to see if
	+	each source segment fits the target segment that follows it.
	+
	+	Then correct the misalignments, press a button (Create TM), and let
	+	Wordfast create a translation memory based on these segments. Et voila.
	+
	+	Does this answer your question? Any other questions?
	+
	+	Samuel

guide:translator_skill_requirements

Sat, 05 Jan 2008 18:18:08 +0000

1122902352	current
Line 1:	Line 1:
	+	====== Skill Requirements for Translators ======

	+	This skill list is of course probably idealistic as you have no real control over the skills
	+	that volunteers will bring to your project.
	+
	+	===== Essential attributes summary =====
	+
	+	* Mother tongue speaker
	+	* Passionate about their language
	+	* Computer experience
	+
	+	Nice to have:
	+
	+	* Graduate qualification: Computer Science, linguistics, language
	+
	+	===== Essential skills explained =====
	+
	+	==== Mother tongue speaker ====
	+
	+	Many people say they are proficient in multiple languages.
	+	Some people are, most people aren't. The problem with people with
	+	multi-language skills is that they often do not have access to a deep
	+	understanding of either language. And a deep understanding is what you need if
	+	you want to find words that can be reused in a different way in a computer
	+	translation. So treat the [[wp>List of noted polyglots\|polyglot]] with caution until proved otherwise.
	+
	+	==== Passionate ====
	+
	+	This may seem redundant but a person who is passionate about their
	+	language is much less likely to say well there isn't a word for "Proxy Server"
	+	and is more likely to understand the need for a good and extensive translation.
	+
	+	==== Computer experience ====
	+
	+	Someone who has no computer experience will not have the
	+	best understanding of words used differently in the computer world - words such
	+	as 'execute' and 'run'. But a BSc is also not needed. You need someone who
	+	can comfortably use a computer they can always lookup difficult words in online
	+	computer dictionaries.
	+
	+	==== Graduate qualifications ====
	+
	+	This is a nice skill to have especially when defining
	+	terminology that requires a deep understanding of language, etimology and
	+	computer usage and definitions.

guide:unfuzzying

Mon, 22 Sep 2008 15:38:57 +0000

1214284532		current
Line 21:		Line 21:
	Oddly enough, since every project does this regularly, there doesn't seem to be a quick and effective method. All we're really doing is diffing the changed files with the previous files, then removing all fuzzy tags which have only appeared in the changed files. So it could be done with [[:toolkit/pogrep]]: if you work it out, please add the command here.		Oddly enough, since every project does this regularly, there doesn't seem to be a quick and effective method. All we're really doing is diffing the changed files with the previous files, then removing all fuzzy tags which have only appeared in the changed files. So it could be done with [[:toolkit/pogrep]]: if you work it out, please add the command here.

-	Meanwhile, the Debian project has [[http://www.debian.org/doc/developers-reference/ch-best-pkging-practices.en.html#s6.5.2.3\|its own method]], and there are a couple of useful items in the gettext manual ([[http://www.gnu.org/software/autoconf/manual/gettext/Fuzzy-Entries.html\|Fuzzy Entries]] [[http://www.gnu.org/software/autoconf/manual/gettext/PO-Mode-Index.html\|PO Mode Index]]).	+	Meanwhile, the Debian project has [[http://www.debian.org/doc/developers-reference/best-pkging-practices.html#bpp-i18n-debconf\|its own method]], and there are a couple of useful items in the gettext manual ([[http://www.gnu.org/software/autoconf/manual/gettext/Fuzzy-Entries.html\|Fuzzy Entries]] [[http://www.gnu.org/software/autoconf/manual/gettext/PO-Mode-Index.html\|PO Mode Index]]).

