Localization/Globalization

[enahar]

Dear Sir/Madam

We are going to do the Localization/Globalisation for our product which has
nearly got 284 forms using satellite Assemblies.
I have got 2 questions
1. How to handle Substrings.
2.What issues can you think could come across during
Globalization/Localisation.
3. How to develop an utility(can anybody share the code) which will display
the only the text values from the resource files and allow to enter new
locale language text values depending on the size of text values.After
enering the new text values for the other language saves into another
resource files.
4.when we set Localizable = true and set the language it creates a very big
resource files .We need only the items which has got only the text field.Can
anyone knows how to filter out only the text values from the resource files.

Kind Regards,

 

[Markus Stoeger]

Dear Sir/Madam

We are going to do the Localization/Globalisation for our product which
has
nearly got 284 forms using satellite Assemblies.

There are companies and tools that are specialized on translating software.
Where I work we used excel files to translate the software ourselves. Each
column contained one language and we used macros to generate code with
string ids from that. But that was quite a lot of work and never worked
really well.
Now we basically develop one single-language version of the software and
send the final .exe and resource .dll's to someone else to translate. They
do the translation and send back the modified files. We find that this is
cheaper and much easier for the developers than translating it yourself.

I have got 2 questions
1. How to handle Substrings.

What are substrings? For strings you should store them all in resource
files. This way they can later be translated without changing the source
code.

2.What issues can you think could come across during
Globalization/Localisation.

Some not so obvious things that I have heard of/seen happen:

- sometimes the same or similar phrases get translated with different words.
That mostly happens when the translation is done by different people or
when some time passes between translations. There are translation tools
that keep track of already-translated phrases in a central database. They
use some fuzzy matching algorithms to tell you when a similar phrase has
been translated before and how.

- in c/c++ you could printf("output %d some %d text", varA, varB);
There are languages which's grammar would require the position of varA and
varB to be swapped. You couldn't do that with %d. Instead in C# you should
use "output {0} some {1} text", varA, varB. If the language's grammar
required it, the translator could change the string to "output {1} some {0}
text", varA, varB.
Also try not to concatenate several strings. That makes translation much
harder because you will not be able to figure out the context of the single
text fragments in the resource file.

- always provide enough space for text on the form. Labels that are wide
enough in the english version can require 130% of that width when
translated to french for example. Not to think of languages that require
your text (or even the whole form) to be mirrored.. there are tools that
can check the width of labels to make sure no text gets cut off after
translation and which can automatically mirror your whole form for
languages that are written from right to left.

hope that helps,
Max