Detecting language please wait for.......
I would like to know what other people experience, but my impression is that there doesn't need to be any segmentation except for the already existing subtitle chunks. Even the full stop rule is rather confusing because it's not clear which line text will be displayed in.
I don't store the translations in a normal TM anyways. I have a separate video TM because the segments are so random (as chunked in the subtitling source) and I don't want them to clutter up my "good" TMs.
The benefit of using a CAT tool for me is in access to Termbase and concordance, and especially the preview with the different quality checks (words per minute etc.). That is such a huge help. Also the controlled process (approving segments etc.).
I really value this new tool, but I think I will turn all segmentation rules off in the respective TM.
In testing the new version with lots and lots of different videos and subtitle files I think I tend to agree with you. I can't comment from a real specialists perspective but not having the segmentation on does help when trying to quickly match things up as I'm playing around.
We should be releasing the updated versions of all the plugins we have developed so far for subtitling this coming week. Many many enhancements!
Yeah, it would be really useful if filetype plugin could enforce specific segmentation.
So far, the only solution I know is to add/create a TM with Paragraph Segmentation at the preparation stage. I believe this needs to be done for each language, which could be a pain when you need to prep for 20-30 langs.