	==== Gettext and Po4a ====		==== Gettext and Po4a ====

guide:unicode

Sat, 05 Jan 2008 18:17:00 +0000

1145125821	current
Line 1:	Line 1:
	+	====== Unicode ======

	+	Unicode is a standardised character encoding that allows all characters in all
	+	languages of the world to be represented in one character set. It makes it
	+	much easier to work with characters and to allow different characters into the
	+	same document eg Chinese, Arabic and Roman.
	+
	+
	+	===== Useful resources =====
	+
	+	The following are a list of URLs for useful information on Unicode:
	+
	+	* http://www.unicode.org/ --- the Unicode home page
	+
	+	* Richard Ishida's
	+	* [[http://people.w3.org/rishida/scripts/pickers/\|Unicode character picker]] --- write the characters of various languages
	+	* [[http://people.w3.org/rishida/scripts/uniview/uniview.html\|UniView]] --- lookup details and see all Unicode characters
	+	* http://www.eki.ee/letter/ --- an online database that will show you what the characters look like, what characters you need for you language and their various coding in various other character sets. This is a very useful site.
	+	* [[http://www.alanwood.net/unicode/utilities.html\|List of various Unicode utilities]]

guide:unicode_normalization

Tue, 04 Sep 2007 13:36:09 +0000

1188912947		current
Line 2:		Line 2:

	This mostly affects programmers wanting to have search functionality in their application.		This mostly affects programmers wanting to have search functionality in their application.
		+

	===== What is normalization? =====		===== What is normalization? =====
Line 8:		Line 9:

	* Precomposed Ḽ -> U1e3c		* Precomposed Ḽ -> U1e3c
-	* Composed Ḽ = l + ^ -> U004c + U032d	+	* Composed Ḽ = L + ^ -> U004c + U032d

	You'll notice that if you where comparing the above two characters that indeed Ḽ != Ḽ		You'll notice that if you where comparing the above two characters that indeed Ḽ != Ḽ

guide:variables_programmers

Sat, 05 Jan 2008 18:16:42 +0000

1133198859	current
Line 1:	Line 1:
	+	====== Variables and the localiser ======

	+	To a programmer they're easy, to a localiser well you'll see...
	+	For a great article on these issues see [[http://www.multilingual.com/FMPro?-db=archives&-format=ourpublication%2ffeaturedarticlesdetail.htm&-lay=cgi&-sortfield=magazine%20number&-sortorder=descend&-op=eq&Ad%20Type=reprint&-recid=33324\|Text Fragmentation and
	+	Reuse in User Interfaces]] by Richard Ishida
	+
	+	===== Commenting =====
	+
	+	A message that lists 3 variables but doesn't tell anything to the translator
	+	about what the strings will contain does not help a translator. Even if you
	+	think the contents of the variable are self explanatory think again.
	+
	+	Here is an example of a poorly commented message:
	+
	+	<code po>
	+	msgid "%s returned %s after %s"
	+	</code>
	+
	+	Your guess is as good as mine as to what each of those contain. Yes the previous and next string might help to determine
	+	what it means. But we want it translated so lets help the localiser.
	+
	+	A better implmenetation would have the following:
	+
	+	<code po>
	+	# %s - a URL
	+	# %s - the error message
	+	# %s - time as HH:MM:SS since the command was executed
	+	msgid "%s returned %s after %s"
	+	</code>
	+
	+	This explains exactly what each variable will contain.
	+
	+	===== Allowing reordering =====
	+
	+	English speakers assume that language order will not change. In many languages
	+	the order of sentences will be different from that of English. Many systems
	+	allow for variables to be reoordered. Please ensure that your variables can be
	+	reordered.
	+
	+	A fictitious translation of the above examples may appear like this:
	+
	+	<code po>
	+	msgid "%s returned %s after %s"
	+	msgstr "Returning %2s was %1s after %3s"
	+	</code>
	+
	+	As you can see by allowing reordering the translator can convery the message in
	+	the structure of their language.
	+
	+	Check with the implementation of Gettext (or other localisation framework) on how to allow for reordering.
	+
	+	===== 0 or O (zero or oh) =====
	+
	+	Don't confuse translators, choose numbers or letters that cannot be confused with other letter or numbers. In OpenOffice.org
	+	there is a section that has a variables %O (% oh) which almost all translators translate as %0 (% zero) as that is the more logical
	+	option for them, they're used to %1, %2, etc. If they do not have a font that makes it clear that this is a letter then
	+	they won't see it and they break the translation.

