True, in DT 1.x the Path of an Import-captured folder was retained, and the Synchronize command could be used to transmit new (but not modified) files added subsequently to the capture, from the Finder folder to the corresponding database group.
In DT 2.0 the Path of an Import-captured group is NOT retained, so there is no communication between the Finder and group counterparts.
IMHO, the DT 2.0 change is appropriate, both because it better fits the logical and practical distinctions between Import-capture and Index-capture, and because it eliminates fairly widespread user misconceptions that often led to mistakes about what happened in DT 1.x when the Synchronize command was applied to an Import-captured group.
Point: Although in DT 1.x an Import-captured group retained the Path to the external folder, the documents in that group did not retain the Path to the corresponding files in the external folder. But in an Index-captured group, the documents within that group, that resulted from the capture of the external folder, retain the Paths to corresponding files in the external folder. The result is that when the Synchronize command is applied to an Indexed group, modifications such as edit changes to the external files are reflected in the database. But when the Synchronize command was applied to an Imported group, edit changes to files in the external folder were not reflected in the database. Too many users have made mistakes, because they assumed that edit changes to external files in a Synchronized group were updated in their database.
Portability: As a database that has Import-captured external files is self-contained, it’s easy to move it from computer to computer, or to run it from external media without the complications that would result from linkages to external folders and files. My main database dates back to the early days of DEVONthink, created initially on a TiBook, then moved to an iMac G5, a Power Mac, a MacBook Pro and currently my ModBook. In DT 1.x the Paths to Import-captured folders became useless vestigial appendices, as the content of the database had less and less correspondence to the external files and folders on each successive computer.
Interestingly, as I’ve moved databases around among my computers, I’ve encountered cases where folders had the same name on two or more computers, and met the Path criteria of a previously Import-captured group on another computer. But the folders had different purposes and content than the original Import-captured folder, so a Synchronize command on a database group would have produced inappropriate results.
Nowadays, more and more users choose Import capture so that they can move databases between their desktop and laptop computers. Retention of the Path to an external folder in such an environment, so that the Synchronize command could be used to bring in new content, becomes not only less useful but a source of potential glitches.