This page created 5 May 2006
Version note: Applies to TMG 8 & 9
This article discusses issues related to use of multiple languages in TMG, for example maintaining data to be output in two different languages.
Other articles, found on the main Languages page, or linked in appropriate places below, provide more detail on related features.
There are at least three different circumstances in which one might want to maintain two or more languages in TMG:
You have relatives or other interested persons who speak different languages, say some use English and others use German, and you wish to create reports for each group in their own language. This application involves all the issues discussed below.
You want to be able to produce reports in two variations of the same language, say a formal version that uses terms like "circa" and an informal version that substitutes "about." This application generally involves only the issue below with Sentence Structures.
You want to be able to create both a "normal" style of narrative and a special-purpose style with special formatting. This method is discussed in may article on Special Format Reports in this series. Because this application generally involves very few Tag Types, the issues discussed below generally do not apply.
In concept, creating output in two or more languages seems pretty simple. You set your user interface to whichever Language you prefer, enter your data, and then when you are ready for output you specify which Language is to be used for the report. You can even save separate report definitions, making it easy to save the Language and related options for use again later. However in practice there are some significant impediments to successful multi-language output. Some of them, and a few suggestions, are discussed below.
One of TMG's strengths is the ability for users to create customized Sentence Structures to produce exactly the type output they want in Narrative reports. Users can customize Sentences "Globally," applying to all Tags of a given type, or "Locally," applying to only a specific Tag. The challenge when one intends to produce narratives in two or more Languages is to consistently apply parallel changes to each Language.
The actual creation of Sentences in each Language is easy enough. There are Language selectors on the screens in which Sentences are edited. The challenge is to maintain the discipline, after making a change in one Language, to remember to make the corresponding change in the "other" Language(s).
If you are intending to produce output in multiple Languages it is probably best to avoid creating Local Sentences, because of the difficulty of continually remembering to keep each Language current, and considerable difficulty in finding and correcting failures to do so. Maintaining only Global Sentence changes is more manageable.
TMG does not provide translation of user-entered data, so the actual data entered can be an issue:
Dates – regular dates are output in the format and using the Language specified in the report, so are generally not an issue. If your Languages require different date formats, there are options on the Dates tab of report Options to use specific formats for a given report. However "irregular dates" (those that TMG couldn't interpret, so just recorded as you entered them) are not translated. So if you enter, say, "third Sunday in May" for your English cousins, the same text will appear if you create a report in French. One option might be to put such dates in the Memo, as described below.
Places – are intended to be entered in the ten fields provided to allow control of output for various types of reports and charts, among other uses. This works fine as long as you can use the same spelling of place names in each language. But if you require different spelling for each language, there is no simple solution. It might be possible to use some place fields for one language and others for a second, using appropriate variables in the Sentences for each language to use only those fields intended for that language. With ten fields to work with, and the Details field sub-dividable into nine segments, such a plan might be workable. But it would require diligence in data entry, and I don't know of any user who has tried it.
Memos – generally hold plain text, and would therefore normally need to be translated when used with different languages. Fortunately, the problem is not that difficult, provided the user is conversant in each language. Since the Memo field can be divided into up to nine segments, one could reserve, say, odd-numbered segments for one language and even-numbered segments for another. Then in the Sentences for the first use [M1], [M3], [M5], etc., and for the other [M2], [M4], [M6], etc. The user would then enter the same material in each pair of segments, once in one language and then in the other.
Names – are not easily accommodated if required to be spelled differently in different languages. Perhaps the best solution is to enter the name as it was spelled by the person at birth for the Primary Name, and create an alternate (perhaps custom) Name Tag that explains the alternate spelling in the other. That alternate tag could then be included reports in either or both languages, as desired.
There is no provision for producing source notes – footnotes or endnotes – in more than one language. Sources can readily be made to accommodate any single language, as I describe in my article on Data Entry Issues. But data elements are output into source notes exactly as input, with no method to specify different elements for different languages. Phrases and punctuation appearing as specified in the Output Templates with no way to change those templates for a second language.
| ReigelRidge Home | Terry's Tips Home | Contact Terry |
Copyright 2000- by Terry Reigel