Hi sskk

The bilingual Excel file type you mention does not segment content using the set segmentation rules if there is any content in the target language column. This is most likely for the same reason:

As soon as you segment both languages, you'd need to have some form of an alignment process.

Example:

Source: This is a rather simple example. But in this context, it may suffice.

Target: Dieses eher einfache Beispiel dürfte unter diesem Umständen ausreichen.

Or, alternative translation: Unter diesen Umständen mag Folgendes ausreichen: Ein eher einfaches Beispiel.

The translations themselves might be deficient, but I hope the point is clear.

I would hope for SDL to develop a routine for all bilingual file types which does the segmentation of the source text and then opens the aligner to let the translator align existing target content. I don't think that would be very hard to do because they have all the building blocks for that, but such a thing does not yet exist, at least not in the SDL universe.

Daniel

Thank you for your reply.

In the case of the bilingual Excel file, if the target cells are blank, Studio conforms to the segmentation rule.

(Only when the target is already filled, the file is segmented as you said.)

And that's enough for me because in most cases, the reason why I use Trados Studio is to make a new translation from scratch.

But PO files are always segmented based on its native TUs....

sskk

It's not only Studio that will do this.  If you open such a file in POEdit for example you'll see something like this:

The problem for Studio (and I assume for others too) is that we have no idea how to segment the file correctly 100% of the time unless it's like this.  If you had three sentences in the source and only two in the target what would you do?  I know a guess might work, but that's not very technical, so the segmentation controls with these sort of files that provide a structural environment in the file itself is to treat them as paragraphs.