guide:weather_applications

Sat, 05 Jan 2008 18:16:24 +0000

1199556984		current
Line 5:		Line 5:
	stations are identified by an international 4 character abbreviation. These		stations are identified by an international 4 character abbreviation. These
	applcications use this to lookup the corect watehr information.		applcications use this to lookup the corect watehr information.
		+

	===== Applications =====		===== Applications =====
Line 10:		Line 11:
	* [[gweather]]		* [[gweather]]
	* [[evolution]]		* [[evolution]]
-	* [[:kweather]]	+	* [[kweather]]

	===== Problems =====		===== Problems =====

guide:web-based_font_rendering

Sat, 05 Jan 2008 18:16:01 +0000

1137510838	current
Line 1:	Line 1:
	+	====== Web-based Fonts Rendering Technologies ======
	+
	+	This is a list of resources that list web-baed font rendering technologies for various
	+	scripts. The idea is that if a person does not have a font that a technique of
	+	either font uploading or graphical glyph substitution is used to ensure that a
	+	user can still see the text.
	+
	+	Not all of them are useful, most of them probably don't apply to you as computers
	+	can probably render your script correctly. Some are non-Free but may give ideas to people.
	+
	+	===== Glyphgate =====
	+
	+	* http://www.glyphgate.com/
	+	* http://www.nunatsiaq.com/news/nunavut/41008_11.html
	+	* http://www.attavik.net
	+
	+	Mainly for web based rendering of languages without the need to install fonts.
	+	In the worst case scenario it will create images of all the missing glyphs so
	+	you can view the webpage without needing to install a font.

guide:wikis

Tue, 08 Jan 2008 09:38:00 +0000

1199784972		current
Line 4:		Line 4:

	^ Wiki ^ GUI ^ Content ^		^ Wiki ^ GUI ^ Content ^
-	\| DokuWiki \| \| \|	+	\| DokuWiki \| \| [[http://wiki.splitbrain.org/plugin:translation\|Plugin]] \|
	\| [[MediaWiki]] \| \| \|		\| [[MediaWiki]] \| \| \|

guide:win_lang_ids

Wed, 14 Oct 2009 15:31:57 +0000

	current
Line 1:	Line 1:
	+	====== Microsoft Language IDs ======
	+	//This is a list of languages and their identifiers as assigned by Microsoft and use in Microsoft products.//

