Eaf file created in Mac opens in Windows with timing differences

We are working annotating some sign language videos. Some of us use the Mac version of ELAN and others are using Windows. When the eaf files saved in Mac are opened in windows, the timing of the annotations have changed. Any ideas on how to solve this problem?

Hello,

I assume (but let me know if I’m wrong) that the start and end time values of annotations created on the Mac are still the same when opened on Windows, but that a different video frame is shown when e.g. activating an annotation?
If that is correct, the problem might be related to the encoding of the videos. On Mac and Windows ELAN uses different media frameworks and these might show different behaviour when e.g. jumping to a specific point in time. This issue is discussed in more detail in this thread and especially the last paragraph of point 1, about GOP and B-frames might be relevant.
But maybe my assumption is completely wrong.

-Han

Thank you. So, if I understand it correctly that thread. Even if the visualization of the annotated video is not exactly the same between windows and Mac, the actual times of the annotations are consistent (which is the most important thing for us). Am I right?

Yes, that should be the case.
The time values are stored as simple numbers in the eaf (XML) file and reading or writing those files does not depend on the platform. Hopefully you can verify this e.g. by writing down the exact times of a specific annotation created on the Mac and activating the same annotation on Windows and then checking the displayed times. Or by creating an annotation from e.g. exactly 10.0 sec. to 11.0 sec on one machine and checking if the times are the same when opened on another machine.

Thank you. We have opened the same file in a Mac and in a Windows computer, we checked the times of the annotations, and they match (although the visualization of the video doesn’t match completely). So no problem. It was just what you thought. Thank you so much for solving this issue.