I sent some files to our translation agency who is
working with Trados, latest version with latest service
pack (which supports FrameMaker 8). When I got the
translated files back, I noticed that text which had a
condition applied and which was hidden (so that it
shouldn't get translated) was changed. All umlauts were
deleted. After some e-mails to our translation agency
(who forwarded them to their Trados support) and to the
German FrameUser mailing list (thanks specifically to
Michael M?ller-Hillebrand!) my point of view is the

o Text in MIF files of FrameMaker 8 in the string
  statement must be encoded in Unicode (UTF-8).
  This info is also in the MIF reference.

o Text in MIF files after export from Trados via the
  S-Tagger is encoded in ANSI, at least text which has
  a condition applied and is hidden.

I confirmed that by the following:
1 Changed the file name extension of the MIF file which
  was created by FrameMaker and of the MIF which was created
  by Trados to txt.
2 Opened that txt file with Word. Word shows a dialog box to
  select the encoding. I selected Unicode (UTF-8).
3 When I searched for text with umlaut, the FrameMaker MIF
  file had the correct umlauts. In the Trados MIF file they
  were deleted. However, the umlauts in the Trados MIF file
  are present, when I select Windows Standard (which is ANSI).

However, the Trados support still says that the umlauts were
present after opening the Trados MIF files with NotePad++.
Therefore the MIF file were OK and it were a FrameMaker fault.

Can anyone either confirm the Trados or my point of view?
May text in MIF strings be encoded only in UTF-8 or also in

Thanks a lot for your help


Reply via email to