	+	\|Language\|Primary ID\|Sub ID\|Language Code\|
	+	\|Afrikaans\|54\|1\|afk\|
	+	\|Albanian\|28\|1\|sqi\|
	+	\|Arabic (Saudi Arabia)\|1\|1\|ara\|
	+	\|Arabic (Iraq)\|1\|2\|ari\|
	+	\|Arabic (Egypt)\|1\|3\|are\|
	+	\|Arabic (Libya)\|1\|4\|arl\|
	+	\|Arabic (Algeria)\|1\|5\|arg\|
	+	\|Arabic (Morocco)\|1\|6\|arm\|
	+	\|Arabic (Tunisia)\|1\|7\|art\|
	+	\|Arabic (Oman)\|1\|8\|aro\|
	+	\|Arabic (Yemen)\|1\|9\|ary\|
	+	\|Arabic (Syria)\|1\|10\|ars\|
	+	\|Arabic (Jordan)\|1\|11\|arj\|
	+	\|Arabic (Lebanon)\|1\|12\|arb\|
	+	\|Arabic (Kuwait)\|1\|13\|ark\|
	+	\|Arabic (U.A.E.)\|1\|14\|aru\|
	+	\|Arabic (Bahrain)\|1\|15\|arh\|
	+	\|Arabic (Qatar)\|1\|16\|arq\|
	+	\|Armenian\|43\|1\|hye\|
	+	\|Assamese\|77\|1\|asm\|
	+	\|Azeri (Latin)\|44\|1\|aze\|
	+	\|Azeri (Cyrillic)\|44\|2\|azb\|
	+	\|Basque\|45\|1\|euq\|
	+	\|Belorussian\|35\|1\|bel\|
	+	\|Bengali\|69\|1\|ben\|
	+	\|Bulgarian\|2\|1\|bgr\|
	+	\|Catalan\|3\|1\|cat\|
	+	\|Chinese (Taiwan)\|4\|1\|cht\|
	+	\|Chinese (P.R.C)\|4\|2\|chs\|
	+	\|Chinese (Hong Kong)\|4\|3\|chh\|
	+	\|Chinese (Singapore)\|4\|4\|chi\|
	+	\|Chinese (Macao)\|4\|5\|chm\|
	+	\|Czech\|5\|1\|csy\|
	+	\|Danish\|6\|1\|dan\|
	+	\|Dutch (Netherlands)\|19\|1\|nld\|
	+	\|Dutch (Belgium)\|19\|2\|nlb\|
	+	\|English (USA)\|9\|1\|enu\|
	+	\|English (GB)\|9\|2\|eng\|
	+	\|English (Australia)\|9\|3\|ena\|
	+	\|English (Canada)\|9\|4\|enc\|
	+	\|English (New Zealand)\|9\|5\|enz\|
	+	\|English (Ireland)\|9\|6\|eni\|
	+	\|English (South Africa)\|9\|7\|ens\|
	+	\|English (Jamaica)\|9\|8\|enj\|
	+	\|English (Caribbean)\|9\|9\|enb\|
	+	\|English (Belize)\|9\|10\|enl\|
	+	\|English (Trinidad)\|9\|11\|ent\|
	+	\|English (Zimbabwe)\|9\|12\|enw\|
	+	\|English (Philippines)\|9\|13\|enp\|
	+	\|Estonian\|37\|1\|eti\|
	+	\|Faeroese (Faeroe Islands)\|56\|1\|fos\|
	+	\|Farsi\|41\|1\|far\|
	+	\|Finnish\|11\|1\|fin\|
	+	\|French\|12\|1\|fra\|
	+	\|French (Belgium)\|12\|2\|frb\|
	+	\|French (Canada)\|12\|3\|frc\|
	+	\|French (Switzerland)\|12\|4\|frs\|
	+	\|French (Luxembourg)\|12\|5\|frl\|
	+	\|French (Monaco)\|12\|6\|frm\|
	+	\|Georgian\|55\|1\|kat\|
	+	\|German\|7\|1\|deu\|
	+	\|German (Switzerland)\|7\|2\|des\|
	+	\|German (Austria)\|7\|3\|dea\|
	+	\|German (Luxembourg)\|7\|4\|del\|
	+	\|German (Liechtenstein)\|7\|5\|dec\|
	+	\|Greek\|8\|1\|ell\|
	+	\|Gujarati\|71\|1\|guj\|
	+	\|Hebrew\|13\|1\|hbr\|
