Comment #⁨18⁩

In reply to Odin Urdland Odin Urdland

I do not recommend anyone to rely on this feature yet, on Windows.
If you export your data, remember to validate that it can actually unpack the archive on the Windows file system.
I am experiencing 3 types of problems, the way it is organizing the data in markdown files in subfolders. Because the naming of folders and files copy the naming in Tana:

  1. It just goes along with the naming (and the naming length) in the Tana graph, so it will INEVITABLY fail sooner or later in a graph, once you hit a combination of a) long node names and/or b) deep hierarchy.

  2. It allows some characters not eligible in the Windows file system and/or it attempts some truncation in this respect that results in duplicate names of files and folders (hence blocking the unpacking, or corrupting its integrity).

  3. Some things with length and/or characters slip through, that work out in the Windows file system, but is not allowed/workable in MS Onedrive. So it is not backed up and it blocks the entire ongoing sync operation of the Onedrive folder.
    (I understand that the Tana team cannot take every potential cloud storage provider into account. But it is quite common for Windows users to have all their user files placed in the Onedrive folder/sync that is installed with Windows out of the box. It is integral or at least widespread part of the Windows user experience/installation.)

(I have filed a bug report some weeks back. Just highlighting it here as well - for people to remember to verify their archives if they use it).