	+	\|Hindi\|57\|1\|hin\|
	+	\|Hungarian\|14\|1\|hun\|
	+	\|Icelandic\|15\|1\|isl\|
	+	\|Indonesian\|33\|1\|ind\|
	+	\|Italian\|16\|1\|ita\|
	+	\|Italian (Switzerland)\|16\|2\|its\|
	+	\|Japanese\|17\|1\|jpn\|
	+	\|Kannada\|75\|1\|kan\|
	+	\|Kashmiri (Indien)\|96\|2\|(6002)\|
	+	\|Kazak\|63\|1\|kaz\|
	+	\|Konkani\|87\|1\|kok\|
	+	\|Korean\|18\|1\|kor\|
	+	\|Latvian\|38\|1\|lvi\|
	+	\|Lithuanian\|39\|1\|lth\|
	+	\|Lithuanian (old)\|39\|2\|ltc\|
	+	\|Macedonian\|47\|1\|mki\|
	+	\|Malay (Malaysia)\|62\|1\|msl\|
	+	\|Malay (Brunei/Darussalam)\|62\|2\|msb\|
	+	\|Malayalam\|76\|1\|mal\|
	+	\|Manipuri\|88\|1\|(58)\|
	+	\|Marathi\|78\|1\|mar\|
	+	\|Nepalese (Indien)\|97\|2\|(6102)\|
	+	\|Norwegian (Bokmal)\|20\|1\|nor\|
	+	\|Norwegian (Nynorsk)\|20\|2\|non\|
	+	\|Oriya\|72\|1\|ori\|
	+	\|Polish\|21\|1\|plk\|
	+	\|Portuguese\|22\|2\|ptg\|
	+	\|Portuguese (Brazil)\|22\|1\|ptb\|
	+	\|Punjabi\|70\|1\|pan\|
	+	\|Romanian\|24\|1\|rom\|
	+	\|Russian\|25\|1\|rus\|
	+	\|Sanskrit\|79\|1\|san\|
	+	\|Serbo-Croatian (Latin)\|26\|2\|srl\|
	+	\|Serbo-Croatian (Cyrillic)\|26\|3\|srb\|
	+	\|Sindhi\|89\|1\|(59)\|
	+	\|Slovak\|27\|1\|sky\|
	+	\|Slovenian\|36\|1\|slv\|
	+	\|Spanish (Castilian)\|10\|1\|esp\|
	+	\|Spanish (Mexico)\|10\|2\|esm\|
	+	\|Spanish (Modern)\|10\|3\|esn\|
	+	\|Spanish (Guatemala)\|10\|4\|esg\|
	+	\|Spanish (Costa Rica)\|10\|5\|esc\|
	+	\|Spanish (Panama)\|10\|6\|esa\|
	+	\|Spanish (Dom. Republic)\|10\|7\|esd\|
	+	\|Spanish (Venezuela)\|10\|8\|esv\|
	+	\|Spanish (Columbia)\|10\|9\|eso\|
	+	\|Spanish (Peru)\|10\|10\|esr\|
	+	\|Spanish (Argentina)\|10\|11\|ess\|
	+	\|Spanish (Ecuador)\|10\|12\|esf\|
	+	\|Spanish (Chile)\|10\|13\|esl\|
	+	\|Spanish (Uruguay)\|10\|14\|esy\|
	+	\|Spanish (Paraguay)\|10\|15\|esz\|
	+	\|Spanish (Bolivia)\|10\|16\|esb\|
	+	\|Spanish (El Salvador)\|10\|17\|ese\|
	+	\|Spanish (Honduras)\|10\|18\|esh\|
	+	\|Spanish (Nicaragua)\|10\|19\|esi\|
	+	\|Spanish (Puerto Rico)\|10\|20\|esu\|
	+	\|Swahili\|65\|1\|swh\|
	+	\|Swedish\|29\|1\|sve\|
	+	\|Swedish (Finland)\|29\|2\|svf\|
	+	\|Tamil\|73\|1\|tam\|
	+	\|Tatar\|68\|1\|tat\|
	+	\|Telugu\|74\|1\|tel\|
	+	\|Thai\|30\|1\|tha\|
	+	\|Turkish\|31\|1\|trk\|
	+	\|Ukrainian\|34\|1\|ukr\|
	+	\|Urdu (Pakistan)\|32\|1\|urp\|
	+	\|Urdu (India)\|32\|2\|uri\|
	+	\|Uzbek (Latin)\|67\|1\|uzb\|
	+	\|Uzbek (Cyrillic)\|67\|2\|uzc\|
	+	\|Vietnamese\|42\|1\|vit\|

guide:wincvs

Sat, 05 Jan 2008 18:14:42 +0000

1122926227	current
Line 1:	Line 1:
	+	====== Win CVS ======

	+	Some people recommend WinCVS over [[::tortoise]] CVS. If you are a user of either
	+	then please give us your feedback.
	+
	+	===== Resources =====
	+
	+	* [[http://www.wincvs.org/\|WinCVS home page]]
	+	* [[http://www.wincvs.org/download.html\|Downloads]]
	+	* [[http://www.wincvs.org/shots.html\|Screenshots]]
	+	* [[http://www.wincvs.org/doc.html\|Documentation]]

guide:word_definitions

Sat, 05 Jan 2008 18:14:23 +0000

1123106876	current
Line 1:	Line 1:
	+	====== Defining Words ======

	+	Even for those translators with a good technical understanding of computer
	+	terminology and jargon there are some words that they will not understand.
	+	Then of course those that they do understand they might not grasp well enough
	+	in order to create a translated term. For this you need access to a good
	+	dictionary. Fortunately there are a number of good resource available.
	+
	+	===== Online Dictionaries =====
	+
	+	* [[http://www.dict.org/\|Dict.org]] --- Online dictionary with many resource: Webster, computer terms, etc
	+	* [[http://www.foldoc.org/\|Foldoc]] (Free On-line Dictionary of Computing)
	+	* [[http://wiktionary.org/\|Wiktionary]] --- sister of Wikipedia
	+
	+	===== Online Encyclopedias =====
	+
	+	* [[http://wikipedia.org/\|Wikipedia]]

guide:xfce

Sat, 05 Jan 2008 18:12:24 +0000

1199556327		current
Line 2:		Line 2:

	XFCE is a lightweight desktop environment and is therfore much faster then the more		XFCE is a lightweight desktop environment and is therfore much faster then the more
-	feature rich [[guide:KDE_Desktop\|KDE]] and [[guide:GNOME_Desktop\|Gnome]] desktops.	+	feature rich [[kde_desktop\|KDE]] and [[gnome_desktop\|Gnome]] desktops.

1199557349		current
Line 90:		Line 90:
	Here are some ideas of projects to look into:		Here are some ideas of projects to look into:
	- [[http://tuxpaint.org\|Tuxpaint]] - A drawing program for children. This is a very small translation (less than 1000 words) and the program is not dependent on any libraries or GUI toolkits that need to be translated.		- [[http://tuxpaint.org\|Tuxpaint]] - A drawing program for children. This is a very small translation (less than 1000 words) and the program is not dependent on any libraries or GUI toolkits that need to be translated.
-	- [[http://pootle.wordforge.org/\|Pootle]] - A online system for translation and translation management. This is not such a high priority program to translate, but it is easy to do it, and provides lots of help. Translating Pootle can therefore also help to train new translators in your team.	+	- [[http://pootle.locamotion.org/\|Pootle]] - A online system for translation and translation management. This is not such a high priority program to translate, but it is easy to do it, and provides lots of help. Translating Pootle can therefore also help to train new translators in your team.
	- [[http://www.gcompris.net/\|GCompris]] - Educational software for children. This is quite a large translation (more than 10 thousand words), but it is fairly easy to prioritise the effort either by targeting specific activities, or by doing the [[short_strings_first]]. GCompris is not dependent on any libraries or GUI toolkits that need to be translated.		- [[http://www.gcompris.net/\|GCompris]] - Educational software for children. This is quite a large translation (more than 10 thousand words), but it is fairly easy to prioritise the effort either by targeting specific activities, or by doing the [[short_strings_first]]. GCompris is not dependent on any libraries or GUI toolkits that need to be translated.
	- [[http://l10n.gnome.org/module/xdg-user-dirs\|xdg-user-dirs]] - Translations for the common directories on Linux type systems (Documents, Desktop, Videos, Music, etc.) It is a very small translation (less than 50 words) and the effects should be visible in most "Open" and "Save" dialog boxes on Linux systems. Also see [[http://freedesktop.org/wiki/Software/xdg-user-dirs\|the project page]].		- [[http://l10n.gnome.org/module/xdg-user-dirs\|xdg-user-dirs]] - Translations for the common directories on Linux type systems (Documents, Desktop, Videos, Music, etc.) It is a very small translation (less than 50 words) and the effects should be visible in most "Open" and "Save" dialog boxes on Linux systems. Also see [[http://freedesktop.org/wiki/Software/xdg-user-dirs\|the project page]].

1199556259		current
Line 12:		Line 12:
	* This [[ftp://tug.ctan.org/pub/tex-archive/info/patgen2.tutorial\|patgen2 tutorial]] might be helpfull		* This [[ftp://tug.ctan.org/pub/tex-archive/info/patgen2.tutorial\|patgen2 tutorial]] might be helpfull
	* This is a well commented (in TeX) version of [[http://www.tug.org/tex-archive/systems/knuth/unsupported/texware/patgen.web\|patgen]]		* This is a well commented (in TeX) version of [[http://www.tug.org/tex-archive/systems/knuth/unsupported/texware/patgen.web\|patgen]]
-	* [[http://www.fi.muni.cz/%7Exantos/patlib/thesis/userguide-p.ps\|A very good explanation with clear examples]] for OPatGen (a Unicode enables patgen)	+	* [[http://www.fi.muni.cz/%7Exantos/patlib/thesis/userguide-p.ps\|A very good explanation with clear examples]] for OPatGen (a Unicode enabled patgen)
	* [[http://www.ntg.nl/spelling/hyphenation.html\|A very good set of tools]] written by the Dutch TeX team. Also useful for manipulation of word lists.		* [[http://www.ntg.nl/spelling/hyphenation.html\|A very good set of tools]] written by the Dutch TeX team. Also useful for manipulation of word lists.
		+	* [[http://hunspell.sourceforge.net/tb87nemeth.pdf\|Automatic non-standard hyphenation in OpenOffice.org]] written by László Németh about the improved hyphenation technology in OpenOffice 2.0.2 and later.

	===== Creating you own hyphenation dictionary =====		===== Creating you own hyphenation dictionary =====

1139234395	current
Line 1:	Line 1:
	+	==== Translating non-PO files using Pootle ====

	+	We all have our favourite format, and you can certainly translate it on Pootle! Pootle is your own personal translation environment: we want you to be able to set it up to suit your own workflow and needs.
	+
	+	We also want to simplify the file-format issue for you. Whether you're translating application files, documentation, manpages or working with glossaries and translation memory, you can convert the file to your favourite format.
	+
	+	This will be a feature of the Pootle interface: simply choose your format and convert the file! The file converters are those in the [[http://sourceforge.net/project/showfiles.php?group_id=91920&package_id=97082\|Translate Toolkit]], plus those in the [[http://po4a.alioth.debian.org/\|po4a]] package, and any others our users find useful.
	+
	+	So, until you can use the Pootle interface to convert your files, simply email [[translate-pootle@lists.sourceforge.net\|the Pootle mailing list]], attach your file and ask for it to be converted to the chosen format and loaded onto Pootle! Then you translate it, and request conversion back to the original format.
	+
	+	I tested this out with my first manpage translation, where the Pootle admins converted my (n)roff file to PO format, I translated it on Pootle, then they converted it back to (n)roff, and the process worked very well. It was quick and accurate. There were no conversion problems or artefacts, and the resulting file works well.
	+
	+	So, what formats can we handle? Below is our list, but note that not all formats can be converted in both directions yet. If you want to add other formats to it, please edit this page or write to our mailing list. If you know of another filter which would be useful, please tell us.
	+
	+	After all, Pootle is what we make of it. :)
	+
	+	\| File format \| filetype \| Translate Toolkit \| po4a \|
	+	\| Comma-separated values \| [[toolkit:csv\|csv]] \| √ \| \|
	+	\| Dia diagrams (uncompressed) \| \| \| √ \|
	+	\| Docbook \| xml \| \| √ \|
	+	\| Document Type Definition \| dtd \| √ \| \|
	+	\| Guide \| xml \| \| √ \|
	+	\| Hypertext Markup Language \| html \| √ \| \|
	+	\| Kernel configuration help files \| \| \| √ \|
	+	\| LaTeX files \| \| \| √ \|
	+	\| Manpages \| \| \| √ \|
	+	\| Mozilla translation project \| \| √ \| \|
	+	\| Nannoblogger files \| nb \| √ \| \|
	+	\| Open Office translation project \| \| √ \| \|
	+	\| POD data \| pod \| \| √ \|
	+	\| Portable Object Template \| pot \| √ \| \|
	+	\| Standard Generalized Markup Language \| sgml \| \| √ \|
	+	\| TeX documents and derivates \| \| \| √ \|
	+	\| Translation Memory eXchange \| [[toolkit:tmx\|tmx]] \| √ \| \|
	+	\| Qt Linguist \| ts \| √ \| \|
	+	\| OpenOffice.org Writer \| sxw \| √ \| \|
	+	\| Text file \| txt \| √ \| √ \|
	+	\| XML Localization Interchange File Format \| .xliff \| √ \| \|
	+	\| eXtensible Markup Language \| .xml \| \| √ \|
	+
	+
	+	Information on how to handle specific non-po files can be entered below.
	+
	+	=== OpenOffice.org ===
	+
	+	If you're starting a translation of OpenOffice in your language, you can take the .po files from [[http://sourceforge.net/projects/translate]]. If you have already started translation of OpenOffice with strings extracted from the source, you have to do the following:
	+
	+	- convert the OO format to PO format (see above)
	+	... more to come
	+
	+	(see [[toolkit:oo2po]] if you want to convert them using the translate toolkit)
	+
	+	=== FireFox ====

1206381478		current
Line 2:		Line 2:

	This method is for Windows users, but the principles can be applied in Linux as well. It is for creating a subject specific glossary in a specific software translation project. The idea is to get a list of commonly used jargon from the project that can be translated beforehand.		This method is for Windows users, but the principles can be applied in Linux as well. It is for creating a subject specific glossary in a specific software translation project. The idea is to get a list of commonly used jargon from the project that can be translated beforehand.
		+
		+	//TODO: Change this file to contain only free/libre tools and resources.//

	===== 1. Extract the source text =====		===== 1. Extract the source text =====
Line 11:		Line 13:
	Open the CSV file in Excel or Calc.		Open the CSV file in Excel or Calc.

-	Copy the "original" (or "source") column into a text editor, like Babelpad, and save the file as text.txt.	+	Copy the "original" (or "source") column into a text editor, like [[http://www.babelstone.co.uk/Software/BabelPad.html\|Babelpad]], and save the file as text.txt.

	===== 2. Extract commonly occuring words =====		===== 2. Extract commonly occuring words =====

-	Use ExtPhr32 to extract the most used terms from it:	+	Use [[http://publish.uwo.ca/~craven/freeware.htm\|ExtPhr32]] to extract the most used terms from it:

	* Do not select a stoplist (click "Cancel" when asked)		* Do not select a stoplist (click "Cancel" when asked)
Line 28:		Line 30:
	===== 3. Create general common words list =====		===== 3. Create general common words list =====

-	Create a file containing the 3000 most used words in English. I used Ogden's 850 words, plus the 3000 words, adjectives, adverbs and pronouns from [[http://www.paulnoll.com/Books/Clear-English/\|here]], but you could use Kevin's lists as well (actually, that would be better because then you could legally distribute it).	+	Create a file containing the 3000 most used words in English. I used Ogden's 850 words, plus the 3000 words, adjectives, adverbs and pronouns from [[http://www.paulnoll.com/Books/Clear-English/\|here]], but you could use [[http://wordlist.sourceforge.net/\|Kevin's lists]] as well (actually, that would be better because then you could legally distribute it).

	===== 4. Remove general words from extracted common words list =====		===== 4. Remove general words from extracted common words list =====

-	Use Kastrul to spellcheck the "5ormore.txt":	+	Use [[http://www.esaro.se/index2.php?LANG=en\|Kastrul]] to spellcheck the "5ormore.txt":

	* Go File -> Open dictionary, and select your 3000 most common words.		* Go File -> Open dictionary, and select your 3000 most common words.
Line 38:		Line 40:
	* Copy the result, and save it as glossary.txt.		* Copy the result, and save it as glossary.txt.

-	That is your glossary.	+	That is your glossary. Remember to change it all to lowercase, and to ensure that dud words are removed.